Related Experiment Videos
Nurse-Led Large Language Model Chatbot for Predicting and Preventing Complications After Coronary Artery Bypass
Salini K1, Ajee K L1, Praveen Kerala Varma2
1Department of Medical Surgical Nursing, Amrita College of Nursing, Amrita Vishwa Vidyapeetham, Amrita Institute of Medical Sciences and Research Center, Kochi Campus, Ponekkara, Edappally, Kochi, Kerala, 682041, India, 91 8335053920.
JMIR Research Protocols
|August 4, 2026
Summary
This study introduces Smart CABGuard, an AI-powered system using large language models (LLMs) and ECG monitoring to prevent unplanned readmissions after coronary artery bypass grafting (CABG). It aims to improve patient care in low- and middle-income countries by offering continuous remote surveillance.
Area of Science:
- Cardiovascular Surgery
- Artificial Intelligence in Healthcare
- Digital Health
Background:
- Thirty-day unplanned readmission after coronary artery bypass grafting (CABG) affects 10%-20% of patients, a critical quality metric, especially in low- and middle-income countries (LMICs) with limited cardiac rehabilitation access.
- Current risk models for post-CABG readmission are static and lack real-time patient engagement.
- No validated large language model (LLM)-based clinical decision support (CDS) system currently exists for preventing post-CABG readmissions.
Purpose of the Study:
- To develop, validate, and evaluate Smart CABGuard, a novel nurse-led, LLM-based CDS chatbot system.
- To predict and prevent postoperative complications and 30-day unplanned readmissions following isolated CABG.
- To integrate continuous remote electrocardiographic (ECG) monitoring for enhanced patient surveillance.
Main Methods:
- A multiphase translational study adhering to TRIPOD-LLM and CONSORT AI guidelines.
- Phase I: Development of a Complication Risk Index (CRI) using ambispective data and logistic regression.
- Phase II: Refinement of the Smart CABGuard chatbot integrating a Mistral-7B LLM, CRI engine, explainable AI, and ECG telemetry.
- Phase III: A prospective, parallel-group, open-label randomized controlled trial (RCT) with 800 patients comparing Smart CABGuard-assisted care to standard care.
Main Results:
- The study protocol outlines the methodology for developing and testing the Smart CABGuard system.
- The primary outcome is the reduction of 30-day all-cause unplanned readmission rates.
- Statistical power calculations indicate a target sample size of 800 patients for the RCT.
Conclusions:
- Smart CABGuard represents the first RCT protocol for an LLM-based nurse-led CDS system aimed at post-CABG readmission prevention in an LMIC.
- The system seeks to establish the utility, safety, and feasibility of AI-augmented postoperative surveillance.
- This initiative aims to significantly improve postoperative care and reduce readmission rates in vulnerable populations.
Related Concept Videos
01:27Coronary Artery Disease V: Interprofessional Care
Interprofessional care for coronary artery disease includes pharmacological therapy and revascularization procedures.Pharmacological therapy for Coronary Artery Disease (CAD) aims to manage symptoms, prevent complications, and improve patient outcomes through various classes of medications:Antiplatelet Agents:Aspirin and Clopidogrel: These medications inhibit platelet aggregation, preventing blood clots, which is crucial for avoiding heart attacks and strokes. Doctors often prescribe these...
01:26Coronary Artery Disease IV: Preventive Measures
Effective preventive measures for coronary artery disease (CAD) focus on controlling modifiable risk factors, including cholesterol abnormalities and lifestyle changes.Cholesterol ManagementFirst, the Mediterranean diet and the American Heart Association advocate for maintaining low-density lipoprotein (LDL) cholesterol levels below 100 mg/dL, with a more stringent recommendation of below 70 mg/dL for individuals at high risk. LDL cholesterol, often termed "bad cholesterol," can lead to the...