Computer Science & Engineering Department

CS791 Topics: Mass Detection in Mammograms (Fall 2025)

Prerequisites

Background in the following areas would be very useful: image processing, computer vision, pattern recognition, machine learning, and deep learning. Knowledge or desire to quickly learn Jupyter Notebook and Python is required. Familiarity with Keras/Tensorflow or Pytorch would be a plus.

Textbook

No textbook will be used in this course. Most materials will be drawn from research papers.

Description and Objectives

The course will focus on the problem of mass detection and classification in mammograms and possibly other modalities such as CT and MRI. The goal is to expose students to some of the main challenges involved in this research area and to recent methods developed by the research community to address these challenges. The course is primarily intended for highly motivated students who are interested in applying pattern recognition, machine learning and deep learning techniques to this research area.

Requirements

Handouts

• Syllabus
• Project Guidelines
• How to Read a Research Paper
• How to read and understand a scientific paper: a guide for non-scientists

Interesting Stories on AI and Mammography (quick reads)

• Calidar Announces Groundbreaking 4D Mammography System: First-in-Human Study Initiated (August 19, 2025)
• Machine learning model accurately predicts DCIS upstaging without invasive surgery (Jan 5, 2022)
• Artificial intelligence may reduce false positives in breast cancer screening (Oct 18, 2021)
• AI helps rule out cancer in dense breasts (Oct 5, 2021)
• Combination of Artificial Intelligence and Radiologists More Accurately Identified Breast Cancer (October 17, 2021)
• The Future is Bright for Collaboration Between AI and Breast Imagers (Sept 1, 2020)
• Digital Mammography (DM) DREAM Challenge (2016-2017)
• DREAM Challenge results: Can machine learning help improve accuracy in breast cancer screening? (2017)

DL Books

• Dive into Deep Learning by Aston Zhang, Zachary C. Lipton, Mu Li, Alexander J. Smola
• Deep Learning by Ian Goodfellow and Yoshua Bengio and Aaron Courville
• Additional books are available on Canvas

DL Courses

• Andrew Ng: Introduction to Deep Learning
• MIT 6.S191 Introduction to Deep Learning
• CS231n: Convolutional Neural Networks for Visual Recognition
• Analyses of Deep Learning (STATS 385)

DL Review Slides (Dr Tavakkoli)

• Introduction to Deep Learning
• Deep Learning Architectures

DL Tutorials

• PyTorch Tutorial 1
• PyTorch Tutorial 2
• PyTorch Tutorial 3
• Python Numpy Tutorial (with Jupyter and Colab)
• Keras with TensorFlow Course - Python Deep Learning and Neural Networks for Beginners Tutorial
• Python for Computer Vision
  Matlab Tutorial (Univ Utah)

DL Environments and Repositories

• KERAS
• Google Collaboratory
• Awesome Deep Vision
• OpenCV

DL Mammography-Related Environments and Repositories

• MONAI: Medical Open Network for Artificial Intelligence
• MediGan: a Python library of pretrained generative models for medical image synthesis
• Awesome GAN for Medical Imaging
• Medical Image Generation
• Awesome Medical Image Anomaly Detection
• CVPR-MIA
• MCGPU: GPU-accelerated Monte Carlo X-ray Imaging Simulator

DL Medical Imaging Related Courses

• Medical Image Computing
• Machine Learning for Healthcare

DL papers written with radiologists in mind

• 2017 - Deep Learning A Primer for Radiologists (nice CNNs intro)
• 2021 - Deep Learning An Update for Radiologists
• 2020 - Artificial Intelligence A Primer for Breast Imaging Radiologists (AI+radiology, risk prediction, tomosynthesis)

Schedule (tentative)

8/25 (1st week) Schedule conflict (no class)

8/27 Course overview
• Introduction to Mammography
• Mammographic Interpretation
• Introduction to Screening Mammography
• Screening Mammography Workflow (How to Read a Mammogram)
  
  
9/1 (2nd week) Labor Day (no classes)

9/3 Introduction to mammography
• 2016 - An_Overview_of_Mammogram_Analysis
• 2020 - A review of various modalities in breast imaging technical aspects and clinical outcomes
• 2018 - A review of computer aided detection in mammography
• 2017 - Why CAD Failed in Mammography
• 2020 - International evaluation of an AI system for breast cancer screening

9/8 (3rd week) Introduction to mammography (cont'd)

9/10 Introduction to mammography (cont'd)

9/15 (4th week) Breast Cancer Detection in Mammography: Challenges and Requirements

• 2018 - Artificial intelligence in radiology
• 2018 - Current Applications and Future Impact of Machine Learning in Radiology
• 2018 - Deep learning and artificial intelligence in radiology Current applications and future directions
• 2020 - Integrating artificial intelligence into the clinical practice of radiology challenges and recommendations
• 2022- Past, Present, and Future of Machine Learning and Artificial Intelligence for Breast Cancer Screening
• 2023 - Artificial Intelligence in Breast Imaging Opportunities, Challenges, and Legal-Ethical Considerations

9/17 Breast Cancer Detection in Mammography - Classical ML Methods

• 2006 - Approaches for automated detection and classification of masses in mammograms (emphasis on mass detection and classification)
• 2007 - A review of computer-aided diagnosis of breast cancer - Toward the detection of subtle signs (discusses microcalcificatons, asymmetries, and architectural distortions too)
• 2010 - A review of automatic mass detection and segmentation in mammographic images (discusses segmentation methods too)
• 2016 - Radiomics Images are More than Pictures They are Data
• 2017PyRadiomics - Computational Radiomics System to Decode the Radiographic Phenotype

9/22 (5th week) Breast Cancer Detection in Mammography - Classical ML Methods (cont'd)

9/24 Breast Cancer Detection in Mammography - Classical ML Methods (cont'd)

9/29 (6th week) Breast Cancer Detection in Mammography - Modern DL Methods

• 2019 - Deep convolutional neural networks for mammography advances, challenges and applications (nice review)
• 2020 - Survey of deep learning in breast cancer image analysis (focuses mostly on feature extraction; discusses histopathology)
• 2021 - Convolutional neural networks for breast cancer detection in mammography A survey (nice review; has a list of FDA approved CAD systems)
• 2021 - Breast Cancer Segmentation Methods Current Status and Future Potentials (nice review of past and current methods)
• 2025 - Deep Learning in Breast Cancer Imaging A Decade of Progress and Future Directions (reviews multiple modalities, not just mammography)

10/1 Breast Cancer Detection in Mammography - Modern DL Methods (cont'd)

10/6 (7th week) Breast Cancer Detection in Mammography - Modern DL Methods (cont'd)

10/8 Breast Cancer Detection in Mammography - Modern DL Methods (cont'd)

10/10 Papers for Presentation

10/13 (8th week) Breast Cancer Detection in Mammography - Modern DL Methods (cont'd)

10/15 Breast Cancer Detection in Mammography - Modern DL Methods (cont'd)

10/17 (Zoom, make-up for Oct 8th) Project Proposal Presentations (10min)

10/19 Upload proposal report on Canvas by 11:59PM

10/20 (9th week) Patrick (Risk Prediction P1)

10/22 Nikola (Weakly Supervised P1)

10/24 (Zoom, make-up for Nov 19th) Carter (VLM P1)

10/27 (10th week) Jinal (Transfer Learning P1)

10/29 Ryota (Multi-modal fusion P1)

11/3 (11th week) Nikola (Weakly Supervised P2)

11/5 Patrick (Risk Prediction P2)

11/10 (12th week) Carter (VLM P2)

11/12 Jinal (Transfer Learning P2)

11/17 (12th week) ISVC'25 (no class)
11/17 Upload interim report on Canvas by 11:59PM

11/19 ISVC'25 (no class)

11/21 (Zoom, make-up for Nov 19th) Project Interim Presentations (15min)

11/24 (13th week) Ryota (Multi-modal fusion P2)

11/26 Nikola (Weakly Supervised P3)

12/1 (14th week) Patrick (Risk Prediction P3)

12/3 Carter (VLM P3)

12/5 (Zoom, make-up for Aug 25th) Jinal (Transfer Learning P3)

12/8 (15th week) Ryota (Multi-modal fusion P3)

12/12 Upload final report on Canvas by 11:59PM

12/15 Project Final Presentations (20min)

Datasets

• Digital Database for Screening Mammography (DDSM)
• CBIS-DDSM (Curated Breast Imaging Subset of DDSM)
• Mammographic Image Analysis Society (MIAS)
• INbreast
• Breast Cancer Digital Repository (BCDR)
• VinDr-Mammo (large-scale benchmark dataset of full-field digital mammography)
• Mammogram Density Assessment Dataset
• EMory BrEast imaging Dataset (EMBED)
• Breast Micro-Calcification Dataset
• Dataset of Breast mammography images with Masses
• Optimam Mammography Image Database
• BancoWeb LAPIMO
• SuReMaPP (Suspicious Regions on Mammograms from Palermo Polyclinic)
• OPTIMAM mammography image database and viewing software (contains annotated 3D tomosynthesis images)
• Virtual Imaging Clinical Trial for Regulatory Evaluation (VICTRE - synthetic dataset)
• Database for Mastology Research with Infrared Image (DMR)
• Breast Cancer Histology Images (BACH)
• Breast Cancer Screening - Digital Breast Tomosynthesis (BCS-DBT); more info can be found here
• Cancer Imaging Archive (very extensive, lots of datasets, not just for breast)

Datasets-related Papers

• 2012 - INBreast: Toward a Full-field Digital Mammographic Database
• 2012 - BCDR A BREAST CANCER DIGITAL REPOSITORY
• 2017 - Data Descriptor: A curated mammography data set for use in computer-aided detection and diagnosis research
• 2018 - A Multi-million Mammography Image Dataset and Population-Based Screening Cohort for the Training and Evaluation of Deep Neural Networks - the Cohort of Screen-Age
• 2020 - Dataset of breast mammography images with masses
• 2020 - OPTIMAM Mammography Image Database: a large scale resource of mammography images and clinical data
• 2021 - A Data Set and Deep Learning Algorithm for the Detection of Masses and Architectural Distortions in Digital Breast Tomosynthesis Images
• 2022 - VinDr-Mammo: A large-scale benchmark dataset for computer-aided diagnosis in full-field digital mammography
• 2023 - The EMory BrEast imaging Dataset (EMBED) A Racially Diverse, Granular Dataset of 3.4 Million Screening and Diagnostic Mammographic Images
• 2023 - Mammography Datasets for Neural Networks Survey

Presentation Tips

1. Presentations should be professional as if it was presented in a formal conference (i.e., powerpoint slides).
2. Your goal is to educate and inform your audience. Make sure your presentation follows a logical sequence. Help the audience understand how successive definitions and results are related to each other and to the big picture.
3. You should have your remarks prepared and somewhat memorized. Reading from your notes excessively should be avoided.
4. Anticipate Questions: think of some likely questions and plan out your answer. Understand the Question: paraphrase it if necessary; repeat it if needed. Do Not Digress. Be Honest: if you can't answer the question, say so.
5. Meet the eyes of your audience from time to time.
6. Vary the tone of your voice and be careful to speak clearly and not talk too quickly.