I am interested in exploring machine learning and deep learning methods for biomedical applications and medical imaging. Limited availability of labeled datasets to train machine learning models and large-scale variability in the datasets are some of the common challenges that I am interested in solving for the successful use of AI in this domain. Currently, I am exploring machine learning methods for genomics, where I am particularly focused on improving prediction for disease risk for Alzheimer’s disease for minority populations.

Professional Education

  • Doctor of Philosophy, Rochester Institute of Technology, Computing and Information Science (2021)
  • BE, Institute of Engineering, Pulchowk Campus, Electronics & Communication (2014)

Stanford Advisors

All Publications

  • Learning to Disentangle Inter-subject Anatomical Variations in Electrocardiographic Data. IEEE transactions on bio-medical engineering Gyawali, P. K., Murkute, J. V., Toloubidokhti, M., Jiang, X., Horacek, B. M., Sapp, J. L., Wang, L. 2021; PP


    This work investigates the possibility of disentangled representation learning of inter-subject anatomical variations within electrocardiographic (ECG) data.Since ground truth anatomical factors are generally not known in clinical ECG for assessing the disentangling ability of the models, the presented work first proposes the SimECG data set, a 12-lead ECG data set procedurally generated with a controlled set of anatomical generative factors. Second, to perform such disentanglement, the presented method evaluates and compares deep generative models with latent density modeled by nonparametric Indian Buffet Process to account for the complex generative process of ECG data.In the simulated data, the experiments demonstrate, for the first time, concrete evidence of the possibility to disentangle key generative anatomical factors within ECG data in separation from task-relevant generative factors. We achieve a disentanglement score of 92.1\% while disentangling five anatomical generative factors and the task-relevant generative factor. In both simulated and real-data experiments, this work further provides quantitative evidence for the benefit of disentanglement learning on the downstream clinical task of localizing the origin of ventricular activation. Overall, the presented method achieves an improvement of around 18.5\%, and 11.3\% for the simulated dataset, and around 7.2\%, and 3.6\% for the real dataset, over baseline CNN, and standard generative model, respectively.These results demonstrate the importance as well as the feasibility of the disentangled representation learning of inter-subject anatomical variations within ECG data.This work suggests the important research direction to deal with the well-known challenge posed by the presence of significant inter-subject variations during an automated analysis of ECG data.

    View details for DOI 10.1109/TBME.2021.3108164

    View details for PubMedID 34460360

  • Sequential Factorized Autoencoder for Localizing the Origin of Ventricular Activation From 12-Lead Electrocardiograms IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING Gyawali, P., Horacek, B., Sapp, J. L., Wang, L. 2020; 67 (5): 1505-1516


    This work presents a novel approach to handle the inter-subject variations existing in the population analysis of ECG, applied for localizing the origin of ventricular tachycardia (VT) from 12-lead electrocardiograms (ECGs).The presented method involves a factor disentangling sequential autoencoder (f-SAE) - realized in both long short-term memory (LSTM) and gated recurrent unit (GRU) networks - to learn to disentangle the inter-subject variations from the factor relating to the location of origin of VT. To perform such disentanglement, a pair-wise contrastive loss is introduced.The presented methods are evaluated on ECG dataset with 1012 distinct pacing sites collected from scar-related VT patients during routine pace-mapping procedures. Experiments demonstrate that, for classifying the origin of VT into the predefined segments, the presented f-SAE improves the classification accuracy by 8.94% from using prescribed QRS features, by 1.5% from the supervised deep CNN network, and 5.15% from the standard SAE without factor disentanglement. Similarly, when predicting the coordinates of the VT origin, the presented f-SAE improves the performance by 2.25 mm from using prescribed QRS features, by 1.18 mm from the supervised deep CNN network and 1.6 mm from the standard SAE.These results demonstrate the importance as well as the feasibility of the presented f-SAE approach for separating inter-subject variations when using 12-lead ECG to localize the origin of VT.This work suggests the important research direction to deal with the well-known challenge posed by inter-subject variations during population analysis from ECG signals.

    View details for DOI 10.1109/TBME.2019.2939138

    View details for Web of Science ID 000530299200028

    View details for PubMedID 31494539

    View details for PubMedCentralID PMC7051887

  • Deep Adaptive Electrocardiographic Imaging with Generative Forward Model for Error Reduction Functional Imaging and Modeling of the Heart - FIMH Toloubidokhti, M., Gyawali, P., Gharbia, O. A., Jiang, X., Font, J. C., Bergquist, J. A., Zenger, B., Good, W. W., Brooks, D. H., MacLeod, R. S., Wang, L. 2021
  • Semi-Supervised Learning for Eye Image Segmentation ACM Symposium on Eye Tracking Research and Applications - ETRA Chaudhary*, A. K., Gyawali*, P., Wang, L., Pelz, J. B. 2021
  • A hybrid machine learning approach to localizing the origin of ventricular tachycardia using 12-lead electrocardiograms COMPUTERS IN BIOLOGY AND MEDICINE Missel, R., Gyawali, P. K., Murkute, J., Li, Z., Zhou, S., AbdelWahab, A., Davis, J., Warren, J., Sapp, J. L., Wang, L. 2020; 126: 104013


    Machine learning models may help localize the site of origin of ventricular tachycardia (VT) using 12-lead electrocardiograms. However, population-based models suffer from inter-subject anatomical variations within ECG data, while patient-specific models face the open challenge of what pacing data to collect for training.This study presents and validates the first hybrid model that combines population and patient-specific machine learning for rapid "computer-guided pace-mapping". A population-based deep learning model was first trained offline to disentangle inter-subject variations and regionalize the site of VT origin. Given a new patient with a target VT, an on-line patient-specific model -- after being initialized by the population-based prediction -- was then built in real time by actively suggesting where to pace next and improving the prediction with each added pacing data, progressively guiding pace-mapping towards the site of VT origin.The population model was trained on pace-mapping data from 38 patients and the patient-specific model was subsequently tuned on one patient. The resulting hybrid model was tested on a separate cohort of eight patients in localizing 1) 193 LV endocardial pacing sites, and 2) nine VTs with clinically determined exit sites. The hybrid model achieved a localization error of 5.3 ± 2.6 mm using 5.4 ± 2.5 pacing sites in localizing LV pacing sites, achieving a significantly higher accuracy with a significantly smaller amount of training sites in comparison to models without active guidance.The presented hybrid model has the potential to assist rapid pace-mapping of interventional targets in VT.

    View details for DOI 10.1016/j.compbiomed.2020.104013

    View details for Web of Science ID 000582723600014

    View details for PubMedID 33002841

    View details for PubMedCentralID PMC7606703

  • Progressive learning and disentanglement of hierarchical representations EIGHTH INTERNATIONAL CONFERENCE ON LEARNING REPRESENTATIONS (ICLR) Li, Z., Murkute, J. V., Gyawali, P., Wang, L. 2020
  • Learning Geometry-Dependent and Physics-Based Inverse Image Reconstruction Medical Image Computing and Computer Assisted Intervention – MICCAI Jiang, X., Ghimire, S., Dhamala, J., Li, Z., Gyawali, P., Wang, L. 2020
  • Semi-supervised Learning by Disentangling and Self-ensembling over Stochastic Latent Space Gyawali, P., Li, Z., Ghimire, S., Wang, L., Shen, D., Liu, T., Peters, T. M., Staib, L. H., Essert, C., Zhou, S., Yap, P. T., Khan, A. SPRINGER INTERNATIONAL PUBLISHING AG. 2019: 766-774
  • Improving Disentangled Representation Learning with the Beta Bernoulli Process Gyawali, P., Li, Z., Knight, C., Ghimire, S., Horacek, B., Sapp, J., Wang, L., Wang, J., Shim, K., Wu IEEE. 2019: 1078-1083
  • Improving Generalization of Deep Networks for Inverse Reconstruction of Image Sequences Ghimire, S., Kumar, P., Dhamala, G., Sapp, J. L., Horacek, M., Wang, L., Chung, A. C., Gee, J. C., Yushkevich, P. A., Bao, S. SPRINGER INTERNATIONAL PUBLISHING AG. 2019: 153-166
  • Generative Modeling and Inverse Imaging of Cardiac Transmembrane Potential Ghimire, S., Dhamala, J., Gyawali, P., Sapp, J. L., Horacek, M., Wang, L., Frangi, A. F., Schnabel, J. A., Davatzikos, C., AlberolaLopez, C., Fichtinger, G. SPRINGER INTERNATIONAL PUBLISHING AG. 2018: 508-516
  • Automatic Coordinate Prediction of the Exit of Ventricular Tachycardia From 12-Lead Electrocardiogram Gyawali, P. K., Chen, S., Liu, H., Horacek, B., Sapp, J. L., Wang, L., IEEE IEEE COMPUTER SOC. 2017
  • Atrial Fibrillation Classification from a Short Single Lead ECG Recording Using Hierarchical Classifier Coppola, E. E., Gyawali, P. K., Vanjara, N., Giaime, D., Wang, L., IEEE IEEE COMPUTER SOC. 2017
  • Deep learning based large scale handwritten Devanagari character recognition 2015 9th International Conference on Software, Knowledge, Information Management and Applications (SKIMA) Acharya, S., Pant, A. K., Gyawali, P. 2015