Assistants: Prithul Sarker
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
We will not use any text in this course; most of the material will be drawn from research papers.
Datasets
Dataset-related Papers
Useful Videos
Interesting Stories (quick reads)
AI in Breast (and Medical) Imaging
Radiomics (i.e., extraction of quantitative features from medical images)
Useful Resources
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
There will be no exams in this course. Grading will be based on paper presentations, class participation, and a team project.
Handouts
Schedule (tentative)
8/28 (1st week) Course overview (Bebis and Tavakkoli)
8/30 Introduction to breast cancer and mammography (Bebis)
9/4 (2nd week) Labor Day (no classes)
9/6 Introduction to breast cancer and mammography (cont'd) (Bebis)
9/11 (3rd week) Introduction to Deep Learning (Tavakkoli)
9/11 Deadline to finalize project teams (enter info on shared Google doc).
9/13 Introduction to Deep Learning (cont'd) (Tavakkoli)
9/18 (4th week) Breast Cancer Detection in Mammography - Overview (Bebis)
9/18 Deadline to select first paper for presentation (enter info on shared Google doc)
Subsequent papers should be selected within two weeks from your previous paper presentation.
9/20 Breast Cancer Detection in Mammography - Review (cont'd) (Bebis)
9/25 (5th week) Deep Learning Architectures (Tavakkoli)
Amit's recorded presentation on Deep Learning Platforms from Spring 2022 can be found here
9/27 Project Ideas (Bebis)
10/2 (6th week) Deep Learning Architectures (cont'd) (Tavakkoli)
10/4 SLIDES An integrated framework for breast mass classification and diagnosis using stacked ensemble of residual neural networks (P1 Team #1)
10/4 SLIDES Two-View Mammogram Synthesis from Single-View Data Using Generative Adversarial Networks (P1 Team #4)
10/4 Deadline for project topic selection (1st deliverable - upload on Canvas by 11:59pm)
10/9 (7th week) SLIDES Breast Cancer in Mammography Classification with local contour features (P1 Team #2)
10/9 (7th week) SLIDES Transformer-based mass detection in digital mammograms (P1 Team #6)
10/11 SLIDES Breast Lesions Detection and Classification via YOLO-Based Fusion Models (P1 Team #3)
10/11 SLIDES A deep learning method for classifying mammographic breast density categories (P1 Team #7)
10/16 (8th week) ISVC'23 (no class)
10/16 Deadline for project proposal (2nd deliverable - upload on Canvas by 11:59pm)
10/18 ISVC'23 (no class)
10/23 (9th week) Project Proposal Presentations (15 minutes each)
Team #1
Team #4
Team #2
Team #6
10/25 Project Proposal Presentations (cont'd)
Team #3
Team #7
Team #5
10/30 (10th week) SLIDES A review of various modalities in breast imaging technical aspects and clinical outcomes (P1 Team #5)
11/1 SLIDES A fully integrated computer-aided diagnosis system for digital X-ray mammograms via deep learning detection, segmentation, and classification (P2 Team #1)
11/1 SLIDES Multi view breast cancer classifications via hyper complex neural network (P2 Team #4)
11/6 (11th week) SLIDES Exploiting Patch Sizes and Resolutions for Multi-Scale Deep Learning in Mammogram Image Classification (P2 Team #2)
11/6 (11th week) SLIDES Rethinking pre-training on medical imaging (P2 Team #6)
11/8 SLIDES 2023 - Convolutional Feature Descriptor Selection for Mammogram Classification (P2 Team #3)
11/8 SLIDES 2022 - Deep Multiple Instance Learning for Automatic Breast Cancer Assessment Using Digital Mammography (P2 Team #7)
11/11 Deadline for interim project report (3rd deliverable - upload on Canvas by 11:59pm)
11/13 (12th week) Interim Project Presentations (20 minutes each)
Team #1
Team #4
Team #2
11/15 Interim Project Presentations (cont'd)
Team #6
Team #3
Team #7
11/20 (13th week) Interim Project Presentations (cont'd)
Team #5
11/22 SLIDES 2022 - Early detection and classification of abnormality in prior mammograms using image-to-image translation and YOLO techniques (P2 Team #5)
11/27 (14th week) SLIDES 2021 - Boundary loss for highly unbalanced segmentation (P3 Team #1)
11/27 (14th week) SLIDES 2019 - High-Resolution Mammogram Synthesis using Progressive Generative Adversarial Networks (P3 Team #4)
11/29 SLIDES 2021 - A deep learning model using data augmentation for detection of architectural distortion in whole and patches of images (P3 Team #2)
12/4 (15th week) SLIDES 2021 - An interpretable classifier for high-resolution breast cancer screening images utilizing weakly supervised localization (P3 Team #3)
12/4 (15th week) SLIDES 2019 - Feature extraction based on empirical mode decomposition for automatic mass classification of mammogram images (P3 Team #7)
12/6 SLIDES 2022 - Ultra-high resolution, multi-scale, context-aware approach for detection of small cancers on mammography (P3 Team #5)
12/6 SLIDES 2023 - Pre-training in Medical Data A Survey (P3 Team #6)
12/11 (16th week) Placeholder (work on your projects)
12/13 Prep Day
12/18 Final Project Presentations (30 minutes) (3:00pm - 5:00pm)
Team #1
Team #4
Team #2
Team #6
12/20 (tentative) Final Project Presentations (cont'd) (3:00pm - 5:00pm)
Team #3
Team #7
Team #5
12/22 Deadline for final project report (4th deliverable - upload on Canvas by 11:59pm)
Papers for Presentation
Review Papers (REV)
Machine Learning (ML)
Deep Learning (DL)
2016 - A deep feature based framework for breast masses classification
2016 - Representation learning for mammography mass lesion classification with convolutiona lneural networks
2016 - The Automated Learning of Deep Features for Breast Mass Classiffication from Mammograms
2016 - Transferring Learned Microcalcification Group Detection from 2D Mammography to 3D Digital Breast Tomosynthesis Using a Hierarchical
2016 -A_multi-view_deep_learning_architecture_for_classification_of_breast_microcalcifications
2016 - Deep Multi-instance Networks with Sparse Label Assignment for Whole Mammogram Classification
2017 - Super-Resolution Imaging of Mammograms Based on the Super-Resolution Convolutional Neural Network
2017 - Breast Density Classification with Deep Convolutional Neural Networks
2017 - Deep learning in breast cancer risk assessment evaluation of convolutional neural networks on a clinical dataset of full-field digital mammograms
2017 - Detecting and classifying lesions in mammograms with Deep Learning
2017 - Automated_Analysis_of_Unregistered_Multi-View_Mammograms_With_Deep_Learning
2017 - A novel and fully automated mammographic texture analysis for risk prediction results from two case-control studies
2018 - Deep Convolutional Neural Networks for breast cancer screening
2018 - Detecting and classifying lesions in mammograms with Deep Learning
2018 - Detection of masses in mammograms using a one-stage object detector based on a deep convolutional neural network (end-to-end, previously presented)
2018 - Detecting and classifying lesions in mammograms with Deep Learning
2018 - Conditional Infilling GANs for Data Augmentation in Mammogram Classification previously presented)
2018 - A fully integrated computer-aided diagnosis system for digital X-ray mammograms via deep learning detection, segmentation, and classification
2018 - Simultaneous detection and classification of breast masses in digital mammograms via a deep learning YOLO-based CAD system
2018 - A context-sensitive deep learning approach for microcalcification detection in mammograms
2018 - High-Resolution Breast Cancer Screening with Multi-View Deep Convolutional Neural Networks
2018 - Adversarial_deep_structured_nets_for_mass_segmentation_from_mammograms (interesting, previously presented)
2018 - Attention U-Net Learning Where to Look for the Pancreas (3D Unet, previously presented)
2018 - Y-Net Joint Segmentation and Classification for Diagnosis of Breast Biopsy Images
2018 - Applying Data-driven Imaging Biomarker in Mammography for Breast Cancer Screening Preliminary Study
2018 - Deep Autoencoding Models for UnsupervisedAnomaly Segmentation in Brain MR Images
2018 - Abnormality Detection in Mammography using Deep Convolutional Neural Networks
2018 - A deep learning method for classifying mammographic breast density categories
2018 - Improving Breast Cancer Detection using Symmetry Information with Deep Learning
2019 - Multi-View_Convolutional_Neural_Networks_for_Mammographic_Image_Classification
2019 - Deep Learning for Breast Cancer Diagnosis from Mammograms: A Comparative Study
2019 - Automatic mass detection in mammograms using deep convolutional neural networks (interesting, previously presented)
2019 - Determination of mammographic breast density using a deep convolutional neural network
2019 - Breast Microcalcification Diagnosis Using Deep Convolutional Neural Network from Digital Mammograms
2019 - Deep learning modeling using normal mammograms for predicting breast cancer risk
2019 - Breast cancer detection using synthetic mammograms from generative adversarial networks in convolutional neural networks (interesting, previously presented)
2019 - Deep Learning to Improve Breast Cancer Detection on Screening Mammography
2019 - Cross-view Relation Networks for Mammogram Mass Detection
2019 - Injecting and removing suspicious features in breast imaging with CycleGAN
2019 - High-Resolution Mammogram Synthesis using Progressive Generative Adversarial Networks
2019 - BREAST DENSITY QUANTIFICATION USING WEAKLY ANNOTATED DATASET
2020 - Joint 2D-3D Breast Cancer Classification
2020 - Deep Learning Computer-Aided Diagnosis for Breast Lesion in Digital Mammogram
2020 - Deep_Neural_Networks_Improve_Radiologists_Performance_in_Breast_Cancer_Screening (interesting, previously presented)
2020 - Deep_Neural_Networks_With_Region-Based_Pooling_Structures_for_Mammographic_Image_Classification
2020 - Classification of breast mass in two-view mammograms via deep learning
2020 - Detecting Asymmetric Patterns and Localizing Cancers on Mammograms
2020 - Segmentation of Masses onMammograms Using Data Augmentation and Deep Learning
2020 - Using Convolutional Neural Network with Cheat Sheet and Data Augmentation to Detect Breast Cancer in Mammograms
2020 - Medical Image Synthesis via Deep Learning (interesting, previously presented)
2020 - Deep Learning Computer-Aided Diagnosis for Breast Lesion in Digital Mammogram
2020 - AUNet attention-guided dense-upsampling networks for breast mass segmentation in whole mammograms (whole mammogram segmentation, previously presented)
2020 - Deep_Neural_Networks_Improve_Radiologists_Performance_in_Breast_Cancer_Screening
2020 - Joint 2D-3D Breast Cancer Classification
2020 - Cross-View_Attention_Network_for_Breast_Cancer_Screening_from_Multi-View_Mammograms (interesting, previously presented)
2020 - Deeply supervised U-Net for mass segmentation in digital mammograms (region+bounary loss, CRF postprocessing, previously presented)
2020 - MultiResUNet Rethinking the U-Net architecture for multimodal biomedical image segmentation (interesting, previously presented)
2020 - Synthesizing lesions using contextual GANs improves breast cancer classification on mammograms (previously presented)
2020 - Evaluation of data augmentation via synthetic images for improved breast mass detection on mammograms using deep learning (previously preented)
2020 - BR-GAN Bilateral Residual Generating Adversarial Network for Mammogram Classification (cycleGAN, previously presented)
2020 - Prediction of breast cancer molecular subtypes using radiomics signatures of synthetic mammography from digital breast tomosynthesis
2020 - Semantic Label Prediction of Mammography Based on CC and MLO Views
2020 - SAP-cGAN Adversarial learning for breast mass segmentation in digital mammogram based on superpixel average pooling
2020 - A multi-context CNN ensemble for small lesion detection
2020 - Detecting Asymmetric Patterns and Localizing Cancers on Mammograms
2020 - Deep Learning Pre-training Strategy for Mammogram Image Classification an Evaluation Study
2020 - New convolutional neural network model for screening and diagnosis of mammograms
2020 - A Deep Learning Approach for Efficient Registration of Dual View Mammography
2020 - Neural networks model based on an automated multi-scale method for mammogram classification
2021 - A framework for breast cancer classification using Multi-DCNNs
2021 - Connected-UNets a deep learning architecture for breast mass segmentation (previously presented)
2021 - A deep learning model using data augmentation for detection of architectural distortion in whole and patches of images
2021 - A computer-aided diagnostic system for mammograms based on YOLOv3
2021 - Multi-View_Mammographic_Density_Classification_by_Dilated_and_Attention-Guided_Residual_Learning (interesting, previously presented)
2021 - Act_Like_a_Radiologist_Towards_Reliable_Multi-view_Correspondence_Reasoning_for_Mammogram_Mass_Detection (graph cnn, multi-views, previously presented)
2021 - MommiNet-v2 Mammographic multi-view mass identification networks (nice paper, previously presented)
2021 - Evaluation of U-net-based Image Segmentation Model to Digital Mammography
2021 - An interpretable classifier for high-resolution breast cancer screening images utilizing weakly supervised localization
2021 - Breast Lesions Detection and Classification via YOLO-Based Fusion Models
2021 - Weakly and semi supervised detection in medical imaging via deep dual branch net
2021 - Rethinking pre-training on medical imaging
2021 - Boundary loss for highly unbalanced segmentation
2022 - ARF-Net An Adaptive Receptive Field Network for breast mass segmentation in whole mammograms and ultrasound images (small mass segmentation, multi-scale attention, previously presented)
2022 - Auto-DenseUNet Searchable Neural Network Architecture for Mass Segmentation in 3D Automated Breast Ultrasound
2022 - YOLO-LOGO A transformer-based YOLO segmentation model for breast mass detection and segmentation in digital mammograms (previously presented)
2022 - BTS-GAN Computer-aided segmentation system for breast tumor using MRI and conditional adversarial networks (interesting, previously presented)
2022 - Breast Cancer Diagnosis in Two-View Mammography Using End-to-End Trained EfficientNet-Based Convolutional Network patch -> whole, previously presented)
2022 - An integrated framework for breast mass classification and diagnosis using stacked ensemble of residual neural networks
2022 - A generative adversarial network for synthetization of regions of interest based on digital mammograms
2022 - Visual and quantitative evaluation of microcalcifications in mammograms with deep learning-based super-resolution
2022 - Multi-View Breast Cancer Classification via Hypercomplex Neural Networks
2022 - Breast_Cancer_Diagnosis_in_Two-View_Mammography_Using_End-to-End_Trained_EfficientNet-Based_Convolutional_Network
2022 - Identifying_Women_With_Mammographically-_Occult_Breast_Cancer_Leveraging_GAN-Simulated_Mammograms
2022 - Early detection and classification of abnormality in prior mammograms using image-to-image translation and YOLO techniques
2022 - Transformers Improve Breast Cancer Diagnosis from Unregistered Multi-View Mammograms
2022 - A generative adversarial network for synthetization of regions of interest based on digital mammograms
2022 - Ultra-high resolution, multi-scale, context-aware approach for detection of small cancers on mammography
2022 - Deep Multiple Instance Learning for Automatic Breast Cancer Assessment Using Digital Mammography
2022 - Two-View Mammogram Synthesis from Single-View Data Using Generative Adversarial Networks
2023 - High-resolution synthesis of high-density breast mammograms Application to improved fairness in deep learning based mass detection
2023 - Transformer-based mass detection in digital mammograms
2023 - Breast Density Transformations Using CycleGANs for Revealing Undetected Findings in Mammograms
2023 - Unsupervised anomaly detection with generative adversarial networks in mammography
2023 - Exploiting Patch Sizes and Resolutions for Multi-Scale Deep Learning in Mammogram Image Classification
2023 - Convolutional Feature Descriptor Selection for Mammogram Classification
2023 - Deep learning performance for detection and classification of microcalcifications on mammography
2023 - M&M Tackling False Positives in Mammography with a Multi-view and Multi-instance Learning Sparse Detector
Breast Cancer Risk
Survival
Other (O)
Different Modalities (DM)
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.
Computer Science & Engineering Department
