Date of Publication :7th February 2026

Abstract: The screening for risk of cardiovascular diseases still continues to be a concern owing to the progressive nature of cardiovascular diseases, as it requires a time-consuming process that relies on clinical assessments. This paper proposes a machine-learning approach that combines natural language processing algorithms for automatically predicting risk for cardiovascular diseases based on unstructured clinical narratives as well as patient histories that span a long period of time. The experimental analysis has been carried out by using established health care benchmarks for convolutional, recurrent, as well as transformer models. The experimental results validate that the use of transformer models for natural language processing has better results than the traditional approach, as recorded by predictive accuracy as well as the F1 score.

Reference :