Date of Award
8-22-2025
Date Published
September 2025
Degree Type
Thesis
Degree Name
Master of Science (MS)
Department
Biomedical and Chemical Engineering
Advisor(s)
Zhen Ma
Second Advisor
Zhao Qin
Keywords
AI;Bioengineering;Cardiomyocyte;Machine Learning;Stem Cell
Abstract
Drug-induced cardiotoxicity remains a critical concern in pharmaceutical development as failed drug candidates result in lost time and resources that could be better spent on viable candidates. We believe that a robust in vitro model that accurately captures human cardiac physiology is an applicable model for examining drug candidates. In addition to those models, there is a need for quick and effective predictive models to analyze data obtained from the model to classify drugs based on their potential for cardiotoxicity. In this study, we conduct a functional analysis of human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) and pass the data collected into a machine learning framework to predict drug-induced contractile dysfunction. Our workflow involves preprocessing contractile waveforms, extracting key biomechanical features, and training an ML-based predictive framework to classify drug-induced effects with high accuracy. This approach showcases the scalability and precision of cardiotoxicity assessment while reducing experimental complexity and time. Additionally, we can analyze how the models came to their predictions to gain a better understanding of how the machine learning model is making its predictions. The findings demonstrate the potential of AI-driven physiological profiling for streamlining preclinical cardiac safety evaluation, providing an ethical and efficient platform for drug screening.
Access
Open Access
Recommended Citation
Watt, Anthony Allan, "Prediction of Drug-Induced Cardiotoxicity Using Machine Learning Analysis of hiPSC-CM Contractile Profiles" (2025). Theses - ALL. 987.
https://surface.syr.edu/thesis/987
