A proposed machine learning model may outperform other approaches at predicting breast cancer with higher accuracy, according to an analysis published in Mathematical Biosciences and Engineering.

Researchers sought to develop a deep extreme gradient descent optimization model to detect breast cancer. The proposed model includes a training stage (data acquisition) and a validation stage (data processing).

The investigators used the Wisconsin Breast Cancer Diagnosis dataset to assess the model and found it demonstrated robust results, with 98.7% accuracy, 99.6% specificity, 99.4% sensitivity, and 99.4% precision.

Reference: Khan MBS, Rahman AU, Nawaz MS, Ahmed R, Khan MA, Mosavi A. Intelligent breast cancer diagnostic system empowered by deep extreme gradient descent optimization. Math Biosci Eng. 2022;19(8):7978-8002. doi:10.3934/mbe.2022373

Link: https://pubmed.ncbi.nlm.nih.gov/35801453/