Diagnostic Accuracy of GPT-4o and Claude for HEART Score Calculation in Chest Pain (NCT07626060) | Clinical Trial Compass
RecruitingNot Applicable
Diagnostic Accuracy of GPT-4o and Claude for HEART Score Calculation in Chest Pain
Turkey (Türkiye)690 participantsStarted 2026-06
Plain-language summary
This prospective observational diagnostic accuracy study evaluates whether large language models (LLMs) - GPT-4o (OpenAI, gpt-4o-2024-11-20) and Claude (Anthropic, claude-sonnet-4-6) - can accurately calculate HEART scores from unstructured Turkish clinical notes and predict 30-day major adverse cardiac events (MACE) in emergency department patients presenting with non-traumatic chest pain.
The study will enroll 600 consecutive adult patients. For each patient, the same anonymized data (free-text anamnesis, ECG report text, troponin value, and age) will be independently processed by both LLMs via separate API calls with deterministic settings (temperature=0, JSON format). A three-expert consensus HEART score - derived through blinded independent scoring by three emergency medicine physicians with majority-vote adjudication - serves as the reference standard for agreement analysis. Actual 30-day MACE (all-cause death, AMI Type 1/2/4b, unplanned revascularization) determined via national health database and telephone follow-up serves as the outcome for diagnostic accuracy analysis.
A secondary documentation-quality sub-study will quantify how spontaneously Turkish emergency anamnesis notes capture HEART score parameters.
Who can participate
Age range
18 Years
Sex
ALL
See this in plain English?
AI-rewrites the medical criteria so a patient or caregiver can understand them. Always confirm with the trial site.
INCLUSION CRITERIA:
* Age \>=18 years
* Chief complaint of non-traumatic chest pain at the emergency department
* Written informed consent obtained from the patient or legally authorized representative
* Availability for 30-day follow-up (reachable by telephone and/or actively registered in the e-Nabiz national health database)
EXCLUSION CRITERIA:
* Traumatic chest pain etiology
* ST-elevation myocardial infarction (STEMI) at presentation requiring immediate reperfusion protocol
* Refusal or subsequent withdrawal of informed consent
* Inability to complete the mandatory 30-day follow-up period
WITHDRAWAL CRITERIA:
* Patient or representative requests data withdrawal after initial consent
* Administrative identification of retrospective data entry after enrollment
Questions worth asking your doctor
Bring these to your next appointment. They're a starting point for a shared conversation — not a sign you qualify or a recommendation to enrol.
1This trial is testing whether AI tools like GPT-4o and Claude can accurately calculate a chest pain risk score — does that mean if I enroll, my care decisions might be influenced by an AI system, and how would that affect my actual treatment?
2Since this study is measuring diagnostic accuracy rather than testing a new treatment, what would my role actually involve — would I just be providing data from my existing chest pain workup, or would there be additional tests or procedures?
3The trial isn't recruiting yet — given my current situation, is it worth waiting for this study to open, or is there a more pressing reason to move forward with standard care now?
4The HEART score is already an established tool used in emergency medicine — what would it mean for my care if the AI version turns out to be less accurate than the traditional method, and how does my doctor currently use the HEART score for patients like me?
5Since this is a diagnostic accuracy study and not a treatment trial, would participating give me any direct medical benefit, or would it primarily contribute to research about AI in emergency medicine?
Generated to help you prepare — always confirm anything about your own eligibility and care with the study team and your doctor.
Questions for the trial coordinator
The trial coordinator is the person who runs the study day to day. These cover the practical side — logistics, costs, and what taking part would actually mean for your life. The study team confirms whether you meet the criteria; these are questions to ask, not a sign you qualify.
1What does taking part actually involve week to week — how many visits, where, and how long does each one take?
2What costs are covered by the study, and what might I have to pay for myself, including travel, parking, or time off work?
3What happens during screening, and what happens if the study team confirms I don't meet the criteria after those tests?
4Who pays for the scans, blood work, and other tests the trial requires — the study, my insurance, or me?
5How will being in the trial affect my regular care, and will my own doctor stay informed and involved?
6Can I leave the trial at any point if I change my mind, and what would happen to my care if I do?
A starting point for the conversation — always confirm anything about your own eligibility, costs, and care with the study team and your doctor.
What they're measuring
1
Area Under the ROC Curve (AUC) of GPT-4o and Claude HEART Score for 30-Day MACE Prediction
Timeframe: 30 days after index emergency department visit
Trial details
NCT IDNCT07626060
SponsorMarmara University Pendik Training and Research Hospital