Research Interests
Education
BS-MS Dual Degree (Inter-Disciplinary)
Indian Institute of Science Education and Research (IISER), Pune, India
Master's Thesis in Data Science: Deep Learning for Breast Cancer Risk Prediction from Mammography
Master's Thesis
Jug Lab, Human Technopole, Milan, Italy
Advisor: Dr. Florian Jug
Research Experience
Biological Age Estimation from Mammography for Breast Cancer Risk Stratification
- Developed an end-to-end deep learning framework for predicting biological age from screening mammograms, achieving MAE of 3.2 years—a 60% improvement over previous state-of-the-art methods.
- Designed a novel architecture combining distilled DINOv3 backbones with hierarchical multi-head attention mechanisms for robust feature extraction.
- Investigating the clinical utility of "age gap" (biological vs. chronological age) as a novel imaging biomarker for personalized breast cancer risk assessment.
- Trained models on large-scale multi-institutional mammography datasets on HPC infrastructure.
Deep CNN Architecture for Histopathology Image Classification
- Designed a custom deep CNN with stacked convolutional blocks, batch normalization, and progressive dropout for multi-class lung and colon cancer classification.
- Achieved near-perfect performance with 99.58% test accuracy and 0.9999 AUC on the LC25000 benchmark, validating the custom architecture against standard ResNet baselines.
- Conducted ablation studies to analyze the contribution of architectural components to model performance.
Single-Cell Transcriptomic Analysis of YAP/TAZ in Breast Cancer
- Developed a reproducible Python-based scRNA-seq analysis pipeline using Scanpy for the laboratory.
- Investigated co-expression patterns and functional redundancy of YAP and TAZ transcriptional co-activators in breast cancer cells versus normal mammary epithelial cells.
- Applied dimensionality reduction, clustering, and differential expression analysis to characterize cell-type-specific transcriptional programs.
Spatial Protein Expression Analysis in Breast Tumor Microenvironment
- Analyzed THBS1 expression patterns across tumor cell types using multiplex immunofluorescence imaging with PanCK and CD3 cell-type markers.
- Performed quantitative image analysis using Halo and inForm software to characterize spatial relationships in the tumor microenvironment.
Computational Optimization for Sustainable Aviation Fuel Production
- Headed Measurements division (statistical modeling) and the Wet Lab team (gene knockouts).
- Designed optimal production conditions using Design of Experiments (DoE) and statistical learning approaches.
- Team achieved: Best Education Award, nomination for Best Climate Crisis Project, and Top-10 global ranking among undergraduate teams at iGEM Jamboree, Paris.
Technical Skills
Machine Learning & AI
- PyTorch, scikit-learn
- Vision Transformers (ViT, DINOv3, Swin)
- CNNs, Attention Mechanisms
- Transfer Learning, Fine-tuning
- Model Interpretability (GradCAM, SHAP)
Programming & Computing
- Python, C++, Bash scripting
- HPC/SLURM, Distributed Training
- Git, Linux
Medical Image Analysis
- Mammography, Histopathology
- Multiplex IF (Halo, inForm)
- DICOM handling, Image preprocessing
- Whole-slide image analysis
Bioinformatics & Data Science
- Single-cell RNA-seq (Scanpy, Seurat)
- Statistical Modeling, DoE
- Data Visualization (Matplotlib, Seaborn)
- Molecular Docking (AutoDock)
Relevant Coursework
Machine Learning
Deep Learning
Computer Vision
Statistical Inference
Linear Algebra
Bayesian Approaches
Computational Biology
Cancer Biology
Molecular Cell Biology
Selected Presentations
Poster: "Deep Learning-Based Mammographic Age Estimation: Towards an Image-Based Biomarker for Cancer Risk Prediction"
Cancer Research UK Data-Driven Cancer Research Conference
Edinburgh International Conference Centre, UK | February 24–26, 2026 (Accepted)
Honors & Awards
- Best Education Award and Top-10 Global Undergraduate Ranking, iGEM Jamboree, Paris (2023)
- Nominated for Best Climate Crisis Project, iGEM Jamboree, Paris (2023)
- G-20 Delegate, "Accessible Science" event, representing IISER Pune (June 2023)
- Graduated with Distinction, 12th Grade (92.1%) - 2020
Leadership & Service
- Athletics Captain, Sports Club, IISER Pune (September 2024 – January 2025)
- Head of Media, Student Media Body, IISER Pune (November 2022 – December 2023)
References
Dr. Florian Jug
Head, Computational Biology Research Centre
Human Technopole, Milan
florian.jug@fht.orgDr. Ruggiero Santeramo
Postdoctoral Researcher (ML Specialist)
Jug Lab, Human Technopole, Milan
ruggiero.santeramo@fht.orgDr. Leelavati Narlikar
Assoc. Professor & Deputy Chair, Data Science
IISER Pune
leelavati@iiserpune.ac.in