Improving Stroke Clinical Trials with Medical Imaging and Advanced Algorithms
Date
2024-11-18
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Clinical trials are essential for advancing healthcare by exploring new treatments to improve patient outcomes. However, recruiting eligible participants for trials remains a significant challenge, particularly in acute conditions like stroke. Stroke is a leading cause of mortality, and timely treatment is critical. While recruiting patients for stroke trials is essential, manual screening for trial eligibility can be slow, prone to errors, and costly. Additionally, the complexity of trial inclusion criteria and the need for rapid decision-making in emergencies make it unrealistic for clinicians to be aware of all ongoing trials. This thesis focuses on three key areas to enhance both the quality and quantity of stroke trial recruitment: 1) the development of automated matching algorithms to improve the number and quality of enrollment; 2) reducing radiation doses in computed tomography perfusion (CTP) imaging to keep patients safer; 3) implementing a multi-site data harmonization pipeline on CTP images to improve the quality of data and ensure consistent and comparable results. The results show that the automated algorithms could quickly and accurately identify eligible participants, reducing human error and bias. The post-processing denoising algorithm effectively reduced the radiation dose by up to 80% while retaining most clinical features. For the harmonization, the variations between sites were mitigated, ensuring more consistent and reliable imaging data. In conclusion, these studies have potential implications that could enhance both the quality and quantity of stroke clinical trials, accelerating medical advancements and promising to improve the outcomes for stroke patients.
Description
Keywords
Stroke, Clinical Trials, Medical Imaging, Matching Algorithms, Denoising Algorithms, Data Harmonization
Citation
Charatpangoon, P. (2024). Improving stroke clinical trials with medical imaging and advanced algorithms (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.