MAISys research group at IIT Jodhpur is constantly working towards the development and deployment of AI-driven medical technologies. We are working in close collaboration with hospitals and health centers. Dedicated AI solutions for clinical problems and translational research are our primary interests.
Publication Updates
- Gajendra Singh, Aiman Farooq, Azad Singh, Deepak Mishra, Dr Rahul Chaudhury, Dr. Pushpinder Singh Khera, Guide2Heart: Proximity Guidance for Standard Echocardiographic View@ MICCAI 2025(ASMUS)
- Aiman Farooq*, Chandisha Das*, Deepak Mishra, Uncertainty-Aware Cross-Modal Attention for Breast Cancer Classification in Ultrasound Imaging @ MICCAI(Deep Breath 2025)
- Mayank Kumar, Deepak Mishra,SFLAME: Federated Learning with Masked Autoencoders and Mean-prototypes Embedding for Sparsely Labeled Medical Images, IEEE TETCI
- Azad Singh, Deepak Mishra,Scale-Aware Adaptive Feature Quantization for Robust Medical Image Representation Learning, ECAI
- Mayank Kumar, Deepak Mishra,AR2FL: Anomaly-Resistant Robust Framework for Federated Learning, IEEE TAI
- Shripad Pate*, Aiman Farooq*, Suvrankar Datta, Dr.Musadiq Aadil Sheikh, Dr.Atin Kumar, and Deepak Mishra, Fine-Grained Rib Fracture Diagnosis with Hyperbolic Embeddings: A Detailed Annotation Framework and Multi-Label Classification Model, Early Accept MICCAI2025
- Vandan Gorade, Azad Singh and Deepak Mishra, OTCXR: Rethinking Self-supervised Alignment using Optimal Transport for Chest X-ray Analysis, WACV'25
- Harini G., Aiman Farooq, and Deepak Mishra, Leveraging Auxiliary Classification for Rib Fracture Segmentation, ICVGIP 2024
- Aiman Farooq, Utkarsh Sharma, and Deepak Mishra, Enhanced Survival Prediction in Head and Neck Cancer Using Convolutional Block Attention and Multimodal Data Fusion, ACCV 2024 Workshop : WiCV(full length paper).
- Aiman Farooq, Deepak Mishra, and Santanu Chaudhury, Survival Prediction in Lung Cancer through Multi-modal Representation Learning, (early accept) WACV'25
- Azad Singh, Vandan Gorade, and Deepak Mishra, MLVICX: Multi-Level Variance-Covariance Exploration for Chest X-ray Self-Supervised Representation Learning, IEEE JBHI'24
- Azad Singh and Deepak Mishra, IDQCE: Instance Discrimination Learning through Quantized Contextual Embeddings for Medical Images, ICPR'24
- Azad Singh and Deepak Mishra, CoBooM: Codebook Guided Bootstrapping for Medical Image Representation Learning, MICCAI'24
- Nirbhay Sharma, Mayank Raj, and Deepak Mishra, Aggregation-Assisted Proxyless Distillation: A Novel Approach for Handling System Heterogeneity in Federated Learning, IJCNN'24
- Aiman Farooq, Deepak Mishra and Santanu Chaudhury, Translating Imaging to Genomics: Leveraging Transformers for Predictive Modeling, CVPR 2024 Workshop : WiCV
- Vandan Gorade, Azad Singh and Deepak Mishra, Large Scale Time-Series Representation Learning via Simultaneous Low- and High-Frequency Feature Bootstrapping. IEEE TNNLS'23
Announcements
- Aiman Farooq is a part of the MICCAI Supporters Outreach program at MICCAI 2025.
- Aiman Farooq receives the MICCAI Registration and Travel grant to attend MICCAI 2025 at Daejeon.
- Azad and Aiman receive the WACV DEI grant to attend WACV2025 in Tucson, USA.
- Aiman receives the IndoML travel grant to present her work at the Graduate Forum IndoML2024.
- Azad Singh receives two prestigious grants to attend MICCAI 2024 in Marrakesh, Morocco.
- MICCAI RISE Registration Grant
- Anusandhan National Research Foundation (ANRF) Grant.
- Ishan Mishra, is visiting the University of Buffalo (SUNY Buffalo) as research intern
- Aiman receives the CVPR-DEI and WiCV grant to attend CVPR2024 in Seattle.
- Our CT fracture rib characterization work gets featured in a prestigious Radiology blog.
Our Research Focus
- Self-Supervised Learning for Medical Imaging: Developing techniques to leverage large volumes of unlabeled radiographs. This approach aims to improve diagnostic models with limited labeled data, enhancing the efficiency of medical image analysis.
- AI for Radiogenomics: Integrating multi-modal data, including medical images and electronic health records, for cancer diagnosis. Our work focuses on predicting critical genomic mutations to guide personalized treatment plans.
- Federated Learning in Medical Imaging: Addressing data privacy and heterogeneity challenges across healthcare institutions. We explore distributed learning techniques to enable collaborative model training while ensuring patient data remains locally secured.
- Semi-Automatic Echo Image Acquisition: Developing AI-driven robotic systems for standard echocardiographic view acquisition. This project aims to reduce physical strain on sonographers and improve image consistency in cardiovascular diagnostics.
- 3D Medical Image Analysis: Creating deep learning models for tasks such as ribcage implant reconstruction from CT scans. Our goal is to automate and improve the accuracy of complex 3D medical image interpretation tasks.
Sponsored Projects
- AI-driven Robot-assisted Cardiac Ultrasound System to Acquire Clinically Useful Standard Echocardiographic Views
- Evaluation and development of Machine Learning (ML) models for the automated detection, localisation and characterisation of traumatic rib fractures on CT scans
- AI Driven ASIC for Real-time Medical Imaging Applications
- Resource Constrained AI