Using Artificial Intelligence to Automatically Extract and Process Clinical Data from Electronic Medical Records in Cardiology Patients

What this Trial Is About

Researchers are evaluating the usefulness of artificial intelligence AI and machine learning to develop computer algorithms that can quickly and accurately extract and process large amounts of clinical data from electronic medical records. This study focuses on hospitalized cardiology patients in Greece and aims to improve automated data analysis, early disease diagnosis, and the development of clinical decision support systems. The study also plans to develop prognostic models for major cardiovascular diseases using AI methods. The study will retrospectively collect electronically registered clinical notes, laboratory results, and imaging exams from about 60,000 patients hospitalized in cardiology wards. Personal identifying information will be removed to protect privacy. Initially, clinical notes will be manually analyzed to create a database and identify keywords for diagnoses. Then, AI techniques, including natural language processing, will be used to automatically extract data from the remaining notes. The accuracy and reliability of these automated methods will be compared with manual extraction for evaluation. Participants data will be analyzed over time, with primary outcomes measuring the accuracy of AI-based data extraction compared to manual methods within one year. Secondary outcomes include tracking times to events such as death, major cardiovascular disease incidents, rehospitalization, stroke, and acute coronary syndrome over up to eight years from hospital discharge. The study aims to improve how clinical data is processed and used for patient care and research without interfering with usual medical treatment.

Artificial Intelligence for Automated Clinical Data Exploration From Electronic Medical Records (CardioMining-AI)