PHD RESEARCHER · WASHINGTON UNIVERSITY IN ST. LOUIS
AI researcher and builder.
I combine images, language, and domain knowledge to build systems for medical imaging and assistive technology. At WashU, my PhD research focuses on multimodal segmentation with limited labeled data.
01 / FLAGSHIP PROJECTS
Ph.D. research & research software
A research platform bringing together report-guided SPECT/CT segmentation, multi-planar visualization, quantitative analysis, and structured export.
My work: segmentation methods and research software at CMIT Lab.
0.73 → 0.80 mean organ Dice after SPECT refinement · independent test cohort
Explore the case study
Assistive technology & entrepreneurship
Assistive computer vision for people with visual impairments, connecting early navigation prototypes with the development of AI-powered devices at Oculosense.
My role: Co-founder & CTO, September 2021–August 2023.
12 FPS object detection in the early navigation prototype
Explore the case studyEARLIER RESEARCH & PROTOTYPES
Master’s thesis · Generative AI
An ACGAN study of class-conditioned synthetic skin-image generation, exploring the effects of generator/discriminator optimization in a small-data setting.
Bachelor’s capstone · Conversational AI
A Telegram-based conversational prototype combining sentiment analysis, rule-based dialogue, and translation support for multilingual interaction.
02 / MY JOURNEY
My work has grown from building physical interfaces to studying how machines interpret images and language. At NIT Hamirpur, I worked on brain-controlled devices, robotics, and assistive vision. With Drishti, I connected computer vision to hardware and went on to co-found Oculosense, serving as CTO from 2021 to 2023.
Alongside that product work, I explored word sense disambiguation, graph-based text classification, and conditional image generation. I joined CMIT Lab at WashU for summer research in 2022 and returned for my PhD in 2023. Today, I develop multimodal segmentation methods and TheraSeg, the software that brings those methods into an inspectable imaging workflow.
2018–2022
Brain-controlled devices, gesture-controlled robots, and hands-on engineering at NIT Hamirpur.
2020–2023
The Drishti navigation prototype, followed by co-founding Oculosense and leading technical development as CTO.
2021–2023
Word sense disambiguation, inductive GNN research at LBR Research Labs, and my conditional-GAN master’s thesis.
2022–Present
Summer research at CMIT Lab, followed by my PhD at WashU and current work on segmentation and TheraSeg.
03 / SELECTED RESEARCH
How can anatomical priors, language context, and learned representations help when supervision is limited? My work explores this question through medical imaging, word sense disambiguation, and graph-based text classification.
CURRENT RESEARCH
Combining SPECT uptake, CT anatomy, and clinical report information to segment organs and tumors without task-specific supervised segmentation training.
Low-resource language learning
Using unsupervised sense induction to improve lexical choices in English-to-Indic statistical machine translation, evaluated across five translation directions.
Graph representation learning
Comparing gated, convolutional, and attention-based graph models that represent each document independently, with experiments on AG News, IMDB, and DBpedia.
04 / RECOGNITION & COMMUNITY
Winner · $25,000 lab research funding
Siteman Cancer Center · MCBP & TIP Research Programs AI Retreat · June 2026
Best Poster
Washington University in St. Louis · March 2025
$5,000 research grant
Society of Nuclear Medicine and Molecular Imaging (SNMMI) · March 2024
$5,000 fellowship
Washington University in St. Louis · August 2023
I’ve taught imaging mathematics, linear algebra, and digital image processing. Through ScienceCoach, I coordinated eight graduate volunteers for hands-on engineering sessions with 22 high-school students and four teachers.
05 / GET IN TOUCH
I’m interested in research collaborations and opportunities involving multimodal learning, computer vision, and useful AI systems.
[email protected] · St. Louis, Missouri