Transcript Cyber-physical modeling of implantable cardiac medical devices

CYBER-PHYSICAL MODELING
OF IMPLANTABLE CARDIAC
MEDICAL DEVICES
SOL YOON
ICE, DGIST
FEB. 6 TH . 2012
OUTLINE

Introduction

Overview of model-based design

Background knowledge

Integrated Heart Model

Heart Model Validation

Pacemaker Model

Closed-loop case study

Conclusion
INTRODUCTION
THE FDA AND MEDICAL
DEVICE SOFTWARE
 FDA: need for rigorous real-time methodologies to validate
and verify medical device software

The use of artificial implantable heart rhythm devices has grown
rapidly over the recent decades
 However, there is no formal methodology or platform to validate and
verify the correct operation of medical device software
 Software is reviewed by the FDA only in the incident of a device
recall.
 Implantable medical devices are a primary example of medical
cyber-physical systems

Safety and efficacy of the device and device software must be
evaluated within a closed-loop context of the patient
CURRENT
TESTING,
VALIDATION, AND VERIFICATION
 The primary approach is unit testing using a playback of
prerecorded electrogram and electrocardiogram signals
 April 2010, the FDA began the “Infusion Pump Improvement
Initiative”
 An effective verification methodology is needed for the risk
analysis and certification of medical device software
 Pacemaker mediated tachycardia (PMT)
 A condition where the pacemaker inappropriately drives the
heart rate toward the upper rate limit
 Can be used for closed-loop system analysis
METHODOLOGY FOR CLOSEDLOOP MEDICAL DEVICE SAFETY
 Developed an integrated functional and formal virtual heart
model (VHM)


Clinically relevant
Timed automata based
 Developed a pacemaker device model for interactive and
clinically relevant test generation
 A set of general and patient condition-specific pacemaker
software requirement to ensure the safety
 Provide a means to test and verify the closed-loop system

The atrial-ventricle synchrony must be enforced
OVERVIEW OF MODEL-BASED
DESIGN
PREVIOUS HEART
MODELING EFFORTS
 The model of the heart should capture the electrophysiological
(EP) properties of the heart and generate functional signals


Conduction
Timing signal
 The heart models have been developed to study the heart
functions from the electrical and mechanical aspects.


Signal propagation, distortion, and attenuation
Cardiac output and valve mechanisms
REQUIREMENTS FOR MODELBASED CLOSED-LOOP V&V
1. Model fidelity: must cover the functioning heart

Normal sinus rhythm, sinus bradycardia, pacemaker mediated
tachycardia, etc.
2. Simplicity



The heart model currently have hundreds of differential equations or
millions of finite elements
Simulation of the models are time consuming and do not interact with
medical devices
The VHM presents an abstraction of the timing and electrical
conduction
3. Physical testbed

Enable to operate the heart on VHDL-based FPGA platform for blackbox closed-loop testing
OVERVIEW OF THE
VHM APPROACH
 Platform provide two interface


A formal signal for medical device software
A functional electrogram for real device implementation
BACKGROUND KNOWLEDGE
BACKGROUND
KNOWLEDGE
 The heart generates electrical impulses which organize the
sequence of muscle contractions during each heart beat
 The heart’s electrical timing is fundamental to proper cardiac
function
 The implantable cardiac pacemaker is a rhythm management
device
 Such devices have improved the condition of patients with
cardiac arrhythmias
CELLULAR-LEVEL
ACTION POTENTIAL
 The heart tissue can be activated by an external voltage applied
to the cell
ELECTRICAL
CONDUCTION SYSTEM

The tissue at the sinoatrial node (SA) periodically and
spontaneously self-depolarizes


The activation signal travels through both atria, causing
contraction and pushes blood into the ventricles.
Then the activation is delayed at the atrioventricular (AV) node
which allows the ventricles to fill fully
CARDIAC ARRHYTHMIAS
 There are anomalies of the conduction and refractory
properties in heart tissue


Bradycardia: failure of impulse generation with anomalies in the
SA node and failure of impulse propagation
Tachycardia: impair hemodynamics caused by anomalies in SA
node or reentry circuit
ARRHYTHMIA DIAGNOSIS
AND TREATMENT
 electrophysiology (EP) testing

Catheters with multiple electrodes on the tip are inserted from the
groin into the heart
 Can locate timing anomalies, using the spatial information from
catheter placement and the temporal information from the timing
difference between the pulses
 Ablation surgery can treat reentry circuit
 Electrocardiography (ECG)
RHYTHM MANAGEMENT
DEVICES
 Implantable pacemakers have been developed to deliver
timely electrical pulses to the heart to treat bradycardia
 The pacemaker has two leads inserted into the heart


One in the right atrium
One in the right ventricle
 By doing timing analysis of the electrogram (EGM) signals
sensed from the two leads


Artificial pacemaker generates electrical pulses when necessary
that can maintain ventricular rate
Enforce atrial-ventricular synchronization
HEART MODEL
A BRIEF OVERVIEW OF
EXTENDED TIMED AUTOMATA
 VHM uses a timed-automata semantics, which is similar to
the semantic extension used in UPPAAL
 The electrical conduction system consists of conduction
pathways with different conduction delays and refractory
period
 The refractory and conduction properties are all timing
based, it is natural to model the electrical conduction
system as a network of timed automata
MODELING THE ELECTRICAL
CONDUCTION SYSTEM
(a) Node automaton that models the refractory properties of heart tissue
𝑡

𝑓 𝑡 = 1 − 𝑇𝑟𝑟𝑝 , 𝑡 ≥ 1, 𝑡 ≥ 𝑇

𝑇𝑚𝑖𝑛 + (1 − (1 − 𝑥)3 ) ∙ 𝑇𝑚𝑎𝑥 − 𝑇𝑚𝑖𝑛 , 𝑖 = 𝐴𝑉
𝑔 𝑥 =
𝑇𝑚𝑖𝑛 + (1 − 𝑥)3 ∙ 𝑇𝑚𝑎𝑥 − 𝑇𝑚𝑖𝑛 , 𝑖 ≠ 𝐴𝑉

𝑇𝑚𝑖𝑛 and 𝑇𝑚𝑎𝑥 are the minimum and maximum values for 𝑇𝑒𝑟𝑝 of the
tissue
MODELING THE ELECTRICAL
CONDUCTION SYSTEM
(b) Path automaton that models the propagation properties of heart tissue
𝑝𝑎𝑡ℎ_𝑙𝑒𝑛/𝑣 ∙ 1 + 3𝑐 , 𝑖 = 𝐴𝑉
𝑝𝑎𝑡ℎ_𝑙𝑒𝑛/𝑣 ∙ 1 + 3𝑐 2 , 𝑖 ≠ 𝐴𝑉

ℎ 𝑐 =

𝑝𝑎𝑡ℎ_𝑙𝑒𝑛 denote the length of the path and vis the conduction velocity
HEART MODEL VALIDATION
ELECTROPHYSIOLOGY
STUDY
1. Catheter Placement



The typical catheter positions used are high right atrium (HRA)
His bundle electrogram (HBE), which is placed across the valve
between atrium and ventricle
Right ventricle apex (RVA), which is placed at the right ventricle apex
to monitor electrical activity of the ventricle
2. Extrastimuli Technique




HRA catheter deliver external pacing signals faster than the intrinsic
heart rate
The interval between two consecutive pacing signals is referred to as
basic cycle length (BCL)
The interval between the extrastimulus and the last pacing signal of
the pacing sequence is referred to as coupling interval
By decreasing the coupling interval gradually, the extrastimulus will
reach the RRP of the tissue, causing changes in conduction delays
CLINICAL CASE STUDY
 Key interval values when the coupling interval shortens for a
real patient



𝐴2 , 𝐻2 , 𝑉2 are the pulse caused by the extrastimulus
The interval 𝐴1 − 𝐴2 is equal to the coupling interval
𝐻1 − 𝐻2 , 𝑉1 − 𝑉2 indicate conduction delay between the His bundle and
the ventricle
Caused by extrastimulus
Coupling interval
VHM SIMULATION
 VHM is able to generate similar result with extrastimuli
technique
PACEMAKER MODEL
PACEMAKER MODEL
 The artificial pacemaker is designed for patients with
bradycardia




Two leads, one in the right atrium and one in the right ventricle, are
inserted into the heart
Two leads monitor the local activation of the atria and the ventricles,
and generate corresponding sensed event (AS, VS) to its software
The software determines the heart condition by measuring time
difference between events and delivers pacing events (AP, VP) to
analog circuit
Analog circuit delivers pacing signals to the heart to maintain heart
rate and A-V synchrony
DDD PACEMAKER TIMING
DIAGRAM
 Five basic timing cycles which diagnose heart condition




Ventricle pace (LRI), ventricle sense (AVI)
Atrial pace (ARP), atrial sense (VRP)
Coordinator between the atrium and ventricle leads (URI)
Each task was assigned a period of 10 ms
CLOSED-LOOP CASE STUDY
ENDLESS LOOP
TACHYCARDIA (ELT)
 The ELT is induced by premature ventricular contraction
(PVC), which is due to abnormal self-depolarization of
ventricular tissue
CONCLUSION
PHYSICAL
IMPLEMENTATION
 Can validate the closed-loop electrical interaction between the
heart (FPGA) and pacemaker (FireFly node)
CONCLUSION AND
FUTURE WORK
 A primary challenge in life-critical real-time systems is with the
design of bug-free medical device software
 Using timed automata
 designed an integrated functional and formal model of the heart
and pacemaker device
 A real-time VHM has been developed to model the
electrophysiological operation of the human heart