Center for Biomedical Signal Analysis and Integrated Diagnostic SAID

1
Center for Biomedical Signal Analysis and
Integrated Diagnostic (SAID) Ranga
Komanduri (OSU-Stillwater) and Bruce Benjamin
(OSU-Tulsa CHS)
2
Cardiovascular Healthcare Needs and Economic
Significance
? Currently, 38 of American women are over 45
years of age ? 21 of them ( 24 million
people) have cardiovascular disease ? U.S.,
Europe, developed Asia and other wealthier
nations 100 million people ? Capture 1 of this
market at the end of year 5 ? 1 million analyses
per year ? At 10 per analysis, 10 million/year
revenue stream and increasing from year 5
3
Interdisciplinary Research Team
OSU
-
CHS

OSU
-
Stil
lwater

Bruce A. Benjamin, Ph.D.

Ranga Komanduri,Ph.D.

Associate Professor

Regents Professor

Chairman, Department of Pharmacology
A.H. Nelson Jr. Endowed Chair in
and P
hysiology

Engineering

Warren
E.
Finn, Ph.D.

Department of Mechanical and Ae
ro
space
Associate Professor,

Department of Pharmacology and
Engineering

Physiology

Lionel M. Raff, Ph.D.

Randy
S.
Wymore, Ph.D.

Regents Professor

Assistant Professor,

Department of Chemistry

Department of Pharmacology and
Satish T.S. Bukkapatnam, Ph.D.

Physiology

Associate Profe
ssor

Randal Reust, D.O.

School of Industrial Engineering and
Assistant Professor

Management

Internal Medicine

Martin T. Hagan, Ph.D.


Professor

OU
-
Tulsa

School of Electrical and Computer
Charl
es
E.
Henley, DO
, MPH, FAAFP

Engineering

Professor and Vice Chair

Mark G. Rockley, Ph.D.

Founders and Associates Research Chair
Professor

in Family Medicine

Department of Chemistry

Doug Ivans, MD

Sundar Madihally

Assistant Professor

A
ssistant

Professor

Department of Family Medicine

Department of Chemi
cal Engineering

Francis Wen, Ph.D.

Robin B. Purdie, R.N., M.S.

Assistant Professor

Director

Department of Family Medicine

Wellness Center


Sa
m L. Earnest, M.D.

Cardiologist, Wellness Center


4
Typical EKG Signals
5
Signal Analysis and Integrated Diagnostics

• ECG and other cardiovascular signals contain a
  wealth of information for medical diagnostics
• - Only about 20 of signal information is
  used in clinical diagnostic applications
• - Diagnostic accuracies heavily depend on
  the experiential knowledge of the individual
  cardiologist
• Significant improvements in diagnosis and
  analysis of cardiovascular conditions possible
  from effective use of information buried in ECG
  and other signals
• - Provides more consistent basis for
  capturing and interpreting current
  experiential knowledge
• - Significant information crucial for
  medical diagnostics can be harnessed (e.g.,
  early detection of Alzheimer's disease)
• Complexity of ECG patterns as well as
  relationships connecting the signals with the
  relevant events pose significant challenges

6
What Will It Take?
? Multiple Disciplines - Researchers from
various branches of Science, Engineering, and
Medicine ? Integration of Multiple Sensor
Signals - e.g., ECG, Pulse, Blood
Pressure, to improve diagnosis ? Multiple
Analysis Methods/Tools - e.g., FFT, Wavelet
Analysis, Principal Component Analysis (PCA),
Independent Component Analysis (ICA),
Neural Networks (NN), Adaptive
Neuro-Fuzzy Inference System (ANFIS) ? Multiple
Disorder Diagnosis Capability - e.g., Heart
Failure, Arrhythmia, Orthostatic Hypotension,
Hypertension, Titration Therapy, myocardial
Infraction, Forensics, Homeland
Security, for comprehensive use of sophisticated
analytical tools
7
Multiple Diagnostic and Clinical Applications
? Hypertension ? Heart failure ? Myocardial
infarction ? Diabetes Type 1 and 2 ? Renal
failure ? Fetal distress ? Forensics ? Sleep
Apnia
? Pregnancy -and pre-eclampsia ? Heart
transplantation ? Arrhythmia ? Preventative
medicine ? Orthostatic hypotension ? Titration
therapy for drugs ? Homeland Security ? Biometric
Identification
8
Multiple Analytical Tools and Approaches

• Signal Representation
• Frequency (FFT) Analysis
• Wavelet Analysis
• Nonlinear Time-Series Analysis
• Feature Extraction
• Statistical Features
• Classical ECG Features
• QRST Cancellation
• Principal Component Analysis (PCA)
• Independent Component Analysis (ICA)
• Classification
• Statistical Characterization
• Neural Networks (NN)
• Adaptive Neuro-Fuzzy Inference Systems (ANFIS)
• Decision Trees and Nonlinear Regression

9
ECG Signal Analysis Methodology
10
Economic Plans and Market
Who will use this? ? Emergency rooms ? Cardiac
patients ? Wellness check ups ? Sports
Facilities ? Telemedicine ? Home monitoring
What is the market size? ? 1 million EKGs in
use ? 365 million analyses per year (at just
1/day) ? 1 of U.S. market share 1 million/yr
(est.)
11
Healthcare Benefits
? Substantial advancements in early diagnosis of
certain dangerous cardiovascular conditions ? New
paradigms in advanced remote and rural
healthcare ? Significant advancements to
telemedicine and home monitoring
National Recognition
? The center servers as a national role model for
creating a new technology-base through
synergistic collaboration among engineers,
scientists and medical practitioners -
Each participant brings a unique expertise and
experience in the respective fields of in
engineering (e.g., signals, materials), science
(e.g., chemical analysis, physiology), and
medical practice (e.g., cardiology,
telemedicine) ? Uniqueness of the collaboration
emerges from the use of multiple alternative
suites of analytical tools and techniques
12
Economic Benefits
? The Signal Analysis and Integrated Diagnostics
(SAID) Center will lead the emergence of
Tulsa region as a national hub for advanced
cardiovascular diagnostics ? Job creation in
Oklahoma, especially in the Tulsa region ? New
investments and ventures commercializing and
supporting the technologies ? Education and
training of Oklahomas best talent to pursue
lucrative careers in emerging healthcare
and biomedical industrial sectors ? Attractive to
NIH and DAPRA for providing additional
funding/grants
Jobs for Oklahoma
? High tech. health research careers ? Software
development ? Computer system development and
operations ? Computer security ? Finance and
business management ? Marketing ? Signal
processing research ? Spin-off technology company
developments