Interpretable Mechanistic and Machine Learning Models for Predicting Cardiac Remodeling from Biochemical and Biomechanical Features - Dissertation

Aug 4, 2026

Upload by Oliver

Biochemical and biomechanical signals drive cardiac remodeling, altering heart physiology and creating precursors to multiple cardiac diseases. Because current treatments such as Cardiac Resynchronization Therapy (CRT) yield highly variable patient responses, this dissertation develops computational, interpretable modeling to support identification of heart-failure patients likely to benefit from therapy. Ensemble machine learning predicts CRT response using demographic, biomarker, and LV-functional variables, while RNA-seq–inferred gene regulatory networks and a logic/OE-based mechanistic fibrosis model enable node prioritization and patient-specific simulation of clinically relevant correlations.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages171

Reading Time7 h 11 min

AccessFree to download

Abstract

Aim 1: Ensemble ML classification for CRT response prediction

Aim 2: Inferring fibrosis-related gene regulatory network from RNA-seq

Aim 3: Logic-based mechanistic model of cardiac fibrosis and perturbation analysis

Dedication and Acknowledgement

INTERPRETABLE MECHANISTIC AND MACHINE LEARNING MODELS FOR PREDICTING CARDIAC REMODELING FROM BIOCHEMICAL AND BIOMECHANICAL FEATURES | A Dissertation
Presented to
The Graduate School of
Clemson University | | --- | | In Partial Fulfillment
of the Requirement for the Degree
Doctor of Philosophy
Biomedical Data Science and Informatics | | by
Anamul Haque
December 2023 | Accepted by: Nina C. Hubig, PhD, Committee Chair William J. Richardson, PhD Brian C. Dean, PhD Bethany J. Wolf, PhD Awe J. Schoepf, MD Abstract Biochemical and biomechanical signals drive cardiac remodeling, resulting in altered heart physiology and the precursor for several cardiac diseases, the leading cause of death formost racial groups in the USA. Reversing cardiac remodeling requires medication and device-assisted treatment such as Cardiac Resynchronization Therapy (CRT), but current interventions produce highly variable responses from patient to patient. Mechanistic modeling and Machine learning (ML) approaches have the functionality to aid diagnosis and therapy selection using various input features. Moreover, 'Interpretable' machine learning methods have helped make machine learning models fairer and more suited for clinical application. The overarching objective of this doctoral work is to develop computational models that combine an extensive array of clinically measured biochemical and biomechanical variables to enable more accurate identification of heart failure patients prone to respond positively to therapeutic interventions. In the first aim, we built an ensemble ML classification algorithm using previously acquired data from the SMART-AV CRT clinical trial. Our classification algorithm incorporated 26 patient demographic and medical history variables, 12 biomarker variables, and 18 LV functional variables, yielding correct CRT response prediction in 71% of patients. In the second aim, we employed a machine learning-based method to infer the fibrosis-related gene regulatory network from RNA-seq data from the MAGNet cohort of heart failure patients. This network identified significant interactions between transcription factors and cell synthesis outputs related to cardiac fibrosis-a critical driver of heart failure. Novel filtering methods helped us prioritize the most critical regulatory interactions of mechanistic forward simulations. In the third aim, we developed a logic-based model for the mechanistic network of cardiac fibrosis, integrating the gene regulatory network derived from aim two into a previously constructed cardiac fibrosis signaling network model. This integrated model implemented biochemical and biomechanical reactions as ordinary differential equations based on normalized Hill functions. The model elucidated the semi-quantitative behavior of cardiac fibrosis signaling complexity by capturing multi-pathway crosstalk and feedback loops. Perturbation analysis predicted the most critical nodes in the mechanistic model. Patient-specific simulations helped identify which biochemical species highly correlate with clinical measures of patient cardiac function. Dedication I want to dedicate this dissertation work to my father, who died from a sudden cardiac arrest on the first day of graduate school. His passion for education has brought me to this country for higher education and helped me find my passion for cardiovascular disease research. I also want to dedicate this to my mother, whose whole life has passed raising our six siblings. My dedication also includes my siblings, especially my elder brother, for his endless contributions to shaping our family's bright future. My Ph. D. journey would not be possible without the support and encouragement of the love of my life, my wife, Nazneen Sultana. She believed me and supported my decision to move to Biomedical Data Science. Without her, this journey would be unthinkable. Finally, I want to dedicate this to my son, Ibrazul Haque. His presence

FAQs

What problem does this dissertation address in cardiac care?

Cardiac remodeling is driven by biochemical and biomechanical signals and is linked to multiple cardiac diseases. Current interventions such as CRT show large patient-to-patient variability, motivating better prediction and understanding of therapy response.

How is CRT response prediction performed in the first aim?

An ensemble ML classification algorithm uses previously acquired SMART-AV CRT clinical-trial data, incorporating demographic/medical-history variables, biomarker variables, and LV functional variables to predict correct CRT response in 71% of patients.

What models are used to study fibrosis mechanisms in the later aims?

The second aim infers fibrosis-related gene regulatory networks from RNA-seq data using machine-learning methods. The third aim integrates this network into a mechanistic cardiac fibrosis signaling framework implemented with ordinary differential equations from normalized Hill functions, enabling perturbation and patient-specific simulations.

Aug 4, 2026

Upload by Oliver