IEEE EMBS Lecture: Deep Learning for Medical Image Analysis

Speaker: Dr. Sandeep Singh Sengar, Senior Lecturer of Computer Vision and Artificial Intelligence at Cardiff Metropolitan University, United Kingdom 

Date : 18th February 2025   03:00 pm - 04:00 pm IST

Venue: Google Meet

No. of attendees: 39

Dr. Sengar elaborated on the dominance of CNNs in medical image processing, emphasizing their effectiveness in various applications. He particularly focused on the U-Net architecture, which is widely used in biomedical imaging due to its contraction and expansion pathways for precise pixel-wise segmentation.

One of the key limitations of traditional U-Net models is their reduced efficiency when applied to 3D voxel-based segmentation tasks. Addressing this challenge, Dr. Sengar introduced the concept of Multi-Planar U-Net, an improved variant of U-Net developed in 2019. This model enhances segmentation accuracy for 3D medical imaging, specifically in knee MRI segmentation. The talk provided insights into how Multi-Planar U-Net overcomes the shortcomings of standard U-Net models and enhances performance in volumetric medical imaging tasks.