Cardiovascular Diseases

Cardiovascular disease remains the leading cause of mortality worldwide, yet clinical risk prediction still relies on static scoring systems designed decades ago. These tools capture a snapshot at a single moment, ignoring how patient physiology evolves, sometimes gradually, sometimes within hours. Our models track patients across the full care trajectory: from early warning signs detectable in primary care, through emergency presentation and triage, intensive monitoring in the ICU or coronary care unit, and to post-discharge follow-up. Rather than producing a single risk score, our approaches generate dynamic predictions that update as new information arrives and explain which clinical factors are driving risk at each stage. This matters because the features that predict deterioration change over time, and what signals danger in the first hour of an ICU stay differs from what matters at the twenty-fourth hour. By making these temporal patterns transparent, we aim to support earlier intervention and more informed clinical decision-making across acute cardiovascular conditions, including heart failure and myocardial infarction.