In this project, we utilize a novel concept that embeds a generative variational auto-encoder (VAE) into the objective function of Bayesian optimization, providing an implicit low-dimensional (LD) search space that represents the generative code of the HD spatially-varying tissue properties. In addition, the VAE-encoded knowledge about the generative code is further used to guide the exploration of the search space. It is applied to estimating high-dimensional tissue excitability in a cardiac electrophysiological model.
I am a Ph.D. student in the Computing and Information Sciences department at Rochester Institute of Technology. My advisor is Dr. Linwei Wang. My research mainly focuses on novel development and adaptation of active learning, deep learning, and machine learning methods in application to inverse problems, model personalization, and uncertainty quantification. I primarily work in the domain of cardiac electrophysiology.
My husband, Kushal Kafle, is also a Ph.D. student at RIT and works on awesome research in the field of visual question answering.
July, 2018: My paper in MICCAI-2018 is selected for ORAL presentation (acceptance rate 4%).
May, 2018: My paper is accepted for publication in Medical Image Analysis (MeDIA).
May, 2018: I passed Ph.D. proposal defense.
May, 2018: My paper is accepted in MICCAI-2018 (early-accept).
Jan, 2018: I will be joining Philips Research, Cambridge, MA for summer internship.
Apr, 2017: Our paper is accepted for publication in CinC 2017.
Apr, 2017: My paper is accepted for publication in IEEE Transactions on Medical Imaging (IEEE TMI).
Mar, 2017: I received IPMI Scholarship and GCCIS Student Travel Fund.
Feb, 2017: My paper is accepted in the prestigious IPMI 2017 (acceptance rate ~20%).
Jun, 2016: I received MICCAI Student Travel Award.
Apr, 2016: My paper is accepted in MICCAI 2016 (early-accept, acceptance rate ~25%).
May, 2015: I passed the GCCIS Ph.D. research potential assesment.
High-dimensional Bayesian Optimization of Personalized Cardiac Model Parameters via an Embedded Generative Model
Jwala Dhamala, Sandesh, Ghimire, John L. Sapp, B. Milan Horacek, and Linwei Wang. “High-dimensional Bayesian Optimization of Personalized Cardiac Model Parameters via an Embedded Generative Model”. In: Proc. MICCAI: International Conference on Medical Image Computing and Computer-Assisted Intervention, LNCS (2018),
Local Parameter Estimation of a Cardiac Electrophysiologial Model
In this project, we developed a multi-scale spatially adaptive optimization approach to estimate the spatially varying cardiac tissue properties. Non-invasive Electrocardiogram (EKG) data is used for model personalization. The optimization is done by using a Gaussian process surrogate of the objective function. The multi-scale modeling of the cardiac mesh is done by using hierarchial clusteriong of the nodes in a cardiac mesh. The adaptive spatial resolution is obtained by computing the reward of refinement or coarsening any spatial cluster.
Jwala Dhamala, John L. Sapp, B. Milan Horacek, and Linwei Wang. “Spatially-Adaptive Multi-scale Optimization for Local Parameter Estimation: Application in Cardiac Electrophysiological Models”. In: Proc. MICCAI: International Conference on Medical Image Computing and Computer-Assisted Intervention, LNCS 9902 (2016), pp. 282-290.[PDF]
Jwala Dhamala, Hermenegild J. Arevalo, John L. Sapp, Milan Horacek, Katherine C. Wu, Natalia A. Trayanova, and Linwei Wang. “Spatially-Adaptive Multi-Scale Optimization for Local Parameter Estimation in Cardiac Electrophysiology”. In: IEEE Transactions on Medical Imaging 36 (2017), pp. 1966-1978.[PDF]
Uncertainty Quantification of Model Parameters
The quantification of uncertainty in model parameters is challenging because the posterior distribution of the parameters given the measurement data is non-Gaussian and the evaluation of the model is computationally expensive. In this project, we build a surrogate of this complicated and computationally expensive posterior distribution and utilize it to obtain a MCMC sampling with higher acceptance rate. The surrogate posterior pdf is used to accelerate the sampling of the true posterior pdf and not as a replacement.
Jwala Dhamala, John L. Sapp, B. Milan Horacek, and Linwei Wang. “Quantifying the Uncertainty in Model Parameters using Gaussian Process-Based Markov Chain Monte Carlo: An Application to Cardiac Electrophysiological Models”. In: Proc. IPMI: International Conference on Information Processing in Medical Imaging, LNCS 10265 (2017), pp. 223-235.[PDF]
Jwala Dhamala, Hermenegild J. Arevalo, John L. Sapp, Milan Horacek, Katherine C. Wu, Natalia A. Trayanova, and Linwei Wang. “Quantifying the Uncertainty in Model Parameters using Gaussian Process-Based Markov Chain Monte Carlo in Cardiac Electrophysiology” In: Medical Image Analysis (2018).
Hackathon on Premature Ventricular Contraction
Consortium of ECG Imaging
Jwala Dhamala, Jaume Coll-Font, Jess D Tate, Maria S Guillem, Dana H Brooks, Rob S MacLeod, and Linwei Wang. “Overcoming Barriers to Quantification and Comparison of Electrocardiographic Imaging Methods: A Community-Based Approach”. In: Proc. CinC: Computing in Cardiology, IEEE (2017) To Appear
Jaume Coll-Font*, Dana H Brooks, Peter M van Dam, Jwala Dhamala, Olaf Dössel, Maria de la Salud Guillem Sánchez, Rob MacLeod, Danila Potyagaylo, Walther Schulze, Jess D Tate and Linwei Wang, “The Consortium on Electrocardiographic Imaging”, Computing in Cardiology (CINC), Vancouver, Canada, September 2016
Women in Computing, Pre-orientation program
Rochester Institute of Technology