CAP 5516 – Medical Image Computing (Spring 2022)

alt text 

Dr. Chen Chen
(Please put [CAP 5516] in the subject line of your emails)
Time: 1:30 - 2:45 PM, Tuesday and Thursday
Location: Remote instruction [Zoom]
Office hours: Tuesday 10:00 - 11:00 AM and by appointment [Zoom]

Course Description

Imaging science is experiencing tremendous growth in the US. Biomedical imaging and its analysis are fundamental to understanding, visualizing, and quantifying medical images in clinical applications. With the help of automated and quantitative image analysis techniques, disease diagnosis will be easier/faster and more accurate, and leading to significant development in medicine in general. This course provides students with the foundation necessary for understanding, visualizing, and quantifying medical images with computational methods. In this course, we will examine some central topics and key techniques in computer vision and medical image processing, in particular employing Deep Learning, through reading, writing reviews on, presenting, discussing the most recent papers published on computer vision and medical imaging conferences (e.g., CVPR, ICCV, ECCV, MICCAI) as well as working on course projects.

The goal of the course is to give students the background and skills for graduate research in medical image computing. Through the class, the students are expected to understand in-depth the state-of-the-art approaches to various topics. By the end of this course, the students will also develop the skills that are vital to their graduate research, such as writing paper reviews, presenting technical papers, analyzing the strengths and weaknesses of the research papers, and identifying open questions and directions for future research.

Course Syllabus

  • Syllabus
  • Course Schedule

    Lecture 1 Course introduction [PDF] [Video]
    Lecture 2 How to review research papers and project ideas [PDF] [Video]
    Lecture 3 Introduction to medical imaging (part 1) [PDF] [Video]
    Lecture 4 Introduction to medical imaging (part 2) [PDF] [Video]
    Lecture 5 Introduction to deep learning (1) [PDF] [Video]
    Lecture 6 Introduction to deep learning (2) [PDF] [Video]
    Lecture 7 Introduction to deep learning (3) [PDF] [Video]
    Lecture 8 Introduction to deep learning (4) [PDF] [Video]
    Lecture 9 Medical image segmentation [PDF] [Video]
    Lecture 10 Generative Adversarial Networks (GAN) [PDF] [Video]
    Lecture 11 Self-supervised learning [PDF] [Video]
    Lecture 12 Adversarial robustness in deep learning [PDF] [Video]
    Lecture 13 Federated learning and its application in MIC [PDF] [Video]
    Lecture 14 Deep learning model efficiency [PDF] [Video]
    Lecture 15 Summary [PDF] [Video]
    Paper Presentation Paper 1: "Beyond COVID-19 Diagnosis: Prognosis with Hierarchical Graph Representation Learning"

    Paper 2: "Big Self-Supervised Models Advance Medical Image Classification"

    Paper Presentation Paper 3: "Data augmentation using learned transformations for one-shot medical image segmentation"

    Paper 4: "Group Shift Pointwise Convolution for Volumetric Medical Image Segmentation"

    Paper Presentation Paper 5: "Towards Robust General Medical Image Segmentation"

    Paper 6: "GLoRIA: A Multimodal Global-Local Representation Learning Framework for Label-efficient Medical Image Recognition"

    Paper Presentation Paper 7: "Annotation-Efficient Cell Counting"

    Paper 8: "3D Transformer-GAN for High-Quality PET Reconstruction"

    Paper Presentation Paper 9: "Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysis"

    Paper 10: "Efficient Medical Image Segmentation Based on Knowledge Distillation"

    Paper Presentation Paper 11: "Visual-Textual Attentive Semantic Consistency for Medical Report Generation" [Video]
    Paper Presentation Paper 13: "DeepOpht: Medical Report Generation for Retinal Images via Deep Models and Visual Explanation"

    Paper 14: "Unified 2D and 3D Pre-training for Medical Image Classification and Segmentation"

    Paper Presentation Paper 15: "Stabilized Medical Image Attacks"

    Paper 16: "T-AutoML: Automated Machine Learning for Lesion Segmentation using Transformers in 3D Medical Imaging"

    Paper Presentation Paper 17: "Medical Aegis: Robust adversarial protectors for medical images"

    Paper 18: "Multi-institutional Collaborations for Improving Deep Learning-based Magnetic Resonance Image Reconstruction Using Federated Learning"

    Paper Presentation Paper 19: "The Way to my Heart is through Contrastive Learning: Remote Photoplethysmography from Unlabelled Video"

    Paper 20: "BioFors: A Large Biomedical Image Forensics Dataset"

    Final Project Presentation Project Presentation (1) [Video]
    Final Project Presentation Project Presentation (2) [Video]
    Final Project Presentation Project Presentation (3) [Video]
    Final Project Presentation Project Presentation (4) [Video]