<![CDATA[Cardiovascular diseases (CVDs) remain the leading cause of morbidity and mortality worldwide, including in Indonesia, largely due to diagnostic challenges such as overlapping symptoms, inaccurate interpretation of test results, and delayed detection linked to socioeconomic limitations and invasive diagnostic methods. To address these issues, this study aims to develop PrediCard, a web-based diagnostic and screening tool integrating Artificial Intelligence (AI) through machine learning algorithms. The research employed a narrative review design, gathering data from databases such as PubMed, Google Scholar, and ResearchGate, followed by descriptive and critical appraisal analysis to design a low-fidelity prototype of the website. The conceptual framework utilizes both pre-existing clinical data from official cardiology guidelines and real patient data input. Algorithms such as decision trees, K-nearest neighbors (KNN), and K-means clustering process these data to produce diagnostic predictions and personalized lifestyle recommendations. A strengths, weaknesses, opportunities, threats (SWOT) analysis revealed that this innovation exerts significant potential for enhancing diagnostic accuracy, promoting early detection, and expanding screening accessibility, though challenges remain regarding internet access and the need for clinical validation. The findings demonstrate that integrating AI into digital diagnostic tools like PrediCard can support medical professionals in confirming diagnoses while enabling high-risk individuals to conduct self-screening efficiently. Ultimately, this innovation is expected to reduce CVD morbidity and mortality rates, improve patient outcomes, and advance progress toward Sustainable Development Goals (SDGs) in health. Further clinical trials and government support for internet infrastructure are required for practical implementation.]]>
