Detecting Cardiovascular Disease from Mammograms with Deep Learning

Cardiovascular disease (CVD) remains a leading global cause of mortality, necessitating innovative, non-invasive methods for early detection to enable timely interventions. Breast arterial calcifications (BAC), visible in mammograms, have emerged as a promising biomarker for assessing CVD risk. This research investigates the application of deep learning, specifically convolutional neural networks (CNNs), to detect BAC in mammograms and stratify CVD risk. We propose a robust CNN-based model trained on a large, annotated mammogram dataset to identify calcifications indicative of cardiovascular risk. The model demonstrates high accuracy, precision, and recall, highlighting its potential to integrate seamlessly into routine mammography screening workflows, thereby enhancing early CVD detection in a cost-effective and scalable manner Key words Alzheimer's disease, dementia, deep learning, neuroimaging, biomarkers, convolutional neural networks (CNNs), recurrent neural networks (RNNs), Explainable AI (XAI), early diagnosis, disease progression modeling.