I am a Ph.D. candidate at B. Thomas Golisano College of Computing and Information Sciences at Rochester Institute of Technology. I am working as a research assistant at Computational Biomedicine Lab (CBL), which is directed by my advisor, Dr. Linwei Wang. My primary research interest is in improving the generalization of deep learning algorithms for biomedical data. Towards this, I consider two general perspectives: learning disentangled representation and regularizing neural networks.

I am looking for full-time research-based positions either in academia or industry. Please let me know if you have any pointer related to my research experience and interests.

TIMELINE EVENTS

Summer 2020 Research Intern at Google.
May 2020 Two papers accepted in MICCAI 2020.
May 2020 Successfully defended proposal for Ph.D. dissertation.
Feb 2020 Our work "A Machine Learning Approach for Computer-Guided Localization of the Origin of Ventricular Tachycardia Using 12-Lead Electrocardiograms" is accepted by  Heart Rhythm Society  (HRS 2020) and selected to participate in the special session to highlight the most novel and innovative abstracts of 2020.
August 2019 Our journal paper "Sequential Factorized Autoencoder for Localizing the Origin of Ventricular Activation From 12-Lead Electrocardiograms" is accepted in IEEE Transactions on Biomedical Engineering.
August 2019 Our paper "Improving Disentangled Representation Learning with the Beta Bernoulli Process" is accepted in ICDM 2019.
June 2019 Our paper "Semi-Supervised Learning by Disentangling and Self-Ensembling over Stochastic Latent Space" is accepted in MICCAI 2019.
Summer 2019 Internship at Verisk AI Lab as AI/ML Research Intern.
December 2018 Presented poster at BNP@NeurIPS and SOCML (highly recommended "unconference").
November 2018 Two papers accepted in ML4H@NeurIPS and BNP@NeurIPS.
Summer 2018 Mentor for Research Experience for Undergraduates (NSF-REU) program at RIT for the project "Multi-modal sensing and quantification of atypical attention in autism spectrum disorder."
November 2017 Our paper "Learning disentangled representation from 12-lead electrograms: application in localizing the origin of Ventricular Tachycardia" was accepted in the Health Intelligence Workshop, AAAI 2018.
June 2017 Our paper "Automatic Coordinate Prediction of the Exit of Ventricular Tachycardia from 12-lead Electrocardiogram" was accepted to CinC 2017. The paper was selected as a semi-finalist for Rosanna Degani Young Investigator Award.
Summer 2017 Mentor for Research Experience for Undergraduates (NSF-REU) program at RIT for the project "Attention and behavior of students in online vs. face-to-face learning contexts."
May 2017 Successfully defended the Ph.D. Research Potential Assesment.
May 2017 Our paper "Disentangling Inter-Subject Variations: Automatic Localization of Ventricular Tachycardia Origin from 12-Lead Electrocardiograms" was accepted to ISBI 2017.
August 2016 Joined Rochester Institute of Technology for Ph.D. in Computing and Information Sciences.

RESEARCH WORKS

    2020

    Ryan Missel, Prashnna K Gyawali, Jaideep Vitthal Murkute, Zhiyuan Li, Shijie Zhou, Amir AbdelWahab, Jason Davis, James Warren, John L Sapp and Linwei Wang. "A hybrid machine learning approach to localizing the origin of ventricular tachycardia using 12-lead electrocardiograms". Computers in Biology and Medicine.

    Prashnna K. Gyawali, Sandesh Ghimire and Linwei Wang. "Enhancing Mixup-based Semi-Supervised Learning with Explicit Lipschitz Regularization". IEEE International Conference on Data Mining (ICDM-20). [Code]

    Prashnna K. Gyawali, Sandesh Ghimire, Pradeep Bajracharya, Zhiyuan Li and Linwei Wang. "Semi-supervised Medical Image Classification with Global Latent Mixing". International Conference on Medical Image Computing & Computer Assisted Intervention (MICCAI-20). (early acceptance). [Code]

    Xiajun Jiang, Sandesh Ghimire, Prashnna K. Gyawali and Linwei Wang. "Learning Geometry-Dependent and Physics-Based Inverse Image Reconstruction". International Conference on Medical Image Computing & Computer Assisted Intervention (MICCAI-20).

    Zhiyuan Li, Jaideep Vitthal Murkute, Prashnna K. Gyawali and Linwei Wang. "Progressive Learning and Disentanglement of Hierarchical Representations". International Conference on Learning Representations (ICLR-20). (accepted for spotlight).

    2019

    Prashnna K. Gyawali, B. Milan Horacek, John Sapp and Linwei Wang. "Sequential Factorized Autoencoder for Localizing the Origin of Ventricular Activation From 12-Lead Electrocardiograms". IEEE Transacations on Biomedical Engineering. (TBE).

    Prashnna K. Gyawali, Zhiyuan Li, Cameron Knight, Sandesh Ghimire, B. Milan Horacek, John Sapp and Linwei Wang. "Improving Disentangled Representation Learning with the Beta Bernoulli Process". IEEE International Conference on Data Mining (ICDM-19). [Code]

    Prashnna K. Gyawali*, Zhiyuan Li*, Sandesh Ghimire, and Linwei Wang. "Semi-Supervised Learning by Disentangling and Self-Ensembling over Stochastic Latent Space". International Conference on Medical Image Computing & Computer Assisted Intervention (MICCAI-19). * equal contribution [Code]

    Sandesh Ghimire, Prashnna K. Gyawali, Jwala Dhamala, John Sapp, B. Milan Horacek, and Linwei Wang. "Improving Generalization of Deep Networks for Inverse Reconstruction of Image Sequences". International Conference on Information Processing in Medical Imaging (IPMI-19).

    2018

    Prashnna K. Gyawali, Cameron Knight, Sandehs Ghimire, John Sapp, B. Milan Horacek and Linwei Wang. "Deep Generative Model with Beta Bernoulli Process for Modeling and Learning Confounding Factors". All of Bayesian Nonparametric workshop at NeurIPS 2018 (BNP@NeurIPS 2018).

    Sandesh Ghimire, Jwala Dhamala, Prashnna K. Gyawali, John Sapp, B. Milan Horacek and Linwei Wang. "Generative Modeling and Inverse Imaging of Cardiac Transmembrane Potential". International Conference on Medical Image Computing & Computer Assisted Intervention (MICCAI-18) (also MLH4@NeurIPS 2018).

    Prashnna K. Gyawali, B. Milan Horacek, John Sapp, and Linwei Wang. "Learning disentangled representation from 12-lead electrocardiograms: application in localizing the origin of Ventricular Tachycardia". Health Intelligence Workshop, 32nd AAAI Conference on Artificial Intelligence (AAAI-18). [Code]

    2017

    Prashnna K. Gyawali, Shuhang Chen, Huafeng Liu, B. Milan Horacek, John Sapp and Wang L. "Automatic Coordinate Prediction of the Exit of Ventricular Tachycardia from 12-lead Electrocardiogram". Computing in Cardiology (CinC-17).

    Erin Coppola, Prashnna K. Gyawali, Nihar Vanjara, Dan Giaime and Linwei Wang. "Atrial Fibrillation Classification from a Short Single Lead ECG Recording Using Hierarchical Classifier". Computing in Cardiology (CinC-17).

    Shuhang Chen, Prashnna K. Gyawali, Huafeng Liu, B. Milan Horacek, John Sapp and Linwei Wang. "Disentangling inter-subject variations: Automatic localization of ventricular tachycardia from 12-lead electrocardiograms". IEEE International Symposium on Biomedical Imaging (ISBI-17).

    2015

    Shailesh Acharya, Ashok K. Pant and Prashnna K. Gyawali. "Deep learning based large scale handwritten Devnagari character recognition". IEEE International Conference on Software, Knowledge, Information Management and Applications (SKIMA-15).   [Dataset]

    Ashok K. Pant, Prashnna K. Gyawali and Shailesh Acharya. "Automatic Nepali Number Plate Recognition with Support Vector Machines". IEEE International Conference on Software, Knowledge, Information Management and Applications (SKIMA).  [Dataset]