Page 10 - JCTR-11-5

Journal of Clinical and

Translational Research

REVIEW ARTICLE
The role of large language models in induced

pluripotent stem cell-derived cardiomyocytes
research and clinical translation

Dhienda C. Shahannaz 1,2 , Tadahisa Sugiura * , and Brandon E. Ferrell 2
2
1 Department of Medicine, Medical Education and Research Institute, Faculty of Medicine, Universitas
Indonesia, Jakarta, Indonesia
2 Department of Cardiothoracic and Vascular Surgery, Montefiore Medical Center, Albert Einstein
College of Medicine, Bronx, New York, United States of America
(This article belongs to the Special Issue: Exploring the Potential of Large Language Models
(ChatGPT) in Cardiovascular Disease Management)

Abstract

Background: Induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) are
redefining cardiovascular regenerative medicine, yet challenges in differentiation
fidelity, functional maturation, and scalable production restrain their full clinical
potential. Aim: This review evaluates the pioneering integration of large language
*Corresponding author: models (LLMs)—including GPT-4, BioGPT, and BioMedLM—into iPSC-CM research
Tadahisa Sugiura and translational therapeutics, with a focus on advancing precision, efficiency, and
(tsugiura@montefiore.org) patient-specific care. Methods: Structured searches across biomedical and artificial
Citation: Shahannaz DC, intelligence-focused databases were conducted to map how LLMs augment literature
Sugiura T, Ferrell BE. The role of mining, experimental design, multi-omics integration, and clinical translation,
large language models in induced including personalized therapy prediction and drug safety assessment. Results: LLMs
pluripotent stem cell-derived
cardiomyocytes research and demonstrably surpass traditional tools in identifying gene-phenotype links, refining
clinical translation. J Clin Transl clustered regularly interspaced short palindromic repeats-based differentiation
Res. 2025;11(5):4-28. protocols, and merging patient-level datasets with iPSC-CM outputs. Limitations
doi: 10.36922/JCTR025230026
include model interpretability, reproducibility across genetically diverse populations,
Received: June 3, 2025 and ethical considerations regarding data privacy and bias. Conclusion: Despite
Revised: July 20, 2025 these barriers, early translational applications demonstrate that LLMs can accelerate
hypothesis generation, optimize laboratory-to-clinic pipelines, and enable high-
Accepted: August 7, 2025
fidelity, patient-specific cardiomyocyte modeling. Relevance for patients: The
Published online: September 2, synergy of LLM intelligence and iPSC-CM biology has the potential to deliver safer,
2025
more effective, and deeply personalized regenerative cardiac therapies—moving
Copyright: 2025 Author(s). the field closer to truly bespoke heart repair.
This is an open-access article
distributed under the terms of the
Creative Commons AttributionNon-
Commercial 4.0 International (CC Keywords: Artificial intelligence; Large language models; Biomedical natural language
BY-NC 4.0), which permits all programs; Induced pluripotent stem cells; Cardiac regenerative medicine
non-commercial use, distribution,
and reproduction in any medium,
provided the original work is
properly cited. 1. Introduction
Publisher’s Note: AccScience
Publishing remains neutral with Cardiovascular diseases (CVDs) remain the leading cause of mortality worldwide,
1
regard to jurisdictional claims in
published maps and institutional necessitating continuous advancements in therapeutic strategies. According to the
2
affiliations. World Health Organization, CVDs account for 32% of global deaths, with Indonesia
Volume 11 Issue 5 (2025) 4 doi: 10.36922/JCTR025230026