The Computer at the Bedside: Monitoring and Diagnostic Applications

1
The Computer at the Bedside Monitoring and
Diagnostic Applications

• Craig L. Scanlan, EdD, RRT, FAARC
• Department of Interdisciplinary Studies
• UMDNJ-SHRP

2
The Need

• To measure electrical, mechanical and/or
  chemical biosignals
• To record, store, and retrieve these data
• To convert the data into useful information
• To apply this information to support decision
  making
• To control equipment/automate processes
• To improve the safety, quality and efficiencyof
  patient care

3
(No Transcript)
4
The Basic Process
5
Transducers
A transducer converts a biosignal into an
electrical signal (voltage or current)
6
Transducers
Example Strain Gauge Pressure Transducer

Converts pressure energy (e.g., blood pressure)
into electrical signal
7
Analog to Digital Conversion
Transforms an analog signal into a series of
numbers (a digital signal) that can be processed
by a computer by taking samples with a certain
sampling frequency
8
Importance of Sampling Frequency
9
Importance of Signal Quality
10
Categories of Signal Analysis

• Output signals only, e.g., EEG
• Evoked signals (stimulus ? response), e.g, nerve
  conduction velocity
• Provocative tests, e.g., exercise ECG, bronchial
  provocation testing
• Modeling/controlling, e.g., closed-loop control
  of mechanical ventilation

11
Uses for Biosignal Analysis

• Basic research - physiologic measurement
• Function analysis - ECG, EEG, spirometry
• Population screening - broad app above
• On-line analysis - continuous monitoring
• Device control - pacemakers, AEDs, ventilators

12
Example Lung Volumes/Flows
Flow Transducer
Flow-Volume Loop
13
Example Blood Pressure Measurement
14
Example Breathing Movements
15
Example Polysomnography
16
Polysomnography - Automated Measurements
17
Polysomnography Automated Diagnosis
18
Using Biosignals to Control Processes
19
Servocontrolled Humidifer (Simple)
20
Automatic Defibrillator (Moderately Complex)
21
Adaptive Ventilator (Complex)
22
Information Integration
23
Decision Support Systems

• Quantitative Decision Support
• based on pre-established data sets and
  well-defined statistical methods
• Qualitative Decision Support
• use symbolic reasoning methods such as "logical
  deduction" which may be best understood in terms
  of Boolean logic or symbolic logic

24
Quantitative Diagnostic Model Input Example
25
Quantitative Model Decision (Output) Example
26
Quantitative Decision SupportOnline Example
DxPlain (UMDNJ Libraries)
27
Qualitative Decision Support

• Examples
• Decision Tables (Truth Tables)
• Flowcharts/Algorithms

28
Expert System (Rule-Based)
29
Expressions Used in the Truth Tablefor
Arrhythmia Diagnosis
30
Truth Table ExampleArrhythmia Diagnosis
31
Truth Table Example Hemodynamic Evaluation
32
Qualitative ModelFlowchart(Partial Open Loop
System)
SIMV2/min
Decrease SIMV by 2/min
Yes
Yes
Rate gt 8 lt 25/min?
Decrease PSV by 4 cm H2O every 5 min
Yes
VE gt 6 lt 14 L/min?
Patient Stable?
Yes
Yes
Yes
SpO2 gt 90?
Extubate at PSV4
Yes