A recent study led by the Cleveland Clinic and Dyania Health (USA) has demonstrated the potential of artificial intelligence (AI) specifically trained in medicine to analyze electronic medical records (EMR) and accurately select eligible patients for clinical trials of rare diseases. Published in 'The Journal of Cardiac Failure', the official journal of the Heart Failure Society of America, the work provides practical evidence that AI-assisted medical record review can optimize the speed, accuracy, and fairness in participant selection for clinical trials. The research evaluated the performance of an AI system developed by Dyania Health, implemented at the Cleveland Clinic, tasked with pre-selecting candidates for the DepleTTR-CM trial, focused on transthyretin amyloid cardiomyopathy (ATTR-CM), a form of heart failure prevalent in older adults. In just one week, the AI reviewed 1,476 medical histories and identified 46 potential candidates. Of the 30 patients selected by the AI and validated by doctors, none had been detected through traditional recruitment methods. The system achieved an accuracy of 96.2% in answering 7,700 specific trial questions across nine clinical areas. Thanks to AI-assisted evaluation, seven patients were enrolled before the trial's quota was filled, compared to ten enrolled by conventional methods over a 90-day period. Medical reviewers rated the AI's justifications for each inclusion or exclusion criterion as 100% accurate and interpretable. Additionally, the AI correctly excluded 198 of 200 ineligible patients, achieving a negative predictive value of 99%. A relevant aspect was diversity: 36.6% of patients identified by AI were Black, compared to 7.1% detected by traditional screening. Only 60% of those selected by AI were previously connected to a heart failure specialist, compared to 92.8% of those identified by conventional methods, suggesting that AI can broaden access to clinical trials for underrepresented groups. Trejeeve Martyn, the principal investigator and director of Population Health with Heart Failure at the Cleveland Clinic, emphasized: “Medical AI allows for large-scale review of medical records, transforming a traditionally labor-intensive process. We can identify quality candidates more efficiently and diversely, and we are exploring how this technology can accelerate research and the implementation of therapies.” The Synapsis AI system, integrated into the Cleveland Clinic's electronic health record, analyzed data from 25 hospitals and 250 outpatient centers in Ohio, Florida, and Nevada, combining structured data and natural language processing to interpret clinical notes and lab results. Each inclusion or exclusion decision was accompanied by detailed and auditable justifications, facilitating verification by research coordinators. Eirini Schlosser, CEO and co-founder of Dyania Health, highlighted that AI can overcome efficiency and equity limitations in patient assignment to trials, even identifying those who typically fall outside traditional processes. The real-world experience of AI in this clinical trial, along with the results on diversity and performance, paves the way for an expansion of these tools for participant selection, population health records, and real-time quality reporting.