Fairness-Aware Interpretable AI in Bioinformatics and Healthcare

Author

• Euiseong Ko, University of Nevada, Las VegasFollow

Award Date

12-15-2025

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Science

First Committee Member

Mingon Kang

Second Committee Member

Kazem Taghva

Third Committee Member

Fatma Nasoz

Fourth Committee Member

Laxmi Gewali

Fifth Committee Member

Mira Han

Number of Pages

84

Abstract

The increasing integration of deep learning in Translational Bioinformatics demands a paradigm shift: moving beyond optimizing raw predictive accuracy to prioritizing fairness and interpretability. Systematic biases inherent in biomedical data, particularly the underrepresentation of diverse ancestries and sexes, pose significant risks. These biases can exacerbate existing health disparities by propagating inequitable outcomes in both research and clinical settings. In this high-stakes domain, interpretability is indispensable. Model inferences must be scientifically rigorous, clinically validated, and transparent to ensure effective translation from foundational research to clinical practice.

To address this complex challenge, we introduce Fairness-aware Interpretable AI frameworks engineered to promote equity in sensitive biomedical applications. These frameworks embed established biological domain knowledge directly into their architecture, thereby enabling intrinsic interpretability across diverse demographic groups. Our proposed approach significantly enhances predictive performance while simultaneously facilitating the discovery of demographic-specific biomarkers and generating personalized insights into underlying biological mechanisms. By unifying fairness and interpretability, this work accelerates innovation in precision medicine, yielding computational tools that are both ethically grounded and scientifically robust, ultimately ensuring more equitable patient care.

Controlled Subject

Bioinformatics; Biomedical engineering; Hospital care

Disciplines

Artificial Intelligence and Robotics | Bioinformatics | Computer Engineering

File Format

PDF

File Size

20700 KB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Repository Citation

Ko, Euiseong, "Fairness-Aware Interpretable AI in Bioinformatics and Healthcare" (2025). UNLV Theses, Dissertations, Professional Papers, and Capstones. 5436.
https://oasis.library.unlv.edu/thesesdissertations/5436

Rights

IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/