Anish Raj

- Developed generalizable algorithms for medical image analysis using CNNs and attention mechanisms to achieve high accuracy with small datasets. - Delivered impactful results across multiple datasets, including 91+ Dice in kidney segmentation, 90+ AUC in aortic dissection detection, and a 4% improvement in Multiple Sclerosis change detection accuracy. - Developed and deployed practical algorithms for clinical use, including Docker-based implementation of aortic aneurysm and dissection detection workflows.

- Developed real-time road condition estimation models using CNNs by fusing road images and reflection maps, achieving 99% accuracy for each class, enhancing ADAS systems for driver safety. - Processed over 500,000 images for model training and validation. - Collaborated with cross-functional teams to integrate algorithm in ADAS systems.

Deep learning based image segmentation, classification and regression using MRI and CT data in small dataset regime

Focus on machine learning, image processing, microscopy and optics

Focus on medical imaging, programming and signal processing