A Multi-Scale CNN and Curriculum Learning Strategy for Mammogram\n Classification

2017/07/21 by William Lotter, Lotter, William, Greg Sorensen +3 · 1 citation
Computer Science · Medicine · #AI in cancer detection #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Radiomics and Machine Learning in Medical Imaging

paper · pdf · doi:10.48550/arxiv.1707.06978

openalex publication_date 2017/07/21 · openalex created_date 2022/09/28 · openalex updated_date 2026/07/28

Abstract

Screening mammography is an important front-line tool for the early detection\nof breast cancer, and some 39 million exams are conducted each year in the\nUnited States alone. Here, we describe a multi-scale convolutional neural\nnetwork (CNN) trained with a curriculum learning strategy that achieves high\nlevels of accuracy in classifying mammograms. Specifically, we first train\nCNN-based patch classifiers on segmentation masks of lesions in mammograms, and\nthen use the learned features to initialize a scanning-based model that renders\na decision on the whole image, trained end-to-end on outcome data. We\ndemonstrate that our approach effectively handles the "needle in a haystack"\nnature of full-image mammogram classification, achieving 0.92 AUROC on the DDSM\ndataset.\n

Citations

• A Threshold Selection Method from Gray-Level Histograms
• National Performance Benchmarks for Modern Screening Digital Mammography: Update from the Breast Cancer Surveillance Consortium
• Rethinking the Inception Architecture for Computer Vision

Cited by

• CURVETE: Curriculum Learning and Progressive Self-supervised Training for Medical Image Classification

Related

• A Two-Stage Multiple Instance Learning Framework for the Detection of Breast Cancer in Mammograms
• Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening
• Deep Curriculum Learning in Task Space for Multi-Class Based Mammography Diagnosis
• Abnormality Detection in Mammography using Deep Convolutional Neural Networks
• MAMMO: A Deep Learning Solution for Facilitating Radiologist-Machine Collaboration in Breast Cancer Diagnosis