Jay Sawant

Hello, I am Jay Sawant, a Master's student in Data Science at University of California, San Diego and an alumnus of IIT Bombay, where I completed a Dual Degree in Electrical Engineering (Communication and Signal Processing).

My research spans medical imaging and autonomous driving. At UCSD Health’s Lab of Cognitive Imaging, I work on deep learning for epilepsy: CNN-based methods for Temporal Lobe Epilepsy classification from MRI, conditional latent diffusion models for generating synthetic 3D T1 brain MRIs, and vision–language models that combine imaging with clinical text.

I am also interested in learning-based autonomous driving. As a Research Engineer Intern at Netradyne, I worked on imitation learning, benchmarking large camera-only and multi-modal models, scaling data for simpler architectures, and exploring temporal attention for closed-loop driving performance on benchmarks such as CARLA Leaderboard and Bench2Drive.

Updates

Jun, 2025	Started working as an AI Research Intern at Netradyne.
Nov, 2024	Started working as a Graduate Student Researcher in the Lab of Cognitive Imaging at UC San Diego Health.
Sep, 2024	Started pursuing the Master of Science in Data Science program at the University of California, San Diego
Jul, 2023	Joined as a SDE at Enphase Energy in Bangalore, India
Jun, 2023	Graduated with a Dual Degree (B.Tech + M.Tech) in Electrical Engineering from IIT Bombay
Jun, 2022	Interned at Qure.ai as a AI Scientist and researched on Self-supervised and Contrastive learning techniques
Jun, 2021	Interned at Qualcomm as a Machine Learning Summer Intern. Received a Pre-Placement Interview offer and a recommendation for 2nd internship at Qualcomm for exemplary performance

Selected publications

JPI
Efficient quality control of whole slide pathology images with human-in-the-loop training

Abhijeet Patil, Harsh Diwakar, Jay Sawant, Nikhil Cherian Kurian, Subhash Yadav, Swapnil Rane, Tripti Bameta, and Amit Sethi

Journal of Pathology Informatics 2023

Abs Bib

Histopathology whole slide images (WSIs) are being widely used to develop deep learning-based diagnostic solutions, especially for precision oncology. Most of these diagnostic softwares are vulnerable to biases and impurities in the training and test data which can lead to inaccurate diagnoses. For instance, WSIs contain multiple types of tissue regions, at least some of which might not be relevant to the diagnosis. We introduce HistoROI, a robust yet lightweight deep learning-based classifier to segregate WSI into 6 broad tissue regions—epithelium, stroma, lymphocytes, adipose, artifacts, and miscellaneous. HistoROI is trained using a novel human in-the-loop and active learning paradigm that ensures variations in training data for labeling efficient generalization. HistoROI consistently performs well across multiple organs, despite being trained on only a single dataset, demonstrating strong generalization. Further, we have examined the utility of HistoROI in improving the performance of downstream deep learning-based tasks using the CAMELYON breast cancer lymph node and TCGA lung cancer datasets. For the former dataset, the area under the receiver operating characteristic curve (AUC) for metastasis versus normal tissue of a neural network trained using weakly supervised learning increased from 0.88 to 0.92 by filtering the data using HistoROI. Similarly, the AUC increased from 0.88 to 0.93 for the classification between adenocarcinoma and squamous cell carcinoma on the lung cancer dataset. We also found that the performance of the HistoROI improves upon HistoQC for artifact detection on a test dataset of 93 annotated WSIs. The limitations of the proposed model are analyzed, and potential extensions are also discussed.
@article{PATIL2023100306, title = {Efficient quality control of whole slide pathology images with human-in-the-loop training}, journal = {Journal of Pathology Informatics}, volume = {14}, pages = {100306}, year = {2023}, issn = {2153-3539}, url = {https://doi.org/10.1016/j.jpi.2023.100306}, doi = {10.1016/j.jpi.2023.100306}, author = {Patil, Abhijeet and Diwakar, Harsh and Sawant, Jay and Kurian, Nikhil Cherian and Yadav, Subhash and Rane, Swapnil and Bameta, Tripti and Sethi, Amit}, keywords = {WSI pre-processing, Weakly supervised learning, Quality control, Generalization} }