Dissertations & Theses (Open Access)
Graduation Date
Summer 2024
Degree Name
Doctor of Philosophy (PhD)
School Name
McWilliams School of Biomedical Informatics at UTHealth Houston
Advisory Committee
Xiaoqian Jiang, PhD
Abstract
Blood testing is an indispensable tool to check the patient's health status. However, the overutilization of lab tests in healthcare settings is a prevalent issue that can lead to redundant information, increased patient risk, and financial strain on healthcare resources. This dissertation presents novel approaches to optimizing lab test utilization by applying deep learning models. The primary objective is to develop and validate methods that accurately predict and reduce unnecessary blood tests.
The study introduces a “selective” mechanism that quantifies the predictability of future lab tests, allowing for the selective reduction of unnecessary tests in the context of specific clinical scenarios. By designing a real-time recommendation system, the research supports healthcare professionals in making informed decisions about lab test ordering, enhancing the model's usability in clinical practices. Furthermore, the dissertation explores the identification of clinically meaningful patient subgroups through the deeplearning- based patient clustering method. By understanding the subgroup patterns identified by the model, physicians can learn about potential health risks for lab test reductions. Experimental results demonstrate that the proposed model achieves high prediction accuracy and maintains robustness during the run-time of lab test reductions. The findings demonstrate the clinical relevance of the approach, showing that stringent model settings ensure patient safety while reducing unnecessary testing. After incorporating the deep patient clustering task into the recommendation algorithm, the model can identify significantly different subgroup patterns. The analysis results of the subgroup patterns present interpretations of the patient’s health conditions.
In conclusion, this dissertation provides a comprehensive framework for optimizing lab test utilization, offering significant implications for reducing the health risk of excessive blood draws and improving resource management in healthcare.
Recommended Citation
Huang, Tongtong, "Designing Recommendation Systems for Identification and Reduction of Unnecessary Blood Tests" (2024). Dissertations & Theses (Open Access). 62.
https://digitalcommons.library.tmc.edu/uthshis_dissertations/62
Keywords
Deep Learning, Clinical Decision Support, Lab Test Reduction, Patient Safety
Comments
Journal article included in this dissertation:
Huang, T., Li, L. T., Bernstam, E. V., & Jiang, X. (2023). Confidence-based laboratory test reduction recommendation algorithm. BMC medical informatics and decision making, 23(1), 93. https://doi.org/10.1186/s12911-023-02187-3