Foundation Award for Research

I. Focus of Research

The Computational Learning Systems (CLS) Lab at Missouri State University focuses on translational machine learning (ML), developing and applying explainable Artificial Intelligence (AI) methods with a strong emphasis on biomedical applications. Driven by practical challenges in healthcare and life sciences, my work aims to design intelligent computational systems capable of analyzing complex, large-scale data and generating meaningful, actionable predictions.

By analyzing multiple modalities of data from electronic health records, blood-based biomarkers, genotype data to imaging, my team has been at the forefront of developing explainable models for both unsupervised and supervised ML models in collaboration with medical domain experts. The clinical applications have been mainly in traumatic brain injury (TBI), Alzheimer’s disease (AD), and autism spectrum disorders. My team also investigates the underlying causes of large language model (LLM) failures to enhance their reliability in ontology-driven knowledge extraction, with a particular focus on biomedical term normalization and clinical note phenotyping. As AI-enabled clinical decision-support systems increasingly impact health outcomes, the need for responsible, trustworthy AI remains critical. Our research emphasizes the integration of ethical principles across the entire AI lifecycle—from problem formulation to deployment.

II. Title and year of major projects

• Integrating explainable AI in trajectory modeling of serum biomarker profiles to enhance management of patients with severe to mild TBI, 2025-ongoing
• Enhancing the utility of large language models for clinical applications, 2024-ongoing
• Genomic variant analysis for prioritization of druggable AD gene targets, 2023-ongoing
• Development of Explainable AI models for neurological disorders, 2020-ongoing
• Data-driven models for analyzing heterogeneity and complexity of TBI using multimodal data, 2019-2023
• Generating standardized benchmark datasets and evaluation metrics for assessing unsupervised learning models, 2018-2020
• Genetic analysis of grapevine species using machine learning techniques to identify disease resistance features 2016 - 2018
• Unsupervised learning models to elucidate heterogeneity in autism spectrum disorder, 2016 – 2020

III. Future directions of research

• Engineering novel features for prediction of TBI recovery trajectories based on pharmacokinetic modeling of blood-based biomarker data.
• Explainable AI framework for neurological disorders that leverages LLMs for enhanced explanations.
• Deriving meaningful metrics for assessing performance of LLMs in ontology data extraction.
• Genetic data-driven prioritization of druggable gene targets for Alzheimer's disease

IV. Topics related to your research and of interest to the broad University Community, for which you are available for presentations and/or consultations.

• Explainable AI/machine learning models
• Application of unsupervised learning methods for large-scale data analysis and knowledge discovery
• Use of large language models in biomedical research
• Ethics of AI
• Impact of AI models in neurological disorder research, in particular TBI and Alzheimer's disease.