Applied Deep Learning for Biomedical Imaging

Course Description

Covers basic concepts, methodology and algorithms in deep learning and their applications to solve various biomedical imaging problems. Introduction to neural networks and their application to supervised and unsupervised learning problems formulated for biomedical imaging will be provided. Connections between general learning methodologies and specific challenges in the field of biomedical imaging, and design, implementation and evaluation of deep neural network-based solutions to imaging problems will be emphasized. Problems covered will include imaging system design and optimization, image recovery and reconstruction (built on the imaging physics and system course – BIOE 483), image processing (e.g., denoising, super-resolution and enhancement) and image analysis (e.g., same-contrast, multi-contrast and multimodal image registration, segmentation, classification and quality assessment). Biomedical application specific problems and solutions will be covered via hands-on problems and team-based projects. 3 undergraduate hours. 4 graduate hours. Prerequisite: MATH 241 or equivalent; BIOE 210, MATH 415 or equivalent; BIOE 310, ECE 310, STAT 410 or equivalent; BIOE 198, CS 101 or equivalent; BIOE 483; BIOE 485; or consent of the instructor.

Code

BIOE 486

Credits

3-4

Subject

Bioengineering

Programs That Include This Course

• Bachelor's Bachelor of Science in Bioengineering
• Bachelor's Bachelor of Science in Computer Science + Bioengineering
• Bachelor's Bachelor of Science in Neural Engineering
• Master's Master of Science in Biomedical Image Computing