COMP 448/548: Medical Image Analysis
Spring 23
Course Description
Imaging has been increasingly used in medicine and biology research. Its primary use is to visualize the human body at different levels (e.g., at organ, tissue, cell, and molecular levels) using different imaging modalities (e.g., those in radiology, pathology, dermatology, ophthalmology, microscopy, and genetics). The explosive growth of this use necessitates to develop computational methods to (semi) automatically analyze images produced by these modalities.
The objective of this course is to provide students with the basic concepts and computational and mathematical methods in medical image analysis. Topics covered in this course include
- An overview of medical imaging modalities,
- An overview of medical image analysis applications and their challenges,
- Traditional image processing techniques for medical image analysis,
- Deep learning for medical image analysis, and
- Case studies.
Lectures: (Slides and reading material will be available)
TOPICS |
CONTENTS |
Introduction |
Introduction, an overview of imaging modalities, an overview of applications and challenges in medical image analysis |
Filters [ Filters ] |
Filters for medical image enhancement |
Medical image segmentation [ Part 1 | Part 2 | Texture analysis ] |
Region-based segmentation, edge-based segmentation, thresholding, clustering, texture analysis, distance transforms, mathematical morphology, segmentation evaluation, case study on cell segmentation in microscopy images |
Medical image representation |
Feature extraction for medical images, morphological features, textural features, structural features |
Medical image classification |
Basics of classifiers, basics of artificial neural networks, classifier evaluation |
Deep learning |
Basics of convolutional neural networks, convolutional neural networks for medical image classification |
Deep learning |
Encoder-decoder networks, semantic segmentation in biopsy images, tumor/organ segmentation in CT images, retinal layer segmentation in OCT images, multi-task networks for instance segmentation in microscopy images |
Deep learning |
Generative modes for medical image synthesis, reconstruction and artifact correction |
Deep learning [ Multiple-instance learning | |
Multiple-instance learning and self-supervised learning for medical image analysis |