University of Southern California Dissertations and Theses

An electronic patient record (ePR) system for image-assisted minimally invasive spinal surgery

(USC Thesis Other)

An electronic patient record (ePR) system for image-assisted minimally invasive spinal surgery

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
AN ELECTRONIC PATIENT RECORD (EPR) SYSTEM FOR IMAGE-ASSISTED
MINIMALLY INVASIVE SPINAL SURGERY

by

Jorge Documet

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOMEDICAL ENGINEERING)

August 2009

Copyright 2009 Jorge Documet

ii
Dedication
I would like to dedicate this manuscript to my wife Cintya for all her support and
understanding; as well as to my kids Andre, Yamna and Pia for being my inspiration and
motivation every day.
I would also like to dedicate this manuscript to my parents Jorge and Maria for all their
continuous advice and enlightment throught my education years.
Last but not least, I would like to thank my siblings Ursula, Daniel, Luis and Jessica as
well as my relatives, colleagues and friends who were continuously providing me the
encouragement needed during my carrer years.
iii
Acknowledgments
This work had been done with the partial contribution and collaboration of the following
collegues and friends:
Aifeng Zhang, PhD IRB applications submission
Anh Le, MS ePR and Gateway Servers installation, Database, Integration Unit
Brent Liu, PhD Advisor
Hanyi Wang, BS Fingerprint Module, Pre-Op Authoring Module
James Sayre, PhD Outcome Analysis
Kevin Wang, BS Integration Unit
Mark Haney, MD Integration Unit
Robert Chinn, BS Integration Unit
Sinchai Tsao, BS Graphical User Interface, Pre-Op Authoring Module
In addition, John Chiu, MD, D.Sc., Chief Surgeon and Director of California Spine
Institute (CSI) provided clinical suggestions, advice, and input; as well as user
requirements and feedback.
A very special acknowledgment for Prof. H.K. (Bernie) Huang D.Sc. for his vision and
guidance.
iv
Table of Contents

Dedication ii
Acknowledgements iii
List of Tables vii
List of Figures viii
Abbreviations x
Abstract xii
Chapter 1: Introduction 1
1.1 Significance 2
1.2 Current Challenges in Image-Assisted Minimally Invasive Spinal 3
Surgery
1.3 General workflow for Image-Assisted Minimally Invasive Spinal 3
Surgery
1.4 Need for data integration 6
1.4.1 Introduction to ePR 6
1.4.2 The ePR System for IA-MISS 7

Chapter 2: Image-Assisted Minimally Invasive Spinal Surgery Workflow and 11
Data Model
2.1 Data model and standards used 12
2.1.1 The Data model 12
2.1.2 Standards used 14
2.1.2.1 DICOM (Digital Imaging and Communications in 15
Medicine)
2.1.2.2 HTTPS (Hypertext Transfer Protocol Secured) 15
2.1.2.3 JPEG (Joint Photographic Expert Group), GIF (Graphics 15
Interchange Format) and PNG (Portable Network Graphics)
2.1.2.4 RS-232 (Recommended Standard 232) 15
2.2 ePR Data flow 16

Chapter 3: Image-Assisted Minimally Invasive Spinal Surgery (IA-MISS) ePR 19
System Architecture
3.1 The Overall IA-MISS ePR System Architecture 19
3.2 Four Major Components in the IA-MISS ePR System 20
3.2.1 Integration Unit (IU) 21
3.2.1.1 Input Data and its Archive 21
3.2.1.2 Out of Range Input Data Alert Message 22
v
3.2.2 The Continuous-Available Gateway server 23
3.2.2.1 Pre-Op Stage 23
3.2.2.2 Intra-Op Stage 24
3.2.2.3 Post-Op Stage 24
3.2.3 The Continuous-Available ePR Server 24
3.2.3.1 Data Storage and Archive and System Database 25
3.2.3.1.1 Input Data and Metadata 25
3.2.3.1.2 Database Schema 27
3.2.3.2 System Security 27
3.2.3.3 System Continuous Availability and Failover 28
3.2.4 Visualization and Display 28

Chapter 4: Pre-Op Authoring Module 30
4.1 Workflow analysis 30
4.2 Participants 31
4.3 Significance of Pre-Op Data Organization 32
4.3.1 Organization of the Pre-Op Data 32
4.3.2. Surgical Whiteboard Data 32
4.4 Graphical User interface 33
4.4.1 Editing 33
4.4.1.1 To Create a Procedure 33
4.4.1.2 To Perform editing 34
4.4.2 The Neuro-Navigator Tool for Image Correlation 35
4.4.3 Display 36

Chapter 5: Intra-Op Module 37
5.1 The Intra-Op Module 37
5.2 Participants 37
5.3 Data Acquired During Surgery 38
5.4 Internal architecture of the Integration Unit (IU) 39
5.5 Interaction with the gateway 41
5.6 Graphical User Interface 41
5.6.1 Rule-based alert mechanism 42

Chapter 6: Post-Op Module 43
6.1 Post-Op Module 43
6.2 Participants 43
6.3 Patient in the Recovery Area 43
6.4 Post-Op Documentation 44
6.4.1 Graphical User Interface (GUI) 44
6.5 Follow-up Pain Surveys 45

Chapter 7: System Deployment 46
7.1 Implementation schedule 46
vi
7.1.1 Planning and design phase 46
7.1.2. Hardware installation 47
7.1.3. Software installation 48
7.1.4 Training 49
7.2 Pitfalls and Challenges during delivery of the system 49
7.3 Training and support to users 51
7.3.1. Users trained 52

Chapter 8: Results 54
8.1 Statistical analysis of data collected before the installation of the ePR 54
System - Only Intra-Op times were recorded
8.2 Preliminary analysis of data collected after the installation of the ePR 55
System – Only Intra-Op times were recorded
8.3 Discussion 56
8.3.1 Data Collection 56
8.3.2 Preliminary Observation 57

Chapter 9: Current Status and Future Plans 60
9.1 Current stage of the deployment of the ePR for IA-MISS at the 60
California Spine Institute (CSI)
9.2 Research and Development After the Prototype 61
9.2.1 Integration Unit Prototype: Version 2 61
9.2.2 Difference between the current IU prototype and Version 2 61
9.3 IRB for patient outcome analysis 62

References 64
vii
List of Tables

Table 3.1 - Default values for the safe ranges of the vital signs 23
Table 8.1 - Summary of Cervical cases with 2 discs operated before the installation 54
of the ePR System

Table 8.2 - Summary of Lumbar cases with 2 discs operated before the installation 54
of the ePR System

Table 8.3 - List of 8 Cervical cases with multiple discs operated collected after the 55
installation of the ePR System

Table 8.4 - List of 16 Lumbar cases with multiple discs operated collected after the 56
installation of the ePR System

viii
List of Figures

Figure 1.1 - The Workflow of the IA-MISS procedure showing all the different 4
stages: before surgery, during the surgery (including the preparation)
and post surgery.

Figure 1.2 - The pre and post digital images for endoscopic discetomies performed 6
on the lumbar, cervical and thoracic vertebras. Arrows show the
locations of the lesions before and after the IA-MISS procedure.

Figure 1.3 - Top (Figure 1.3a) Generic Operating Room Layout with Personnel 10
and Live Surgical Monitoring and Imaging Equipment. Bottom
(Figure 1.3b) Proposed Locations of the Pre-op and Intra-op
Image/Data on two Opposite Side Large Screens in the OR.

Figure 2.1 - The Algorithm of Spine Care for degenerative and herniated spinal 11
discs, and spinal stenosis. The proposed research will move us closer to
the third bracket using IA-MISS Operation.

Figure 2.2 - Part of the DICOM data model used for this ePR. The diagram above 13
shows the relationship among the patient, study, series and image. The
DICOM data model in reality is more extensible that the one shown
above, which is only displaying the entities used in the proposed ePR.

Figure 2.3 - The data model (or database schema) of the ePR system. It extends the 14
schema of DICOM to accommodate surgical information including live
waveform and several standard surgical forms.

Figure 2.4 - Dataflow of IA-MISS ePR System. 16

Figure 3.1 - The ePR system architecture showing three operation phases: Pre-Op, 20
Intra-Op and Post-Op (Left); as well as four operation modules, some
modules are bundled up together for ease of data transfer and
Continuous-Available back-up. The arrows show the data flow during
the three phases of operation. The outside light gray color side-way “U”
band is the Display module backbone with five subunits. Inside the
opening of the “U” in dark gray are the Integration Unit (IU),
Continuous-Available Gateway, and Continuous-Available ePR Server.
Within the Gateway and the ePR Server, the Database and Filesystem
software are interrelated and shared by both components.

ix
Figure 3.2 - The Dual-system back-up Schema with two hardware pieces: ePR 26
Server hardware and Gateway hardware. Each hardware piece has two
softwares: ePR Server software, and Gateway software; and a tandem
database with hard drive for data and metadata archive.

Figure 3.3 - The home page of the ePR system showing three surgical stages: 29
Pre-Op, Intra-Op and Post-Op.

Figure 4.1 - The Pre-Op authoring module page. The left text list depicts the surgical 34
data model showing the studies and procedures. After the user clicks an
item in the list, the proper image, i.e a sagittal MRI would be shown on
the right.

Figure 4.2 - The Neuro-navigator tool allows the correlation of the position of the 35
lesion in the sagittal (left) and the axial view (right).

Figure 4.3 - The Pre-Op display organized during patient consultation as seen on the 36
Pre-OP display monitor in the OR during Intra-OP. Top Text Row:
Patient General Information, Second Text Row: Whiteboard
information, Center: Images and annotation during Pre-Op consultation.

Figure 5.1 - The hardware and software interconnectivity diagram for a surgical 40
procedure using the ePR system. The IU in the middle accepts different
input devices and sends them to the Intra-Op Live Display.

Figure 5.2 - A mock-up example of the Intra-Op Live Display as seen on the 42
Intra-Op large monitor in OR. Top row: Waveforms of six vital signs,
BIS, and IVF, the horizontal axis is time. Middle row: Waveform of
EMG, Fluoroscopic image, and endoscopic image. Bottom row: Laser
output values.

Figure 6.1 - The Post-Op authoring module displaying a Post-Op document showing 45
data acquired during the surgery. This page can be synchronized with
the surgeon Post-Op dictation.

Figure 7.1 - Server installation and final location of servers at the Server Room at 47
CSI.

Figure 7.2 - Integration Unit installed at the OR with the different input sources 48
connected as well.

Figure 7.3 - A training session with clinical staff at the California Spine Institute. 52
General workflow explanation and Pre-Op authoring module were
introduced.
x
Abbreviations

API Application Program Interface
BIS Bispectral Index System
CO
2
Carbon dioxide
CR Computed Radiography
CSI California Spine Institute
CSS Cascading Style Sheet
CT Computed Tomography
DICOM Digital Imaging and Communications in Medicine
EMG Electromyography
ePR Electronic Patient Record
GIF Graphics Interchange Format
GUI Graphical User Interface
HIPAA Health Insurance Portability and Accountability Act
HIS Hospital Information System
HTML HyperText Markup Language
HTTP HyperText Transfer Protocol
HTTPS HyperText Transfer Protocol Secured
ICT Information and Communication Technology
IA-MISS Image-Assisted Minimally Invasive Spinal Surgery
IPILab Image Processing and Informatics Laboratory
IRB Institutional Review Board
xi
IT Information Technology
IU Integration Unit
IVF Intravenous Fluid
JPEG Joint Photographic Expert Group
LCD Liquid Crystal Display
MB Megabytes
mmHg Millimeters of Mercury
MRI Magnetic Resonance Imaging
OR Operating Room
PACS Picture Archiving and Communication System
PHP PHP: Hypertext Preprocessor
PNG Portable Network Graphics
RAM Random Access Memory
RIS Radiology Information System
RS232 Recommended Standard 232
SDK Software Development Kit
TB Terabytes
USC University of Southern California
VAHE Veterans Affairs Healthcare Enterprise
VAS Visual Analog Scale
VGA Video Graphics Array
xii
Abstract

Recent developments in medical imaging informatics have improved clinical workflow in
Radiology enterprise. However, there still remains gaps in the clinical continuum from
diagnosis to surgical treatment through post-operative follow-up that can be addressed by
a variety of advanced technologies. One solution is the development of an electronic
patient record (ePR) that integrates key imaging and informatics data during the pre,
intra, and post-operative phases of clinical workflow. One application is in image-guided
minimally invasive spinal surgery (IA-MISS) where spinal discectomy procedures are
performed for decompressing nerve roots affected by spinal disc protrusions. This
procedure utilizes a variety of still and real-time acquisition systems including X-Ray,
CT, MRI, digital fluoroscopy and digital endoscopic video. The integration of these data
together with waveform and other related informatics data is necessary during the entire
surgical procedure for evaluation, treatment planning, and review.
This dissertation presents an ePR System design and implementation tailored to IA-MISS
but it can also be expanded to other surgical procedures. The ePR System is currently
being utilized for more than 6 months at the California Spine Institute (CSI) at Thousand
Oaks, California.
1
Chapter 1
Introduction
Image-Assisted Minimally Invasive Spinal Surgery (IA-MISS) is a microdecompressive
spinal discectomy procedure for decompressing nerve roots constricted by spinal disc
protrusions. This image-assisted procedure utilizes a variety of still and live real-time
imaging devices and methods including X-Ray, CT, MRI, 3-D modeling, digital
fluoroscopy and digital endoscopic video. These techniques, when integrated, provide
magnification, guidance and real-time course intervention to assist the surgeon with the
insertion of a small tube (6 mm diameter) into the disc to remove its offending portion.
IA-MISS is different from standard spinal disc surgery because there is no traumatic
muscle dissection, bone removal, or bone fusion. The incision is tiny enough to close
with sutures and then covered with a small band-aid. Therefore, most of the
complications that occur with conventional spine surgery are virtually eliminated with the
IA-MISS procedure. Over the past eight years, there have been stepwise technical and
clinical advances in IA-MISS demonstrating the potential for significantly improved
spinal surgical outcomes. However, there still remain large gaps along the clinical
continuum from diagnosis to surgical treatment to postoperative follow-up that can be
more fully addressed by the integration of a variety of advanced medical and imaging
informatics technologies. The following paragraphs in this chapter will first introduce the
IA-MISS workflow, data utilized, and the challenges relating to this clinical procedure.
Then, a general introduction to the concept of the image-intensive electronic patient
record (ePR) will be presented as a potential solution for IA-MISS and related issues.
2

1.1 Significance
Back and neck pain is a common occurrence with human beings due to poor posture,
prolonged sitting, lifting, repeated bending, obesity, and injuries from accidents.
Approximately 85% of the Western world are afflicted with some degree of back or neck
pain at some point in their lives [1]. In fact, 31 million Americans experience low-back
pain at any given time [2]. About 25% of the United States population has been
incapacitated for two weeks or more due to back pain and an estimated 8 to 10 million
people have had permanent disability as a result. [3, 4, 5, 6] The economic impact is
obvious. In most cases, simple treatments such as bed rest, exercise, physiotherapy, and
pain medication bring relief. However, if one or more of the vertebral discs ruptures and
presses on nerve roots, the pain radiating from the back or neck and down the limbs can
be incapacitating and severe. Until recently, the only treatment was surgical removal of
parts of the ruptured disc, a major operation that required general anesthesia, the
dissection of muscle, removal of bone, manipulation of nerve roots, and, at times, bone
fusion. This surgical procedure takes several hours followed by between 2 to 6 months of
convalescence and rehabilitation. In an effort to overcome the disadvantages of
traditional surgical techniques, the scientific medical community began exploring the use
of endoscopy (arthroscopy) during surgical procedures. An endoscope provides clear
visualization and magnification of deep structures. This technology, first used in knee
surgery, has been astonishingly successful in relieving pain. Because of advanced
scientific technology and miniaturization, including fiber optics, video imaging
3
technology, laser treatment and experience gained through minimally invasive spinal
surgery, there is a discectomy procedure that is less traumatic for some patients with disc
problems. In the recent years, development of this image-Assisted surgical procedure has
improved the treatment precision and reduced surgical tissue trauma.
1.2 Current Challenges in Image-Assisted Minimally Invasive Spinal Surgery
Despite the overall benefits from IA-MISS in terms of recovery time and sucessful
patient outcomes, there are still challenges remaining that need to be addressed and are
described as follows:
1) Current scattered systems within the operating room (OR) including multiple sources
of images, video, and waveforms. This is similar to general surgery.
2) The need for enhanced workflow with data acquisition, management, and distribution
all in a single system.
3) The need for an integrated data repository in a one-stop source during all stages of the
surgical workflow (eg, before, during, and after).
4) The need to develop outcomes analysis for patients undergong IA-MISS since it is a
relatively new field of expertise.
5) The need for training of new adopters in IA-MISS.
1.3 General workflow for Image-Assisted Minimally Invasive Spinal Surgery
The following paragraphs describe the general workflow with the steps involved in a
typical IA-MISS procedure. Figure 1.1 depicts this workflow [7].
4

Figure 1.1 - The Workflow of the IA-MISS procedure showing all the different stages: before
surgery, during the surgery (including the preparation) and post surgery.
The workflow can be broken down into three phases: 1) Before surgery; 2) During
surgery (including the preparation) 3) Post surgery which will be discussed below.
• Before surgery (Pre-Op): This phase is the workflow involved prior to the actual
surgical procedure. In the Pre-Op workflow, usually the patient presents with a
problem and is evaluated by the physician to determine whether IA-MISS is
needed and whether it would be helpful to the patient. If this case is true, then a
procedure is scheduled. At this stage the surgeon or surgeons in combination with
the physician assistant plan the surgical procedure using digital diagnostic images
such as CR, CT and MRI. In addition to the information obtained from the
medical studies, the patients also fill out a set of surveys that determine the level
of pain that they feel.
• During surgery (Intra-Op): During the surgical procedure, the surgeon(s) operate
on the different disc(s) that need to be corrected. While operating, there is a
significant amount of data being acquired that help monitor the body response of
the patient to the procedure. This includes video and image data acquired with the
endoscope. A single vertebrae procedure usually lasts 30 minutes on average.
5
• After surgery (Post-Op): During this stage the patient recovers from surgery. The
patient is continuously monitored in the recovery area to assure all vitals signs are
stable. In addition, a set of tests are also performed to assess the outcome of the
surgical procedure which includes an additional set of forms that the patient fills
out.
The recovery period after surgery lasts from 45 minutes to one hour. The patient is then
discharged. Therapy can begin the next day and the patient can go back to work within 2
to 3 days.
Figure 1.2 depicts three cases of IA-MISS: one lumbar, one cervical and one thoracic.
For each of these cases, the upper row shows the Pre-Op diagnosis using MRI and the
lower row shows their corresponding image after the surgical procedure (Post-Op). The
illustration also depicts annotations that pinpoint the location of the lesions obtained from
pateint pre-surgery consultation (arrrows). It can be seen that after the procedure (Post-
Op) the lesions are no longer present from the post-surgery annotations (arrows). All
these cases were treated with percutaneous disectomies. From the figure, it is obvious the
drastic improvement that the patient encounters after a successful IA-MISS procedure.
6

Figure 1.2 - The pre and post digital images for endoscopic discetomies performed on the lumbar,
cervical and thoracic vertebras. Arrows show the locations of the lesions before and after the IA-
MISS procedure.
1.4 Need for data integration
As mentioned in section 1.2 there exists challenges that need to be addressed in order to
improve the overall workflow efficiency of IA-MISS and patient outcomes. One such
concept that addresses these challenges and shed some light on possible advancements
within the field is the concept of an image-intensive ePR (Electronic Patient Record)
system with image distribution which will be described in the following paragraphs.
1.4.1 Introduction to ePR
In general, an ePR is an effort to re-design HIS (Hospital Information System) with
patient as the focus [8]. It tries to bring all necessary information across different
departments in hospitals to a single interface. It stores textual and graphical data as well
7
as images. Tradionally, the former has been in terms of reports, patient demographics,
clinical test results, to mention a few; and the latter in terms of digital images such as CT,
MR, CR, X-ray C-arm fluorography, and endoscopy.
The components of an ePR include an information model, a clinical data repository, a
web-based application for users, along with a security model and built-in decision
support. The inclusion of imaging data and built-in decision support makes the ePR
stand out amongst general clinical information systems such as HIS and RIS (Hospital
and Radiology Information Systems). Furthermore, some of the benefits of adopting an
ePR as a repository of clinical records of patients include an increase in the efficiency of
patient care and ease-of-use to find information about the patients. In addition, the
imaging data within the ePR data model has opened new doors to the possibility of
improvement in clinical decision outcomes of the future. The United States Department
of Veterans Affairs developed the first Healthcare Enterprise (VAHE) information
system, Vista, ten years ago for their enterprise-level ePR integrated with images. The
Hong Kong Hosptial Authority implemented the most advanced ePR with image
distribution starting in 2003 until now for their 44 enterprise-level hospitals.
1.4.2 The ePR System for IA-MISS
As described in the previous section, an ePR serves the purpose of integrating clinical
data into a single source, by combining data from different sources to a centralized server.
Even though ePR servers have been implemented by initiatives coming from different
departments, such as radiology or cardiology; there is currently no such ePR system that
has been specifically tailored for surgical data. This concept was first proposed in 2001
8
[7], but required technologies that were not readily available for system design and
clinical implementation. Thus, one contribution in my dissertaiton is to create an ePR
system, from design to clinical implementation, containing pertinent Pre-Op; Intra-Op
real-time data, wavefroms and images; and Post-Op information, particularly for the IA-
MISS procedure.
The proposed ePR for IA-MISS has been designed to overcome the challenges presented
in section 1.2 which are currently not being designed or implemented by any available
system, either from the research arena or as a commercial product.
1) Scattered systems in the OR: The proposed ePR acquires all Pre-Op, Intra-Op, and
Post-Op pertinent available data and presents them on two organized large LCD
monitors, one for Pre-Op and the second for real-time Intra-Op. The data is also
organized and saved in the ePR based on the DICOM (Digital Imaging and
Communications in Medicine) data model. This challenge is covered in the detailed
description of the Intra-Op module in Chapter 5.
2) The need for enhanced workflow: The proposed ePR can perform workflow analysis
of a surgical procedure in OR. Additionally, it provides the necessary infrastructure
to properly acquire, manage and distribute all contents to the users. This challenge is
covered in detail by chapters 3 through 6.
3) The need for an integrated data repository: The proposed ePR keeps all relevant
information from Pre-Op, Intra-Op, and Post-Op, and stores the data in a database
with a filesystem. This is explained in chapter 3.
9
4) The need for training of new personnel for IA-MISS: The proposed ePR is aimed to
be a simple-to-use but powerful application that will make the utilization of IA-MISS
more attractive for general surgery personnel to adopt. This ePR is currently being
implemented at a clinical site. More details are presented in chapter 7.
5) The need to develop outcomes analysis: The proposed ePR brings a new and unique
application in IA-MISS surgery that will allow patient surgical outcomes analysis.
Actual data collection for the outcomes analysis is in progress. Some of this is
described in chapter 9.
Figure 1.3 shows the concept of my disseration in which the ePR system will transform
the current IA-MISS setup at the OR shown in Figure 1.3a to the ePR system depicted in
Figure 1.3b

10

Operating Table

136 Endoscope
Displa /
Storag

142 Laser
Generator

138 EEG/
Displa

2800 mm .

120 Large
screen intra-
op image/data

143 Selected
Imaging/ dictation
system

133 Video
Mixing
Equipmen

132 Surgical
Video
Camer /
Displa

141 EKG/
Displa

139
signs and
Displa

137 Authoring
document
module

Fluoroscopic
Display
/
Storage
134 C-ARM
-
Surgical
Instrument
table
110 Large
screen Pre-
op

140
EMG/
Displa
11

135
Pt Biom
131
Neuro
Physio
(SSEP)
133 Fluid
Intake/
Output
Pre-OP 52” LCD
Intra-op 52” LCD
1) Anesthesiologist
2) Assistant
3) Surgeon
4) Scrub nurse
5) Circulator

1
2 3 4
5

Figure 1.3 - Top (Figure 1.3a) Generic Operating Room Layout with Personnel and Live Surgical
Monitoring and Imaging Equipment. Bottom (Figure 1.3b) Proposed Locations of the Pre-op and
Intra-op Image/Data on two Opposite Side Large Screens in the OR.
Current Digital Endoscopic OR suite facility
Courtesy of : Dr. John Chiu
MD’s
Staff
RN, Tech
EMG Monitoring
C-Arm Fluoroscopy
MRI Image - PACS
C-Arm Images
Image Manager -
Report
Video Endoscopy
Monitor
EEG Monitoring
Left side of OR
Image view boxes
Teleconferencing
- telesurgery
Laser
generator
11
Chapter 2
Image-Assisted Minimally Invasive Spinal Surgery Workflow and Data
Model
Image-Assisted Minimally Invasive Spinal Surgery (IA-MISS) is a technique applied to
treat cases of herniated lumbar discs, post fusion junctional disc herniation, neural
compression, osteophytes, spinal stenosis, vertebral compression fractures, spinal tumor,
synovial cysts and other types of spinal traumas. Over the last few decades its popularity
has been gaining more prevalance as a preferred option of treatment due to the benefits
that it provides against open surgery as explained in chapter 1. According to medical
professionals endoscopic IA-MISS procedures are expected to increase in number to
85% for spinal surgery cases [9].
Figure 2.1 shows the decision tree for treatment of spinal pain cases where the main goal
of patient outcomes is to favor less traumatic treatment alternatives. Among the options
presented here when a patient requires a surgical procedure, the most preferable is IA-
MISS, then Spinal Arthroplasty, followed by the last resort option of open surgery.
Figure 2.1 - The Algorithm of Spine Care for degenerative and herniated spinal discs, and spinal
stenosis. The proposed research will move us closer to the third bracket using IA-MISS
Operation.

Redefining Algorithm in Spine Care: MISS
Pain
Management
Conservative
Treatment
Image-Assisted
MIS Surgery
Open Spinal
Surgery
The last resort
Spinal Arthroplasty
Disc Replacement
Artificial Disc
12
2.1 Data model and standards used
After the general workflow for IA-MISS (presented in section 1.3) was identified, the
next step for the ePR system was to design a data model and identify the standards to be
used.
2.1.1 The Data model
The data model for the proposed ePR for IA-MISS has been designed to extend the
DICOM data model due to the similarities that the data has with currrent medical imaging
data utilizing DICOM. The DICOM data model is a good basis for data modeling due to
the following:
1) DICOM data model is the de-facto standard for the storage, handling, and distribution
of medical image and imaging informatics data and
2) There is a new DICOM working group for Surgery [9] which utilizes the existing
DICOM data model as well, and is a concrete example of the importance of following
trends towards standardized data.
From the DICOM data model the ePR for IA-MISS follows the relationship among the
patient, the medical studies, their series and images as seen in Figure 2.2.

13

Figure 2.2 - Part of the DICOM data model used for this ePR. The diagram above shows the
relationship among the patient, study, series and image. The DICOM data model in reality is
more extensible that the one shown above, which is only displaying the entities used in the
proposed ePR.

The final data model utilized for the proposed ePR includes entities that describe the data
required for IA-MISS. These new elements that are added to the data model include the
surgical procedure type, waveforms, the key image, survey forms for pain, and user
information for access to the system. The data model is shown in Figure 2.3 and this data
model will be explained in greater detail in the chapter 3: System Architecture.
14

Figure 2.3 - The data model (or database schema) of the ePR system. It extends the schema of
DICOM to accommodate surgical information including live waveform and several standard
surgical forms.
2.1.2 Standards used
Interoperability of devices is important when data is to be shared, accessed, and stored in
an efficient and convenient manner. In addition, standards are a comprehensive set of
rules that allow devices and especially software components to communicate between
each other. Thus, standards become crucial for a robust system integration that will
eventually lead to reduced costs, risks, and developmental time.
The proposed ePR has been designed with the concept of utilizing available standards
whenever possible. Currently we are using the following standards:

15
2.1.2.1 DICOM (Digital Imaging and Communications in Medicine)
As previously mentioned in the ePR data model design, DICOM is the de facto standard
for communications in medicine for medical images. Even though it was originally
mainly utilized by Radiology, other clinical departments are beginning to utilize it on a
daily basis. For the proposed ePR system the DICOM standard is used as the
communication protocol when receiving images from the PACS (Picture Archiving and
Communication System) archive in addition to the aforementioned data model design.
2.1.2.2 HTTPS (Hypertext Transfer Protocol Secured)
The ePR itself is a web application that allows users to obtain clinical information about
patients from a single interface. The users connect to the ePR server and utilize a client
browser, (ie Internet Explorer or Mozilla Firefox). All the communication are established
using HTTPS for security reasons.
2.1.2.3 JPEG (Joint Photographic Expert Group), GIF (Graphics Interchange Format)
and PNG (Portable Network Graphics)
Due to the web-based nature of the ePR and the lack of native support of DICOM images
on the web browser, all images are converted to formats web browsers can understand
and then presented properly to the clients. The format of choice for medical images is
jpeg. For other type of images such as icons the gif or png formats are used.
2.1.2.4 RS-232 (Recommended Standard 232)
Commonly know as the standard to obtain data out of a serial port. This standard is used
by some of the peripheral devices on the OR to transmit data out, however, even though
the transmission is standard, the protocols from machine to machine can vary
16
significantly. Thus, it is important to note that currently the vendors do not have
implemented a common standard protocol to share the acquired data (data point values
and video streams) for the devices within the OR. This has been a major challenge in the
clinical environment and one of the major reasons for this research work in system
integration and ePR development.
2.2 ePR Data flow
The initial Data flow utilized by the ePR system was based on the concept of the
workflow presented in section 1.3 and it is shown in Figure 2.4.

Figure 2.4 - Dataflow of IA-MISS ePR System.

The figure above shows the main components used in this application: The Pre-Op
Image/Data module, the Intra-Op module, the Post-Op module, the Continuous-Available
Gateway, the Continuous-Available ePR and the visualization and display module. These
components will be explained in more detail in chapter 3 when the system architecture is
presented.
Pre-Op
Historical
1

1) Archive/database

2)Monitoring
Module
Intra-Op (IU)
Real-time
5
Post-Op
6
9
11
10
8
Pre-Op Input
Gateway
Intra-Op Input
Gateway
Post-Op Input
Gateway
Continuos
Available ePR
Server
2
4
7
3

Pre-Op Image/Data
Post-Op authoring and
Display
Intra-Op Image/Data

Post-Op Data
Visualization &
Display module
Continuous
Available
Gateway Server
Pre-Op data Display
Intra-Op image/data
Display
17
The following paragraphs will describe the data workflow in detail as well as introducing
the major components of the ePR system that will be involved with the data workflow.
Numerical Workflow Steps in the paragraphs refer to the numbered labels of each arrow
line in Figure 2.4.
Pre-Op Workflow
1) Historical medical imaging studies are acquired from the PACS. The studies being
retrieved are in the DICOM format.
2) The Gateway Server, which is a component of the ePR system, receives the DICOM
images and processes them accordingly. The original image is kept in the ePR and a jpeg
version is utilized for display purposes via the web interface of the ePR. All DICOM
header information and metadata is extracted and recorded in the database.
3) and 4) Pre-Op authoring is performed by the surgeon(s) and the physician assistants.
The surgical procedure information is entered into the ePR. At this stage the patient’s
survey pain forms are also entered into the system. The surgeon selects some key images
and authors annotations overlayed on the key images that will ultimately be utilized
during surgery.
Intra-Op Workflow
5) The Integration Unit (IU) is connected to the clinical devices in the OR and
continuously gathers live data signals from them during the entire surgical procedure.
After the surgical procedure has ended, the IU sends all data points and surgical images
to the Gateway server. Chapter 5 explains in greater detail the functionality of the IU. All
values collected in addition with the endoscopic video stream and the C-ARM captures
18
are displayed during the surgical procedure in real-time on a large 52-inch digital display
in the OR that is visible by all personnel in the operation.
6) The Gateway Server receives the data from the IU and stores the data values and
images at the database and filesystem respectively.
7) The images selected at steps 3) and 4) during Pre-Op in combination with the
demographics and clinical history of the patient are displayed in the OR utilizing a
second 52-inch digital display.
Post-Op stage
8) While the patient is in the recovery area, the system continues gathering some vital
signs that are transferred to the Gateway Server.
9) The Gateway Server receives the data sent from the previous step and stores the data
values of the signals received into the database.
10) and 11) These steps are performed using the Post-Op authoring module that allows
the surgeon to create a final report out of the data gathered during the pre-, intra-, and
Post-Op stages. The final report will be kept in digital format at the ePR as the patients’
permanent surgical record.

19
Chapter 3
Image-Assisted Minimally Invasive Spinal Surgery (IA-MISS) ePR
System Architecture
3.1 The Overall IA-MISS ePR System Architecture
Following Figure 2.4 which depicts the workflow and data flow model of the IA-MISS
ePR system represented by a 3 x 4 dimension matrix model. The three rows are Pre-Op,
Intra-Op, and Post-Op workflow process stages; and the four columns are the input data
integration unit (IU), input gateway, ePR server, and Image/Data display. From the
system architecture point of view, the ePR system should be designed for efficiency,
effectiveness, and reliability of system operations. For these reasons, although system
workflow is separated into Pre-Op, Intra-Op, and Post-Op stages, some modules which
handle multiple workflow stages may be combined to share system workload and
reliability. For example, the Continuous-Availability requirement of each component in
the system is better designed to support other existing components of the system for easy
system back-up and cost containment. Also, although there are four major components
and three operational workflow phases in the ePR system, many of the software have
similar design backbones, and some software may be bundled together for easier
programming effort and faster system execution time. The ePR System architecture has
been reorganized and modified from the dataflow shown in Figure 2.4 to Figure 3.1 for
better system implementation and operation. Detailed examination of these two figures
demonstrate the existence of four components and three operational workflow phases.
However some components in Figure 3.1 are bundled together for design and operational
20
convenience. This chapter describes the system design and architecture. Chapters 4, 5,
and 6, present the detailed technical descriptions of the Pre-Op, Intra-Op and the Post-Op
modules, respectively, based on the system architecture shown in Figure 3.1 for
consistency.
Figure 3.1 - The ePR system architecture showing three operation phases: Pre-Op, Intra-Op and
Post-Op (Left); as well as four operation modules, some modules are bundled up together for ease
of data transfer and Continuous-Available back-up. The arrows show the data flow during the
three phases of operation. The outside light gray color side-way “U” band is the Display module
backbone with five subunits. Inside the opening of the “U” in dark gray are the Integration Unit
(IU), Continuous-Available Gateway, and Continuous-Available ePR Server. Within the Gateway
and the ePR Server, the Database and Filesystem software are interrelated and shared by both
components.

3.2 Four Major Components in the IA-MISS ePR System
The four major components in the IA-MISS ePR System are: 1) Data Input Integration
Unit (IU); 2) Continuous-Available Gateway server; 3) Continuous-Available ePR
Server, and 4) Visualization and Display. Both the Input Gateway and the ePR Server
include data storage and archive, system database, system security, system continuous
21
availability and failover.The GUI and display module resides within the ePR Server. All
data input systems like medical imaging, surgical video, vital signs waveform recorders,
and textual data recorder generate Pre-Op, Intra-Op, and Post-Op data, and they are all
categorized as input data. The imaging and data systems that generate information are
existing peripheral surgical supported equipment already within the OR but they do not
belong to the IA-MISS ePR system. However, the ePR system must integrate these
systems in order to receive the input data that is acquired before, during, and after surgery
to support the surgical procedure.
3.2.1 Integration Unit (IU)
3.2.1.1 Input Data and its Archive
This component is responsible for acquiring all data from different peripheral devices that
are presented in the OR during surgery (Intra-Op) that continuously measure all live vital
signs, waveform signals, and surgical related images of the patient undergoing a
procedure. The data acquired by the IU from all input devices are synchronized through a
master clock and displayed live onto a customized interface using a 52” LCD (Liquid
Crystal Display) screen (called Intra-Op Live Display) in the OR. The data gathered
during surgery include the following:
W Digital C-ARM fluorographic images
W Digital endoscopic video images
W Surgical video

22
W Waveform signals: EMG (Electromyography), BIS (Bispectral Index), and Vitals
(Blood Pressure, Heart Rate, Respiratory Rate, PulseOX, Body Temperature and
Partial Pressure of Carbon Dioxide).
The images, videos and data points mentioned above are transferred automatically and
continuously from the various input sources of the different data systems in the OR
during operation that are attached to the data input IU. The data is immediately saved
into IU memory. The IU software displays the waveforms, images, and streamed videos
properly every second (which is a default value) on the large Intra-Op LCD, and also
makes a copy from the memory to the IU local hard drive with 1.5 TB (Terabytes) of
storage space every five seconds (which is also a default value). These two default values
can be adjusted interactively depending on clinical demands.
Normal Procedures for a single vertebra surgery takes about 30 minutes on average. This
data is sent continuously to the Gateway where the images are processed if needed and
then placed in a data folder shared with the ePR server where they will be permanently
archived. The data values are also extracted and saved to the ePR system database.
3.2.1.2 Out of Range Input Data Alert Message
In addition to the one second input display described in the last section, the IU features a
rule-based alert-software that checks each input waveform for data that is out of the
normal range. The IU has a set of rules based on clinical accepted medical practice that
determines when a given signal is considered within the normal range for a patient (See
Table 3.1). If at any given time during the surgical procedure, a signal falls outside the
safe range, the IU will trigger an alert message on the Intra-Op Live display (See Section
23
3.2.4 Display). This assists the surgeon and key personnel in the OR to take necessary
actions during the surgical procedure. The values shown in Table 3.1 are considered
normal values within the safe ranges for the signals used in the IU. It is noted that those
values might not be considered normal for all patients; thus, during the Pre-Op patient
consultation time, these default values need to be revised and properly adjusted as
necessary.
Table 3.1 - Default values for the safe ranges of the vital signs
Signal Lower Value Upper Value Units
Blood Pressure 80 120 mmHg
Partial Pressure 35 45 mmHg
Heart Rate 60 100 Beats/minute
Respiratory Rate 10 16 Breaths/minute
PulseOX 92 %
BIS 40 70 Score value
Body Temperature 36.1 38 Celcius

3.2.2 The Continuous-Available Gateway server
The functions of the Input gateway are receiving, staging, managing, and transferring
input data during the three workflow stages of the surgery: Pre-Op, Intra-Op, and Post-
Op.
3.2.2.1 Pre-Op Stage
The gateway receives DICOM images and diagnostic reports from PACS. Once images
are received by the gateway, a Pre-Op script is automatically launched by the gateway to
properly extract all the information from the headers of the DICOM files. This data is
then saved into the database. This whole process is automated at the gateway and does
not require any user intervention.
24
3.2.2.2 Intra-Op Stage
During Intra-Op, the gateway receives live data as described in Section 3.2.1 from the IU
using an API (Application Program Interface). The transfer protocol used is the HTTPS
(HyperText Transfer Protocol Secure) Standard. Before any data is sent to the Gateway,
the IU needs to properly authenticate itself in order to avoid conflict with other possible
input devices. Once the data is received by the Gateway server, the API will place the
data in a specific location in the ePR where a script will be executed to process the data
accordingly.
3.2.2.3 Post-Op Stage
During Post-Op, the patient is under observation in the recovery area by a nurse and the
surgeon. The vital signs and other monitoring equipment are used to evaluate the patient
Post-Op condition. During the one-half to one hour observation, live data of the patient is
continuously received and displayed at the bedside monitor by the Post-Op module.
3.2.3 The Continuous-Available ePR Server
The ePR Server is the heart of the IA-MISS ePR System, the basic components are
shown in Figure 3.2. The ePR Server is the front-end of the system where the users will
login to perform all the necessary tasks during the surgical workflow. The ePR Server
allows access to the Pre-Op authoring module, the Pre-Op display in the OR and the Post-
Op authoring module (see Figure 3.1). Administrative tasks such as giving the users
access to the system, registration of patient information, scheduling, fingerprint
registration and identification, among others are also included.
25
The ePR by definition allows the participants to obtain any necessary information about
the patient from a single interface, i.e., the information follows the patient. The ePR goes
beyond PACS because it does not only show information about the medical examinations
for the patients, but also any other related data such as clinical history, pain surveys,
biometric information and data acquired during surgical procedure.
The ePR is developed utilizing PHP (PHP: Hypertext Preprocessor) as the backend
programming language. The data values are stored using a MySQL database. The web
pages are structured with HTML (Hyper Text Markup Language), and they are styled
using CSS (cascading style sheet). The interfaces are dynamically updated using
JavaScript.
3.2.3.1 Data Storage and Archive and System Database
3.2.3.1.1 Input Data and Metadata
Managing the data acquired by the ePR system is a critical task. A dual-system back-up
mechanism is implemented as follows (see Figure 3.2). First, the ePR server has the
server hardware, and the Gateway has the Gateway hardware. Two identical server
software packages are implemented, one in the ePR server hardware as the primary and
the other in the Gateway hardware as the back-up. By the same token, two Gateway
software packages are implemented, the primary package is in the Gateway hardware,
and the secondary package is in the ePR Server hardware as the back-up. Refer to the
middle row of Figure 3.1, the Gateway and the ePR server each has its own hardware,
where each hardware piece is housing both the ePR server software and the Gateway
26
software; one is the back-up of the other. Figure 3.2 depicts the dual-system back-up
mechanism.
The input data first comes to the Gateway hardware, where the Gateway software
categorizes them by images, live waveform information, and textual information. Images
include DICOM images in their original DICOM format as well as in jpeg format for web
display; as well as endoscopic videos, endoscopic single frame images, and digital C-arm
fluoroscopic images. The metadata in the DICOM images and other data are stored in the
database disks. All acquired data and metadata are immediately backed up by the ePR
Server hardware.

Figure 3.2 - The Dual-system back-up Schema with two hardware pieces: ePR Server
hardware and Gateway hardware. Each hardware piece has two softwares: ePR Server software,
and Gateway software; and a tandem database with hard drive for data and metadata archive.

27
3.2.3.1.2 Database Schema
The database schema is another important component of the system; it contains all values
and relationships of the data objects that are stored. Figure 2.3 presented in previous
chapter shows the database schema and all relationships among data tables based on the
DICOM Data Model Standard. The database schema is supported by the filesystem
shown in Figure 3.1 where the images (DICOM, jpeg, gif, etc), videos and text files will
be stored. The database holds all metadata, data values and pointers to the location of the
objects stored in the filesystem as well. This database schema was first introduced for an
ePR for Radiation Therapy for decission support [10].
3.2.3.2 System Security
The system security has been considered carefully during the design in order to comply
with the HIPAA (Health Insurance Portability and Accountability Act) requirement. Only
the users who have been granted permission are allowed access to the system. At the
same time the privacy of the communications are kept to avoid any non-authorized
receiver to obtain a given patient’s private information. To guarantee the security of the
data, web access to the ePR is established with HTTPS that encrypts all communication
between the server and the clients (web browsers). In addition, the ePR system handles
permissions that will allow users to perform different tasks on the system. Different user
groups in the system have a different set of enabled permissions, however, permissions
can be overwritten for individual users by the system manager if necessary providing a
greater level of flexibility.

28
3.2.3.3 System Continuous Availability and Failover
The information that is kept in the ePR is unique and cannot be obtained from any other
sources if lost; even images that were acquired from the PACS become unique once they
were annotated in the ePR and can not be found anywhere else. To overcome any
possible lost of data, a Continous-Available solution that replicates the data of the ePR to
more than one place has been implemented. The primary Gateway serves as the backup
for the primary ePR server, and vice versa.
In addition to having the data being stored with more than one copy, system redundancy
with automatic failover mechanism has been designed to access the data in case of the
failure of any component in the system to guarantee system continuous availability.
3.2.4 Visualization and Display
The last of the four components in the ePR System is the Graphic User Interface (GUI)
and Display. In order to have the ePR system to be utilized as an effective tool that can
improve the workflow of the Surgery Department, it is important to have a user-friendly
GUI that presents all necessary contents for the surgery in an easy to use manner. For this
reason, the ePR system is designed with this concept to achieve this goal.
Because the entire ePR system operates in three interrelated stages during a surgical
procedure, from planning (Pre-Op), to surgery (Intra-Op), to patient recovery (Post-Op);
the interface design between these three stages are critical. The Display interface design
(see Figure 3.1) includes the main page, the Pre-Op display at the patient consultation
room, the Pre-Op display at the OR, the Intra-Op display at the OR, the Post-Op at the
29
patient recovery area, the Post-Op at the OR for surgical documentation, and the
administrative pages.
Figure 3.3 presents this general concept by displaying the main page of the ePR System
with three workflow steps listed. Pre-Op lists surgical patients scheduled to be
performed; Intra-Op details the next scheduled patient’s procedures; and Post-Op
presents procedures of the patient which had been recently performed. The details of
which will be described in Chapters 4, 5, and 6.

Figure 3.3 - The home page of the ePR system showing three surgical stages: Pre-Op, Intra-Op
and Post-Op.

30
Chapter 4
Pre-Op Authoring Module
The Pre-Op stage of a IA-MISS procedure is where all necessary information prior to the
surgery procedure is collected and organized in a patient e-folder. The Pre-Op happens
days prior to the surgery and involves querying, interviewing, collecting, and storing of
pre-surgical medical images, patient demographic information as well as other pertinent
data value that would assist the surgery during the procedure.
4.1 Workflow analysis
The Pre-Op authoring software module is composed of the following steps:
1) The patient goes to the surgery planning center and proper medical examinations are
performed. Usually they are X-Ray and MRI exams that are stored by the PACS.
These images are then reviewed by a medical specialist to determine the proper
treatment course.
2) Once the patient has been recommended for a IA-MISS procedure, the patient visits
the surgeon for a consultation session. At this point, the patient registers in the IA-
MISS ePR system to have his/her fingerprint recorded for biometric verification and
authentication. Also, the patient fills out all necessary forms for pain and clinical
history, which will later be entered into the ePR system.
3) Next, the IA-MISS clinical personnel create the surgical procedure in the ePR system.
The ePR interface allows them to obtain studies from PACS if those images have not
yet been downloaded onto the ePR server.

31
4) After the studies have been sent from the PACS to the ePR, the key images including
axial and sagittal MRI series images are required for the surgical procedure and are
selected and annotated by a physician assistant. The key images are labeled to
properly identify the surgical region of the body.
5) The Pre-Op authoring phase is complete once all the necessary image and text
information is entered into the ePR system, and can been viewed at the Pre-Op
authoring and display module at the consultation workstation.
4.2 Participants
The personnel at the IA-MISS clinical site involved in the Pre-Op authoring module
include the following:
1) Receptionists: They register the patient in the ePR system by entering the patient
demographics, such as name, date of birth, sex, gender, patient ID, and the accession
Number; as well as the pain survey data. They also take a fingerprint of the patient for
biometric registration (used later at the OR for patient identification). They are in
charge of scheduling the procedures and making sure there are no time conflicts.
2) Physician Assistants: They retrieve pre-surgical PACS images of the patient using a
query/retrieve function in the Pre-Op authoring and display module of the ePR
system. After the images of the patient have been transferred to the ePR system, they
will select the key images that will be used during the Intra-Op stage. At the display
module they create the surgical procedures by entering annotations which would be
later displayed with the images in the Pre-Op 52-inch LCD display monitor in the
OR.
32
3) The Surgeon: He/she approves the Pre-Op contents prior to display on the Pre-Op
display monitor in the OR. The surgeon works closely with the physician assistants to
determine which images should be included in the procedures.
4) Administrators: They are in charge of assigning user access rights to the ePR system.
4.3 Significance of Pre-Op Data Organization
4.3.1 Organization of the Pre-Op Data
Traditionally, surgeons have been relying on their memory for localization of where the
procedure should be performed. They review the MRI and X-Ray images the day before
surgery and studied the approach to be taken during the procedure. These images are also
brought to the OR for reference. But they are displayed in an unorganized fashion
scattered throughout the OR. The next few paragraphs focus on the organization of the
Pre-Op patient information which requires a preparation process described earlier in this
chapter. This process should not be done during the time of surgery and the information
should be saved in advance with the display streamlined and organized for efficiency
purposes.
4.3.2 Surgical Whiteboard Data
In addition to input data described earlier, one type of Pre-Op data which is critical
during surgery is the hand written whiteboard information located at the entrance of the
OR which contains a very short summary of the patient such as name, gender, age,
weight, height, any allergies, comorbidity and pain. The Pre-Op authoring module
described has been designed to integrate the whiteboard information onto the same Pre-
33
Op screen for display in the OR during the surgery. The following survey measures are
also included:
1. Visual Analog Scale (VAS): Is a psychometric response scale to describe the amount
of pain a patient is feeling from a specific part of her/his body.
2. Oswestry Disability Index: A survey to identify how the pain in the back or legs is
affecting the patient in his/her daily activities.
4.4 Graphical User interface
The design concept of the ePR system is user-friendly but effective at the same time. For
these reasons, the criterion of the user interface is to minimize the number of mouse
clicks needed to perform a certain task and to aggregate information adequately into a
single interface whenever possible. The current Pre-Op authoring module is a self-
contained interface where the users can download, edit, add, and delete the contents as
needed. The Pre-Op has two major interfaces, one for editing and one for display in the
OR.
4.4.1 Editing
4.4.1.1 To Create a Procedure
The interface allows the users to create the surgical procedures by first selecting the key
images as well as adding annotations to those key images as shown in Figure 4.1. On this
screen the PACS image and surgical procedures had been combined into one display.
Image studies related to the surgical procedure are shown on the left hand side based on
the surgical data model (see Figure 2.3). To view an image in a study, the users can either
drag the study shown on the list from the left to the viewing pane on the right hand side
34
or by double clicking the study from the list on the left. Figure 4.1 displays a sagittal MRI
image with patient’s ID above the image. The toolbar with icons at the top of the viewing
pane allows the users to perform certain tasks accordingly to the current status of the
editing module.

Figure 4.1 - The Pre-Op authoring module page. The left text list depicts the surgical data model
showing the studies and procedures. After the user clicks an item in the list, the proper image, i.e
a sagittal MRI would be shown on the right.
4.4.1.2 To Perform editing
1) To view images in a study: The two icons on both the right and the left sides allow the
user to preview images in the study series.
35
2) To perform image manipulation: The toolbar for the Pre-Op include some basic image
manipulation tools such as window/level, pan, and zoom. With this functionality, the
images can be displayed optimally at the exact location of the lesion.
4.4.2 The Neuro-Navigator Tool for Image Correlation
During a IA-MISS operation, it is important to correlate the Axial view with the
corresponding Sagittal view of an MRI study. The neuro-navigator tool allows such
correlation through the display as show in Figure. 4.2

Figure 4.2 - The Neuro-navigator tool allows the correlation of the position of the lesion in the
sagittal (left) and the axial view (right).

36
4.4.3 Display
A IA-MISS surgical procedure requires multimedia data during the Pre-Op stage
including patient history, images, and consultation results. These data should be
organized and displayed in the Pre-Op display during surgery. An example is shown in
Figure 4.3 which depicts the patient general information at the top row; below that row
the whiteboard information related to the surgical procedure is displayed. This layout
allows to maximize the viewing real estate of the screen to display the selected key
images and their annotations (center of the screen). A toolbar that allows to perform some
image manipulation (Window/Level, Pan, Zoom) is located on the left hand side of the
screen. The term that is used for this display is the Pre-Op Display since the Pre-Op
authored data is actually displayed during the Intra-Op workflow stage.

Figure 4.3 - The Pre-Op display organized during patient consultation as seen on the Pre-OP
display monitor in the OR during Intra-OP. Top Text Row: Patient General Information, Second
Text Row: Whiteboard information, Center: Images and annotation during Pre-Op consultation.
37
Chapter 5
Intra-Op Module
5.1 The Intra-Op Module
The Intra-Op stage is defined by the time during which the surgical procedure is being
performed by the surgeon in the OR. All information collected during the Pre-Op are
displayed in the Pre-Op display Monitor in OR. In addition live data from different input
devices during the surgery are collected by the Integration unit (IU) and displayed in the
Intra-Op monitor screen. Before the surgery starts, the patients’ fingerprint is captured in
the OR and verified with the biometric data acquired during the Pre-Op stage. This is an
important step towards reducing errors and insuring correct patient identification prior to
surgery.
5.2 Participants
In addition to their standard assigned surgical tasks, the surgical staff involved in the ePR
Intra-Op Module include:
1) Surgeons and medical staff: The chief surgeon operates on the patient, and makes
decisions based on the ePR Pre-Op and Intra-Op displays while the procedure is
taking place. Other surgeons and medical staff would alert the chief surgeon based on
alert messaging on the Intra-Op display. Thus, the ePR system facilitates their
multiple tasking by displaying any relevant information in an organized and efficient
easy-to-view way.
2) Anesthesiologist: He/she provides anesthesia to the patient and monitors the vital
signs during the procedure.
38

3) Nurses and Technicians: Each has assigned tasks to perform during the surgery. In
addition, one is in charge of activating the Pre-Op ePR module that has been authored
and displays it on the large Pre-Op Display monitor in the OR. One is assigned to
monitor the Intra-Op Live Display monitor. One is also assigned to verify the patient
by means of the fingerprint scanner and verification toolkit.

5.3 Data Acquired During Surgery
During surgery, while the surgeon is operating, all important data are automatically
acquired and saved by the IU (see Section 3.2.1). These data include:
1) BIS (bispectral index system): BIS signals comes from an array of electrodes that are
attached to the patient's forehead that monitor the patient consciousness. The value
that comes out is in the range of 0-100, where zero means no brain activity and 100 is
full consciousness. The safe range is considered to be between 40 and 70. Currently
the output data from this device is obtained from a serial port (RS232).
2) Endoscopic Images and Video: The IA-MISS procedure is performed using an
endoscope through a small incision made on the patient’s body. The video stream of
the camera inserted in the endoscope is displayed continuously on the Intra-Op Live
Display and is also saved at the IU for later review. At the same time, the IU takes
snapshots of the endoscopic video and saves them as images at the same interval as
the other data are acquired. The resolution of the video and images can vary from
vendor to vendor; currently the ePR system is acquiring the output in a VGA
39
resolution of 640x480 pixels. The output data from this device is obtained from an S-
Video port.
3) C-Arm fluoroscopic images: The C-Arm imaging device takes X-Rays images that
assist the surgeon to pinpoint the exact positions of the spinal column where the
lesion is located in real-time. These images are also encapsulated in a DICOM file
and sent to the PACS.
4) EMG: The Electromyography (EMG) device monitors the electrical activity of the
muscles and nerves by inserting a set of small needles into the body of the patient
with the goal to prevent any damage of the spinal cord while the surgeon is operating
with the endoscope. Currently the output data from this device is obtained from an
Ethernet connection.
5) Laser: The laser is inserted in the endoscope and is used to cauterize a wound inside
the body of the patient for both cutting and healing purposes, but its usage depends on
the type of treatment. The current available output from this device is from a serial
port (RS232)
6) Vital Signs: The data obtained from the vital signs device include the respiratory rate,
the heart beat rate, the temperature, the blood pressure, the pulseOX and the CO
2

respiratory emission levels of the patient (see Table 3.1 for other possible data).
Currently the data output from this device is obtained from a serial port (RS232).
5.4 Internal architecture of the Integration Unit (IU)
Figure 5.1 shows the interconnectivity between the current input devices present in the
OR during surgery. The left hand side shows the data inputs or sources, as explained in
40
the previous section; the middle is the IU, and the right hand side depicts the connectivity
between the IU and the Intra-Op Live Display monitor. The Pre-Op Display shown at
top-right also presents in the OR. The current setup allows the IU to support the available
input devices at the implementation site, however, the IU has been designed to be flexible
enough to support various vendors’ input devices using different interfaces (serial ports,
VGA, Ethernet, etc).

Figure 5.1 - The hardware and software interconnectivity diagram for a surgical procedure using
the ePR system. The IU in the middle accepts different input devices and sends them to the Intra-
Op Live Display.
As seen above in the figure, there is a interactive input device called the Surgeon Keypad
that permits surgeons to manage some functionality of the IU. With this keypad the
41
surgeon can take a screenshot of all the data and images any time during the surgery, and
it can also be used to have a side-by-side freeze frame display of the endoscopic video
with a prior endoscopic screenshot for comparison purposes.
5.5 Interaction with the gateway
During the surgery the IU saves all the data acquired from different connected devices in
its hard drive, after completion of the surgery, it connects and sends all data to the
Gateway server. The protocol used to send the data over is HTTPS. If the primary
gateway is down, the IU will try to connect to the secondary gateway which is the ePR
Server for failover. In order to reduce the total sending time, the data is previously
compressed and combined in a single zip file; once the data is correctly received by the
Gateway server it is then uncompressed and processed accordingly.
5.6 Graphical User Interface
Figure 5.2 shows a mock up example of the Intra-Op Live Display with waveforms and
images. The horizontal axis is time. There are eight groups of waveform in the top: six
vital signs with heart rate, blood pressure, respiratory rate, pulse oxygen concentration,
pCO2, and temperature, as well as BIS, IVF (Intravenous Fluid). Every dot in the
waveform represents a data point over a one second interval. There are two images, the
C-Arm fluoroscopic and the endoscopic video images, and one EMG waveform in the
middle row. In the lower row, there are the laser energy in joules. The video is updated
on the Intra-Op Live Display with a frame rate of 30 per second (a default value).

42
5.6.1 Rule-based alert mechanism
If a signal falls outside its safe range a 3 stage mechanism will alert personnel in the OR
about that situation.
1) Warning mode: If the numeral falls outside the safe range it will change its color to
red (as seen with the PulseOX and Blood Pressure in the figure)
2) Emergency mode: If the condition falls to a value greater or lower in 25% of the safe
range then the Intra-Op Live Display will place an alert message on top of the screen.
3) Critical mode: If the data signal value is either greater or lower in 50% to the values
in the safe range then the alert message will cover the whole screen.
Figure 5.2 - A mock-up example of the Intra-Op Live Display as seen on the Intra-Op large
monitor in OR. Top row: Waveforms of six vital signs, BIS, and IVF, the horizontal axis is time.
Middle row: Waveform of EMG, Fluoroscopic image, and endoscopic image. Bottom row: Laser
output values.
43
Chapter 6
Post-Op Module
6.1 Post-Op Module
The Post-Op stage takes place after the completion of the surgical procedure. There are
three substages: 1) Patient in the recovery area and then discharged, 2) the Surgeon
documents the surgical results, and 3) follow up pain surveys.
6.2 Participants
The personnel involved in the Post-Op module include the following:
1) Surgeons: The chief surgeon reviews the patient’s ePR record including Pre-Op,
Intra-Op, and Post-Op files and images, and then dictates the surgical report. He/she
also determines which images from the ePR should be included in the report. They
are ultimately responsible for the contents of the report.
2) Physician Assistants: They assist the surgeon while creating the surgical report.
3) Nurses and Front-desk personnel: They will perform follow-up surveys several times
after the surgery, and enter the pain surveys data into the ePR system as a follow-up
of the progress of the patient.
6.3 Patient in the Recovery Area
While the patient is in the recovery area, some fo the the patient’s vital signs (pulseOX,
Heart Rate and Blood Pressure) are still being recorded with the Vitals device and can be
displayed on the recovery monitor within the recovery area (see Figure 3.1). In addition,
an immediate pain survey before the patient is discharged will be conducted by the nurse
44
as a follow-up monitoring the progress and effectiveness of the surgery. These data is
entered to the ePR Post-Op module for statistical outcomes as the patient recovery record.
6.4 Post-Op Documentation
6.4.1 Graphical User Interface (GUI)
When the surgeon performs Post-Op documentation, he/she can retrieve information
from the Post-Op module pertinent to the surgery using the GUI. This process involves 4
major steps:
1) Finding the Patient from the ePR System. The correct patient can be found from the
ePR by clicking the first line of the bottom block section of GUI (Figure 3.3).
2) Selecting images. From this GUI, the surgeon can select endoscopic images that will
be included in the final report by clicking the star at the top left corner of the viewing
pane. As shown in Figure 6.1, that image has been selected for the final report.
3. Selecting Waveforms. The waveforms are displayed at the bottom of the interface.
They can be dynamically selected by clicking their corresponding boxes on the upper
right side of the interface.
4. Data Synchronization. A blue slider at the bottom of the graph would allow for
synchronized viewing of all the image and waveform data being displayed.
Figure 6.1 presents a documentation page that has been authored by the surgeon from the
ePR system using the Post-Op GUI. The data displayed and documented include an
endoscopic image, heart rate, diastole and systole blood pressures, respiratory rate, BIS
score value, oxygen pressure, partial pressure of carbon dioxide, and temperature. There
45
are also two curves shown in the bottom (Respiratory Rate and pulseOX) which are
waveforms obtained from another Intra-Op device.
During the documentation, the surgeon also dictates the report, which will be
synchronized with the authorized image and waveform paged.

Figure 6.1 - The Post-Op authoring module displaying a Post-Op document showing data
acquired during the surgery. This page can be synchronized with the surgeon Post-Op dictation.

6.5 Follow-up Pain Surveys
Nurses and Front-desk personnel perform surveys several times after the surgery and
enter the pain surveys data into the ePR system as a follow-up of the progress of the
patient. The collected information can be used for future patient outcome analysis.

46
Chapter 7
System Deployment
This chapter covers the system deployment and implementation of the ePR for Image-
Assisted Minimally Invasive Spinal Surgery (IA-MISS) at a clinical site. The clinical site
is the California Spine Institute (CSI) located in Thousand Oaks, California, which is the
only clinical site in Southern California that performs IA-MISS.
The first few sections will discuss the implementation schedule followed by pitfalls and
challenges faced during the delivery of the system and finally the training and support to
the users.
7.1 Implementation schedule
7.1.1 Planning and design phase
The ePR system for IA-MISS was developed at the IPILab, Department of Radiology,
USC, research facility with close collaboration with the deployment site mentioned
above. To test the functionality of the ePR system, a prototype for each of the ePR
components was developed and integrated; in addition, mockup data was used during this
phase that includes an Intra-Op signal simulator for the IU device. Once the system was
tested fully in the laboratory environment, it was then deployed in the clinical site to
obtain user feedback and clinical evaluation. The component pieces of the ePR system
include a combination of hardware and software in the clinical implementation. The
hardware components include the Gateway Server, the ePR server, the IU and the
fingerprint readers. The software pieces are the ePR web pages, the IU application, the
fingerprint application, the database, and the web server.
47
The final goal of the clinical implementation stage was to deploy the ePR system while
minimizing the risk of any possible disruption of live clinical services through
coordinating of tasks with clinical staff.
7.1.2. Hardware installation
The ePR and Gateway servers were installed at CSI on a rack at their Server Room.
Figure 7.1 shows the installation and the final location of those servers.

Figure 7.1 - Server installation and final location of servers at the Server Room at CSI.

In addition to the two servers above, the IU was also installed in one of the ORs at CSI.
The IU needs to be connected to peripheral devices that are present in the OR for
monitoring the real-time patients’ response during the clinical procedure. The IU and
connections to other devices are shown in Figure 7.2
Installing the
ePR server
ePR Server
Gateway Server
48
Integration
Unit
installed in
the OR
Vital signs being
connected
All input sources connected to
The Integration Unit
Figure 7.2 - Integration Unit installed at the OR with the different input sources connected as
well.
7.1.3. Software installation
Once the servers were installed at the clinical facility, the next step was to configure all
necessary software components of the ePR system. Those components include:
1) ePR server: A web application that provides the User Interface for the Pre- and Post-
Op authoring module and the Pre-Op display (during surgery) as well. In addition, this
server requires the installation of the web server (Apache [11]) and the database (MySQL
[12]).
2) Gateway server: A listener software that receives incoming DICOM studies sent from
the PACS and a set of scripts to extract the metadata information and store it at the
database. This server also receives the real-time surgical data including live images and
waveforms collected during a procedure, which is sent by the IU by means of an API.
3) Integration Unit: It acquires the surgical data in the form of data points, images and
videos from the surgical peripheral devices at the OR. A program developed in C++ is
utilized to make low level system calls to the different interfaces depending on
performance requirements to display real-time data in the OR.
49
4) Fingerprint module: Allows the registration and identification of patients at the
clinical facility. The registration is performed during the Pre-Op stage and the
identification and verification at the Intra-Op stage. This module makes use of a
fingerprint SDK (Software Development Kit) from CrossMatch [13].
7.1.4 Training
Training the users of the ePR system was a primary objective during deployment, Section
7.3 explains in greater detail.
7.2 Pitfalls and Challenges during delivery of the system
Lessons learned during the ePR system implementation at CSI are as follows:
1) Browser incompatibilities for visualization and display module: To access the Pre-
Op and Post-Op authoring modules users needed access to a web browser, which could
sometimes be different versions or applications that was used for developing. Thus, some
work was needed to to make sure the application runs in a similar fashion on all major
web browsers such as Internet Explorer versions 6 and 7, Mozilla Firefox versions 1.5
and 2 and 3.
2) Graphical User interface challenging for new users: When a new application was
developed, training becomes crucial since users were not familiar with the interface and
functionality.
3) Clinical environment was different than laboratory environment: The
development of the ePR system was initiatally in a laboratory environment where most of
the settings could be controlled and modified as needed where testing, debugging and
implementation of new requirements could be done quickly. However, the reality was
50
very different when it was implemented in the clinical environment. Clinical staff’s first
priority was to deliver the best service to patients, and needed to accomplish their daily
responsibility and activities timely, thus time became a precious resource to them that
needed to be used wisely and efficiently.
4) Clinical institutions were not always in control of their computer and ICT
(Information and Communication Technology) equipment: Installing new
applications in clinical computers might sometimes require administrative priviledges. In
addition, configuring and adding new servers to CSI might need to be performed by a
third party IT (Information Technology) team that could cause implementation issues.
5) Users acceptance: When a new application was implemented at CSI, users might be
reluctant to fully embrace the system since it would temporarily disrupt their normal
routine clinical workflow even if the system would ultimately improve their clinical
workflow in the long term.
6) Lack of standards for obtaining data from peripheral devices in the OR: This was
the major ostacle during the implementation phase. It had been a major challenge for
acquiring data from peripheral devices at the OR. Different vendors exported their data in
different ways, thus adding complexity to the mechanisms for data retrieval and limiting
the interoperability to certain vendors and products.
7) System refinements: Due to the fact that the real data for Intra-Op came from third
parties’ peripheral devices at the OR, it was quite challenging to retrieve real-time data
without disturbing an ongoing surgical procedure. Thus, special arrengements were made
to make sure the Intra-Op module was properly tested and debugged without
51
compromising the surgical outcome of IA-MISS procedures. In the same fashion,
deciding what changes from users’ feedback needed to be incorporate in the ePR system
as fast as possible was a decision that had to be made between the clinical team and the
technical team. The ePR could incorporate new requirements that would make the
application more useful, but there was also a need to freeze software releases to make
sure promised features would be put in place, this decision might disappoint some users
but was necessary. Finally, due to the nature of software development it was expected to
have some pieces of software code not functioning as expected. Thus, there should be a
clear mechanism to address the fixing of bugs for the application at due time.
7.3 Training and support to users
One key and important aspect in the adoption of a new technological solution is the
training and support given to users. During the preliminary inception of a new system
that will replace an existing one there will be some repetition on the daily activities that
will eventually and gradually will be eliminated. During the ePR implementation, I have
planned a very meticulous training program to the CSI personnel, an example during the
Pre-Op training is shown in Figure 7.3. The training sessions were aimed to specific users
who were heavily involved in the clinical workflow. The trainig sessions needed to work
around the daily clinical service to avoid the conflict with their duties.

52

Figure 7.3 - A training session with clinical staff at the California Spine Institute. General
workflow explanation and Pre-Op authoring module were introduced.

7.3.1. Users trained
The following users were among the ones being trained for the usage of the ePR system,
which are also considered the user types of the ePR system:
1) Surgeons: They received training on the Pre and Post-Op authoring modules (Post-
Op was demonstrated using mockup data) on how to use them by adding, editing and
deleting contents. In addition, they were taught to properly interpret the two large
LCD displays presented at the OR: the live Pre-Op Display and the Intra-Op Display.
(See Figure 4.3 for Pre-Op Display and Figure 5.2 for Intra-Op Live Display)
53
2) Physician Assistants: They were involved in the training sessions for Pre-Op
authoring module including Query/Retrieve functionality, image manipulation, key
image selection, procedure management, annotations for key images selected.
3) Nurses: They were trained to properly enter information related to the patient’s
whiteboard data (see Section 4.3.2)
4) Front-desk assistants: They received proper training for scheduling of surgical
procedures, input of pain surveys for Pre-Op and Post-Op stages as well as patient
registration which included registration of patients’ fingerprints.
5) Technicians: The training given to them included verifying the patients’ identity at
the OR using the fingerprint module. In addition, they were instructed on how to
correctly understand the data presented at the LCD monitor displays at the OR.
6) Administrative staff: They were given training on how to add or remove a user
from the ePR system and to manage the permissions for the different user types or a
specific user as well.
54
Chapter 8
Results
This chapter provides some preliminary analysis comparing the IA-MISS operation time
required before the ePR system was installed and IA-MISS operation time after the ePR
system was installed and used. This dissertation presented only the Intra-op time.
8.1 Statistical analysis of data collected before the installation of the ePR System - Only
Intra-Op times were recorded
Table 8.1 and Table 8.2 show some randomly chosen retrospective surgical patient data
collected before the implementation of the ePR System at CSI between 2007 - 2008 who
had the IA-MISS operation of two discs. Table 8.1 displays data from Cervical cases
while Table 8.2 displays data from Lumbar cases. Both tables refer to surgical cases
where 2 discs were operated on. The average time for Cervical cases was 75.25 minutes,
while the average time for Lumbar cases was 76.82 minutes.
Table 8.1 - Summary of Cervical cases with 2 discs operated before the installation of the ePR
System
Surgery Type Number of cases Average (minutes) Std. Deviation
Cervical 12 75.25 49.21

Table 8.2 - Summary of Lumbar cases with 2 discs operated before the installation of the ePR
System
Surgery Type Number of cases Average (minutes) Std. Deviation
Lumbar 29 76.82 39.31

55
The data displayed on Tables 8.1 and 8.2 was acquired from the manual notations taken
by the nurses during the surgical procedures. This method of data collection was a
manual task and could lead to errors from different sources including the definiton of
start and completion of the operation; when the nurses actually recorded the time, etc.
8.2 Preliminary analysis of data collected after the installation of the ePR System –
Only Intra-Op times were recorded
Among others, one goal of the ePR System was to automatically and objectively record
these important timestamps based on the ePR system master clock. Data collection after
the installation of the ePR system is defined by the fact that the surgical team relies 100%
on the Pre-Op display screen and the Intra-Op Live Display screen to assist the surgical
porcedure. Although the surgical team still monitored the individual imaging and clinical
devices, they used the two large LCD displays exclusively as their guide to perform the
surgical procedure. The data shown in Tables 3 and 4 was collected from the manual
notes that the nurses took during the surgery. The data had been collected from March -
April 2009.
Table 8.3 - List of 8 Cervical cases with multiple discs operated collected after the installation of
the ePR System.
Surgery Type Duration (minutes) Number of discs
Cervical 51 2
Cervical 77 2
Cervical 64 2
Cervical 42 2
Cervical 100 4
Cervical 32 2
Cervical 47 2
Cervical 28 1
The total number of cases was 8, the average time per case was 55.12 minutes and Std.
Dev. was 24.19.
56
Table 8.4 - List of 16 Lumbar cases with multiple discs operated collected after the installation of
the ePR System.
Surgery Type Duration (minutes) Number of discs
Lumbar 72 2
Lumbar 150 8 *
Lumbar 62 4
Lumbar 49 3
Lumbar 44 3
Lumbar 68 5
Lumbar 120 2
Lumbar 80 4
Lumbar 59 4
Lumbar 109 4
Lumbar 72 4
Lumbar 101 4
Lumbar 106 3
Lumbar 126 5
Lumbar 17 1
Lumbar 24 2
The total number of cases was 16, the Average time per case was 78.68 minutes and Std.
Dev. was 37.36.
* Denotes a bilateral case (procedure operating right and left sides)

8.3 Discussion
8.3.1 Data Collection
In order to compare the time spent during a IA-MISS operation, the start time and the
completion time of the operation were obtained based on the traditional method of having
the nurses take the notation during each operation. This method of data collection allows
us to directly compare the effectness of the ePR system to assist the surgical team during
the operation with less bias. Although no conclusive results have been revealed up to this
point because of limited data collected, the ePR system does pave the way for a very
57
rigorous data collection methodology with the system master clock and real-time image
and data archive to automatically record timestamps and events.

8.3.2 Preliminary Observation
Observation 1, Cervical cases: Even though only eight cervical cases with multiple
discs operations were collected after the installation of the ePR (Table 8.3), the average
time spent per case for the Intra-Op operation, 55.12 minutes, was signifcantly less than
75.25 minutes, which is the operation time spent without the ePR system (Table 8.1).
Although we can not conclusively say that the ePR system improves efficiency of the IA-
MISS operation for such a small data collection, the preliminary signs are certainly
positive.
Observation 2, Lumbar cases. We had collected sixteen lumbar cases after the
installation of the ePR. Unfortunately many of them required more than two discs for
operation and only 1 out of the 16 cases had less than two disc. If we ignore the number
of discs operated, the average time required to perform a IA-MISS operation with the
ePR system was 78.68 minutes (Table 8.4), only 2 minutes longer than that of two disc
only lumbar cases, 76.82, (Table 8.2) collected prior to the installation of the ePR system.
In general, without the ePR system, the data collection reveals that it takes about 30
minutes per disc. Based on this observation, it looks promising that the ePR may also be
able to improve the efficiency of the IA-MISS lumbar operation.

58
Observation 3. Other data collection benefits observed of the ePR System.
1) It is observed that the recorded Intra-Op time has very long duration times for a small
group of procedures, for example, in Lumbar 2, 7, 10, 12, 13 and 14. Currently, the
ePR System is being validated against those times to determine if it can accurately
and precisely detect the starting and ending points of a procedure which would give a
better representation of the average IA-MISS operation times without using the
manual nurse-taking-notation method. For those cases which require much longer
time to perform, the ePR system could allow the surgical team to trace the recorded
data from the ePR database and document the possible causes for the extra time
required and ultimately help to streamline the surgical workflow. The data collection
is ongoing with respect to this study. Currently the ePR System is being validated to
assure that the times automatically captured reflect the real start and end times of the
procedure.
2) To pinpoint the exact start and completion of a IA-MISS operation, we can start to
document the exact time required to perform a one disc operation as follows: We first
define the start time of an operation of a disc as the time the ePR records the first
fluoroscopic image which the surgeon needs in order to insert the cannula as close to
the location of the herniated disc as possible. The completion time of the operation
can be defined as the time when the ePR system detects the loss of multiple vital sign
signals of the patient signalling that the surgical team is removing the connection of
these devices from the patient. These times have been well documented in the ePR
59
database based on the system master clock, which can be traced back to related events
from the database automatically after the operation.
60
Chapter 9
Current Status and Future Plans
9.1 Current stage of the deployment of the ePR for IA-MISS at the California Spine
Institute (CSI)
The ePR system was deployed at CSI for clinical evaluation in November, 2008, and as
of April 2009, the ePR system has been used daily. Currently, the Pre-Op authoring
module and the Intra-Op module are being used for routine daily surgical operation. The
Post-Op module is under clinical evaluation and user training.
Pre-Op Stage: For the Pre-Op stage the users can register new patients, create new
surgical procedures, add key images to procedures with corresponding annotations, add
patients survey forms into the ePR, include whiteboard data and register the patients’
fingerprints for verification. All these data are displayed on the large Pre-Op LCD
monitor during live surgical procedure and archived.
Intra-Op Stage: The Intra-Op module is also in daily clinical use. Due to the fact that
the data requires accuracy and timely display, the IU is undergoing some performance
enhancements to handle various input sources adequately and efficiently. The
performance enhancements are being developed in the background where the
programming has no burden on the daily operations of the system. The feedback loop for
enhancements and refinements are carried out with the collaboration of the clinical staff
in order to assure their confidence with the ePR system. Certain refinements related to
interconnectivity of some peripheral surgical devices are also being performed iteratively
with clinical feedback.
61
Post-Op Stage: This module is currently under clinical evaluation and user training. The
data collected from the Pre-Op and Intra-Op stage provide the basis to create a clinical
report for the surgery. The report is created by the surgeon after the surgery and has been
completed. The chief surgeon at the clinical site has already performed a clinical case
with all the data collected during all the stages and some refinements to the final display
of the data is currently undergoing.
9.2 Research and Development After the Prototype
9.2.1 Integration Unit Prototype: Version 2
The Integration Unit (IU) is a very critical and complex component in the ePR system
which needs refinement to satisfy the original ePR System requirements (see Section
3.2.1). The prototype Version 2 of the IU has been designed to provide system
continuous availability; and to improve the performance for data collection, visualization
and storage with expanded flexibility in order to support a larger set of future devices and
system configurations.
9.2.2 Difference between the current IU prototype and Version 2:
The main difference between the two versions of the Integration Unit is the capability of
handling system continuous availability as the first prototype has no system continuous
availability built in. If any component of the system goes down, the complete system
goes down. IU Version 2 provides continuous availability by adding a second set of key
components in the IU. The principles for continuous availability design follow the
methodology developed for PACS in Radiology. Among these components added are:
1) A more robust UPS (Uninterruptible Power Supply);
62
2) Using NAS (Network Attached Storage) for archive;
3) The main processing unit is provided by means of two identical clustered blade
servers. In case of any failover the recovery is performed automatically by the server;
4) It is assembled in a self-contained mobile cart that enhances the flexibility for
implementation in different environments;
5) The number of I/O ports has been increased to accommodate more peripheral devices
in the Operating Room;
6) In software, the acquisition, display, and archive software have been upgraded to
handle new future system infrastructure.
Upon completion of Version 2, one copy will replace the existing first prototype within
CSI, and the second copy will be deployed at a second clinical site to be determined.
9.3 IRB for patient outcome analysis
Two IRB (Institutional Review Board) applications had been submitted and approved in
order to receive permission to conduct research studies on human subjects while
respecting their rights and welfare. With the approval of these two IRB applications, we
will adequately and lawfully allow the proper patient outcomes analysis of the data
collected by using the ePR system. Among others, the major outcome analyses are:
1) The benefits of using the ePR ssytem for IA-MISS operation in terms of system
operation.
2) The effectiveness and efficiency of using the ePR system to record time stamps on
every process during the IA-MISS operation.
63
3) Record the patient pain level before and after the IA-MISS operation for several years
as a parameter for measuring patient outcomes after the IA-MISS operation.
The IRBs are tied to the clinical institutions where the research has been and will be
performed, in this case, they are the California Spine Institute (CSI) for system research
and development, and clincial operation; and USC (University of Southern California) for
statistical analysis. The two IRB are:
1) “OR SurgMatix, An Electronic Patient Record (ePR) System for minimally invasive
spinal surgery (MISS)” WIRB (Western Institutional Review Board), submitted
April, 2008, approved August 26, 2008. [14]
2) “A minimally Invasive Spinal Surgery (MISS) Cinical Database for Patient Outcomes
Analysis” USC IRB, submitted August 29, 2008, approved October 23, 2008. [15]
With these two approved IRB’s listed above, the outcome analysis path becomes clear
from the institutional and regulatory requirements, and thus future patient outcomes
statistical analysis of the ePR system can be implemented.
64
References
[11] Apache web server, Apache Software foundation http://httpd.apache.org
[3] Chiu J. Endoscopic Lumbar Foraminoplasty In: Kim D, Fessler R, Regan J, eds.
Endoscopic Spine Surgery and Instrumentation. New York: Thieme Medical Publisher;
2004: Chapter 19, pp. 212-229.
[4] Chiu, J., Savitz, MH. Use of Laser in Minimally Invasive Spinal Surgery and Pain
Management. In: Kambin P, ed. Arthroscopic and Endoscopic Spinal Surgery – Text and
Atlas. Second Edition. New Jersey: Humana Press; 2005: Chapter 13, pp. 259-269.
[5] Chiu J., Anterior Endoscopic Cervical Microdiscectomy. In: Kim D, Fessler R, Regan
J, eds. Endoscopic Spine Surgery and Instrumentation. New York: Thieme Medical
Publisher; 2004:5 pp. 48-58.
[6] Chiu J., Clifford T, Greenspan M. Percutaneous microdecompressive endoscopic
cervical discectomy with laser thermodiskoplasty. Mt Sinai J. of Med 2000; 67:278-282.
[9] Chiu J., The Decade of Evolving Minimally Invasive Spinal Surgery (MISS) and
Technological Considerations. The Internet Journal of Minimally Invasive Spinal
Technology. 2008. Volume 2 Number 3
[13] CrossMatch Technologies Inc. http://www.crossmatch.com
[7] Huang H.K., 2001. PACS, Informatics, and the Neurosurgery Command Module. J.
Mini Invasive Spinal Technique. Vo1 1, 62-67.
[8] Huang H.K., March, 2004. PACS and Imaging Informatics: Principles and
Applications. John Wiley & Sons, Hoboken, New Jersey. 704 pages.
[2] Jensen M, Brant-Zawadzki M, Obuchowski N, et al. Magnetic Resonance Imaging of
the Lumbar Spine in People Without Back Pain. N Engl J Med 1994; 331: 69-116
[10] Liu B.J., Law Y.Y., Documet J., Gertych A., Image-Assisted Knowledge Discovery
and Decision Support in Radiation Therapy Planning, Computerized Medical Imaging
and Graphics, 31:4-5, pp. 311-321, 2007.
[12] MySQL AB, Sun Solaris Microsystems Inc. http://dev.mysql.org
[15] “A minimally Invasive Spinal Surgery (MISS) Cinical Database for Patient
Outcomes Analysis” USC IRB, submitted August 29, 2008, approved October 23, 2008
[1] Vallfors B. Acute, Subacute and Chronic Low Back Pain: Clinical Symptoms,
Absenteeism and Working Environment. Scan J Rehab Med Suppl 1985; 11: 1-98
65
[14] “OR SurgMatix, An Electronic Patient Record (ePR) System for minimally invasive
spinal surgery (MISS)” WIRB (Western Institutional Review Board), submitted April,
2008, approved August 26, 2008.

Abstract (if available)

Abstract Recent developments in medical imaging informatics have improved clinical workflow in Radiology enterprise. However, there still remains gaps in the clinical continuum from diagnosis to surgical treatment through post-operative follow-up that can be addressed by a variety of advanced technologies. One solution is the development of an electronic patient record (ePR) that integrates key imaging and informatics data during the pre, intra, and post-operative phases of clinical workflow. One application is in image-guided minimally invasive spinal surgery (IA-MISS) where spinal discectomy procedures are performed for decompressing nerve roots affected by spinal disc protrusions. This procedure utilizes a variety of still and real-time acquisition systems including X-Ray, CT, MRI, digital fluoroscopy and digital endoscopic video. The integration of these data together with waveform and other related informatics data is necessary during the entire surgical procedure for evaluation, treatment planning, and review.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Mining an ePR system using a treatment plan navigator for radiation toxicity to evaluate proton therapy treatment protocol for prostate cancer

PDF

A medical imaging informatics based human performance analytics system

PDF

Knowledge‐driven decision support for assessing radiation therapy dose constraints

PDF

Imaging informatics-based electronic patient record and analysis system for multiple sclerosis research, treatment, and disease tracking

PDF

Development of an integrated biomechanics informatics system (IBIS) with knowledge discovery and decision support tools based on imaging informatics methodology

PDF

Molecular imaging data grid (MIDG) for multi-site small animal imaging research based on OGSA and IHE XDS-i

Asset Metadata

Creator Documet, Jorge (author)

Core Title An electronic patient record (ePR) system for image-assisted minimally invasive spinal surgery

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Biomedical Engineering

Publication Date 08/01/2009

Defense Date 04/30/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag electronic patient record,intra-operative,minimally invasive spinal surgery,OAI-PMH Harvest,post-operative,pre-operative,system integration

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Huang, Han K. (committee chair), Khoo, Michael (committee member), Liu, Brent (committee member), McNitt-Gray, Jill (committee member)

Creator Email documet.ipilab@gmail.com,documet@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2445

Unique identifier UC1451954

Identifier etd-Documet-3092 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-172222 (legacy record id),usctheses-m2445 (legacy record id)

Legacy Identifier etd-Documet-3092.pdf

Dmrecord 172222

Document Type Dissertation

Rights Documet, Jorge

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu