1 Virtual Physiological Humans From Medical Images to Personalized Cardiac Models Nicholas Ayache http://www-sop.inria.fr/Asclepios/ Bio-ICT: The Heart in the Computer 2 April 2009 Credits Asclepios Team past/current members Academic, Clinical and Industrial Partners
25
Embed
Credits - ETSI · ARTreat CV/Atheroschlerosis IP preDiCT Heart/CV disease STREP ContraCancrum Cancer STREP ARCH Vascular/AVF & haemodialysis STREP PASSPORT Liver/surgery STREP PredictAD
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Virtual Physiological Humans
From Medical Images to
Personalized Cardiac Models
Nicholas Ayachehttp://www-sop.inria.fr/Asclepios/
Bio-ICT: The Heart in the Computer
2 April 2009
Credits
Asclepios Team past/current members
Academic, Clinical and Industrial Partners
2
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
3
X. PennecH. Delingette
O. ClatzG. Malandain
M. Sermesant
Project-Team Asclepios – March 2008
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
4
Medical Imaging Today
• Large Choice of in vivo modalities• High temporal and spatial resolution• Large parameter spaces• Large Databases• Emerging Modalities/Therapies
• Quantity of information too high : requires Computer Science
200 microns
Colon crypts
3
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
5
Computational Medical Image Analysis (1980 - Today)
• Assist Diagnosis• Objective quantitative measurements• fusion of multimodal, multidimensional,
euHeart: A Technical Project with a Clinical Focus
7
CONFIDENTIAL Philips Research Europe - Aachen 13
Nicholas Ayache Ercim Infinity April 09
13
The Consortium
Universities and research institutes
• INRIA, Sophia Antipolis, FR• INSERM, Rennes, FR• University of Karlsruhe, DE• UPF, Barcelona, SP• University of Sheffield, UK• University of Oxford, UK• Amsterdam Medical Center, NL
Hospitals and clinics
• KCL, London UK• DKFZ, Heidelberg, DE• INSERM, Rennes, FR• HSCM, Madrid, SP• Amsterdam Medical Center, NL
Industrial partners• Berlin Heart, DE• HemoLab, NL• Philips Healthcare, NL & SP• Philips Research, DE• PolyDimension, DE• Volcano, BE
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
14
Towards a Patient-Specific Computational Heart Model
Multi-level modeling
8
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
15
Average structure
1. Geometry & Statistics • Heart Database (E. McVeigh, NIH)
DTI Image Statistical Analysis J.M. Peyrat, M. Sermesant, X. Pennec, H. Delingette, C. Xu, E. McVeigh, N. Ayache A Computational Framework for the Statistical Analysis of Cardiac Diffusion Tensors: Application to a Small Database of Canine Hearts. IEEE Transactions on Medical Imaging, 26(11):1500-1514, November 2007
b) Phenomenological models FitzHugh-Nagumo, Aliev-Panfilov, Mitchell-Schaeffer,
c) Eikonal EquationKeener, Colli-Franzone
Cellular automata
PDE
FMA
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
20
Anisotropic Fast Marching for Depolarization Fronts
40 000 elements1s/beatEpicardial Pacing
Pseudo-potentialBlue: excitable
Red: depolarisedYellow: refractory
Depolarisation FrontBlue: depolarised side
Red: excitable sideBlack: Repolarisation Front
M. Sermesant, E. Konukoglu, H. Delingette, Y. Coudiere, P. Chinchaptanam, K.S. Rhode, R. Razavi, and N. Ayache. An anisotropic multi-front fast marching method for real-time simulation of cardiac electrophysiology.In Functional Imaging and Modeling of the Heart 2007 (FIMH'07), volume 4466 of LNCS, pp 160-169, June 2007.
11
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
21
2. Electrophysiological models
u action potentialD Diffusion tensor (fiber orientation)z repolarization variable
a) Ionic modelsHodgkin-Huxley, Luo-Rudy, Noble
b) Phenomenological models FitzHugh-Nagumo, Aliev-Panfilov, Mitchell-Schaeffer,
c) Eikonal EquationKeener, Colli-Franzone
( ) ( )
( )⎪⎪⎩
⎪⎪⎨
⎧
−=∂∂
−+∇=∂∂
czubtz
zufuDdivtu
Cellular automata
PDE
FMA
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
22
Electrical Simulation
Color : action potential u
( ) ( )
( )⎪⎪⎩
⎪⎪⎨
⎧
−=∂∂
−+∇=∂∂
czubtz
zufuDdivtu
12
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
23
3. Electro-mechanical Model
Kc stiffnessu action potentialεc strainσc stress
J. Bestel, F. Clément, and M. Sorine. A Biomechanical Model of Muscle ContractionMICCAI 2001.
Inspired byrheological modelOf Hill-Maxwell
Model of Bestel-Clément-Sorinenano
micro
méso
macro
ATP
sarcomeres
fibers
organ
active non-linear viscoelastic anisotropic incompressible material.
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
24
Electro-Mechanical Simulation
• Action potential u controls contractile element:
M. Sermesant, H. Delingette, N. Ayache. An Electromechanical Model of the Heart for Image Analysis and Simulation. IEEE Transactions on Medical Imaging. 2006 May;25(5):612-25.
M. Sermesant, H. Delingette, N. Ayache. An Electromechanical Model of the Heart for Image Analysis and Simulation. IEEE Transactions on Medical Imaging. 2006 May;25(5):612-25.
D. Chapelle, P Moireau, M. Sermesant, M. Fernandez, H. Delingette: The Digital Heart – INRIA DVD
14
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
27
Pathology SimulationLeft Bundle Branch Block
Slowed
9 times
D. Chapelle, P Moireau, M. Sermesant, M. Fernandez, H. Delingette: The Digital Heart – INRIA DVD
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
28
Personalized Cardiac Models
Clinical Data• Electrophysiology• Image Motion
in silico Heart Model• Simulated Electrophysiology• Simulated Motion
Feedback
Patient Parameters• Electrical• mechanical
Compare simulation & measurements to learn model parameters
• Moreau-Villeger, Delingette, Sermesant, Mc Veigh, N.A. et al., IEEE Trans. on BioEng. 2006
•Sermesant, Peyrat, Chinchapatnam, Billet, Mansi, Rhode, Delingette, Razavi, Ayache, Toward Patient-Specific Myocardial Models of the Heart, Heart Failure Clinics, July 2008.
15
29
In vivo clinical measurements (XMR Imaging)
Razavi R, Hill DL, Keevil SF, Miquel ME, Muthurangu V, Hegde S, Rhode K, Barnett M, van Vaals J, Hawkes DJ, Baker E. Cardiac catheterisation guided by MRI in children and adults with congenital heart disease. Lancet. 2003 Dec 6; 362(9399): 1877-82.
Reza Razavi, Guy’s and St Thomas’ Hospitals, London
Personalisation of Local Conductivity with Division of Imaging Sciences, KCL London
• Mean error ~3 ms,
•correlation with ischemic regions from late-enhancement MRI
P. Chinchapatnam, K. Rhode, M. Ginks, C.A. Rinaldi, P. Lambiase, R. Razavi, S. Arridge, M. Sermesant.Model-based Imaging of Cardiac Apparent Conductivity and Local Conduction Velocity for Diagnosis and Planning of Therapy. IEEE Transactions on Medical Imaging, in press
Prediction of BiV Pacing IsochronesJoint work with the Division of Imaging Sciences, King’s College London, UK
P. Chinchapatnam, K. Rhode, M. Ginks, C.A. Rinaldi, P. Lambiase, R. Razavi, S. Arridge, M. Sermesant. Model-based Imaging of Cardiac Apparent Conductivity and Local Conduction Velocity for Diagnosis and Planning of Therapy. IEEE Transactions on Medical Imaging (in press).
Depolarisation Isochrones
Conductivity Map
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
32
Minimize objective function to reproduce observed displacements:
Personalization of Mechanics
measurementsobservations from simulations
parameters state variables
I : set of instants when there are observations
• M. Sermesant, P. Moireau, O. Camara, J. Sainte-Marie, R. Andriantsimiavona, R. Cimrman, D.L.G. Hill, D. Chapelle, R. Razavi, Cardiac Function Estimation from MRI using a Heart Model and Data Assimilation: Advances and Difficulties, Medical Image Analysis, Vol 10, April 2006, pp 642-656.
• F. Billet, M. Sermesant, H. Delingette and N. Ayache: Cardiac Motion Recovery by Coupling an Electromechanical Model and Cine-MRI Data: First Steps, in Miccai’08 workshop on biomechanics, 2008.
Constraint :
17
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
33
Personalized Physiological Curves on a Control
Sermesant, Peyrat, Chinchapatnam, Billet, Mansi, Rhode, Delingette, Razavi, Ayache, Toward Patient-Specific Myocardial Models of the Heart, Heart Failure Clinics, July 2008.
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
34
Personalized Physiological Curves on a Patient with Right Ventricle
Overload
Sermesant, Peyrat, Chinchapatnam, Billet, Mansi, Rhode, Delingette, Razavi, Ayache, Toward Patient-Specific Myocardial Models of the Heart, Heart Failure Clinics, July 2008.
18
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
35
Clinical dataClinical data
Simulation
3D+t Segmentation
Personalised simulationof the heart motion
Patient with Chronic Pulmonary Valve Regurgitations (repaired Tetralogy of Fallot)
3D+t MRI
LV
RV
From segmentat
Health-e-Child European Project (coordinator : Siemens)
T. Mansi et al.
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
36
• Valve replacement : suppress regurgitations• Surgery Simulation of RV using SOFA using SOFA
framework framework www.sofa-framework.org
Simulating Surgery
T. Mansi, B. Andre, M. Lynch, M. Sermesant, H. Delingette, Y. Boudjemline, N. Ayache.Virtual Pulmonary Valve Replacement Interventions with a Personalised Cardiac Electromechanical Model, Workshop on 3D Physiological Human, in press.
Health-e-Child Project(coordinated by Siemens)
19
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
37
Assessing the Effect of Surgery
Before & after surgery
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
38
Patient with a LBBB
Sermesant M, Rhode K, Sanchez-Ortiz GI, Camara O, Andriantsimiavona R, Hegde S, Rueckert D, Lambiase P, Bucknall C, Rosenthal E, Delingette H, Hill DL, N.A., Razavi R. Simulation of cardiac pathologies using an electromechanical biventricular model and XMR interventional imaging. Med Image Anal. 2005 Oct;9(5):467-80.
Man, 68, Left Bundle Branch Block
20
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
39
Simulating Resynchronization on LBBB Patient
antero septal Infarct +Left Bundle Branch Block
EF = 41%
bi-ventricular resynchronization
EF = 47%
Slowed down
3 times
standard EF = 57%
LVBlood Volume
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
40
Resynchronization
scars
Woman, 60 years, LBBB
F. Billet, M. Sermesant, H. Delingette, and N. Ayache. Cardiac Motion Recovery by Coupling an Electromechanical Model and Cine-MRI Data: First Steps. Workshop MICCAI-2008, September 2008.
Conductivity map
Fiber map
Personalized mechanics
21
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
41
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
42
22
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
43
Simulation of BV pacing
Depolarization isochrones
ms
Coronary sinus catheter
Endocardial catheter
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
44
Simulation/measure of P & dP/dt
withoutpacing
withpacing
M. Sermesant, et al., Personalised Electromechanical Model of the Heart for the Prediction of the Acute Effects of Cardiac Resynchronisation Therapy, FIMH 2009.
23
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
45
Medical Perspectives
• Quantitative Diagnosis• From electrical and mechanical
parameters of 3-D model
• Realistic Simulation of Therapy• Surgery• BiV Cardiac Resynchronization• RF Ablation, • etc.
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
46
FIMH 20093-5 June
Nice, France
5th International Conference
Functional imaging and Modeling of the Heart
Philips, Siemens, Microsoft Research
24
Nicholas Ayache Ercim Infinity April 09
Towards Personnalized Cardiac Models
47
Conclusion: Virtual Physiological Patient
• Combining • in vivo digital images • in silico models of life
• Provides new tools• To analyze and simulate patient condition • To quantify diagnosis • To optimize therapy