Research Impact

With 152 citations across 7 published papers, my research journey has been a fascinating exploration of how artificial intelligence can revolutionize healthcare and biology. From teaching machines to read mammograms to predicting protein behavior, each project has been an adventure in pushing the boundaries of what’s possible.

Key Research Areas

🏥 Healthcare AI & Medical Imaging

My work in medical imaging has been particularly close to my heart. Working with doctors and clinicians, I’ve developed AI systems that can detect early signs of disease in medical scans. There’s something incredibly fulfilling about knowing that lines of code could potentially save lives by catching diseases earlier than human eyes might.

Notable Achievement: Developed a weakly-supervised segmentation model using U-Net that achieved 97.86% accuracy in mammogram analysis, providing explainable predictions that doctors can trust and understand.

🧬 Computational Biology & Protein Analysis

The microscopic world of proteins fascinates me endlessly. My research in protein-ligand interactions has achieved 95.99% accuracy in predicting how proteins and molecules interact - think of it as matchmaking at the molecular level! This work has implications for drug discovery and understanding diseases like arthritis, cancer, and depression.

Impact: Created an open-source web server that automates feature generation for protein analysis, making complex computational biology accessible to researchers worldwide.

🔬 Computer Vision in Clinical Settings

During my time at IIIT Hyderabad’s Center for Visual Information Technology, I collaborated with medical professionals to develop AI systems for analyzing ultrasound scans and X-rays. This work involved creating landmark detection systems for fetal ultrasound scans - essentially teaching computers to see what doctors see.

Research Philosophy

My approach to research is driven by practical impact rather than just academic achievement. I believe in:

  • Explainable AI: Making sure doctors and researchers understand how AI makes decisions
  • Open Science: Sharing code and methodologies to accelerate collective progress
  • Interdisciplinary Collaboration: Working closely with domain experts to solve real problems
  • Reproducible Research: Ensuring others can build upon and validate my work

Publications & Citations

My research has been published in peer-reviewed journals and conferences, with work spanning:

  • Deep learning for medical image analysis
  • Protein-ligand interaction prediction
  • Computer vision applications in healthcare
  • Weakly-supervised learning techniques

Google Scholar Profile: 152 citations and counting

Current Research Interests

I’m constantly exploring new frontiers in AI/ML, particularly:

  • Multimodal AI: Combining different types of data (images, text, signals) for better predictions
  • Edge AI: Deploying sophisticated models on resource-constrained devices
  • Sustainable AI: Developing more energy-efficient machine learning models
  • AI in Manufacturing: My current role at Merck has opened new avenues in process optimization

Collaboration & Mentorship

Research is never a solo endeavor! I’ve had the privilege of working with brilliant minds across different institutions and am always open to collaborations. I also enjoy mentoring students and junior researchers - there’s nothing quite like seeing someone have their “eureka!” moment.