Dr. Mohsin Furkh Dar
Dr. Mohsin Furkh Dar is currently serving as an Assistant Professor in the School of Computer Science at UPES Dehradun. His academic and research work focuses on deep learning, medical image analysis, and the design of efficient and generalizable neural architectures. His doctoral contributions include EfficientU-Net, UMA-Net, fuzzy–rough set based loss functions, and SGAN frameworks for medical image segmentation and classification.
Citations by Year
Research Focus
About Me
Bridging AI Innovation with Clinical Impact
Dr. Mohsin Furkh Dar is currently serving as an Assistant Professor in the School of Computer Science at UPES Dehradun. His academic and research work focuses on deep learning, medical image analysis, and the design of efficient and generalizable neural architectures. His doctoral contributions include EfficientU-Net, UMA-Net, fuzzy–rough set based loss functions, and SGAN frameworks for medical image segmentation and classification. He has published in SCIE-indexed journals and presented at international conferences.
In addition to his research pursuits, Dr. Dar has significant teaching experience at both undergraduate and postgraduate levels, delivering courses in computer science, artificial intelligence, and data-driven technologies with a strong emphasis on hands-on learning. He combines theoretical model development with practical system implementation and has built AI-driven applications such as ShifaAI. He also supervises student projects in computer vision, deep learning, and applied machine learning.
Deep Learning
Advanced neural architectures for medical image analysis
Medical Imaging
Multi-modal imaging: CT, MRI, X-ray, ultrasound
Clinical Collaboration
Translating research into real-world medical applications
Research Leadership
Mentoring students and leading interdisciplinary projects
Featured Research
EfficientU-Net: A Novel Deep Learning Method for Breast Tumor Segmentation and Classification in Ultrasound Images
A lightweight encoder-decoder architecture achieving state-of-the-art performance in breast cancer detection from ultrasound images. Features efficient skip connections and optimized feature extraction pathways.
Deep Learning and Genetic Algorithm-Based Ensemble Model for Feature Selection and Classification of Breast Ultrasound Images
An evolutionary feature selection approach combined with deep learning ensemble models for improved breast ultrasound classification. Demonstrates superior accuracy through optimized feature subsets.
Adaptive Ensemble Loss and Multi-Scale Attention in Breast Ultrasound Segmentation with UMA-Net
A novel U-Net Multi-scale Attention architecture with adaptive ensemble loss functions for precise medical image segmentation. Addresses class imbalance through dynamic weight adjustment mechanisms.
Let's Collaborate
Interested in my research or exploring collaboration opportunities? I'd love to hear from you.