Transparency in Computer-Aided Breast Cancer Diagnosis Tool Development: A CBIS-DDSM Case Study

Abstract Accessible mammography datasets and innovative machine learning techniques are at the front line for computer-aided breast cancer diagnosis. However, the opacity surrounding private datasets and the unclear methodology behind the selection of subset images from publicly available databases for model training and testing, coupled with the arbitrary incompleteness or inaccessibility of code, markedly intensifies the obstacles in replicating and validating the model's efficacy. These challenges, in turn, erect barriers for subsequent researchers striving to learn and advance in this field. To address these limitations, we provide a pilot codebase covering the entire image preprocessing to model construction and evaluation pipeline, utilizing publicly available Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM) mass subset, including both full images and regions of interests (ROIs). The pipeline achieves an area under the receiver operating characteristic curve (AUROC) of 86.3% [95% CI: 0.842, 0.885], representing the state of the art performance for benign and malignant classification with selected mammography images. Collectively, our efforts hold promise in accelerating global software development for breast cancer diagnosis by leveraging our codebase and structure, while also integrating other advancements in the field.