Abstract:
Patients often visit multiple hospitals throughout their lives, many of which are located in different cities across Saudi Arabia. These visits can range from regular che...Show MoreMetadata
Abstract:
Patients often visit multiple hospitals throughout their lives, many of which are located in different cities across Saudi Arabia. These visits can range from regular checkups and second opinions to emergency cases. A significant issue with the current healthcare system in Saudi Arabia is that each hospital maintains its own system with separate patient information. Due to the lack of communication between hospitals and physicians, patients frequently undergo unnecessary checkups, tests, or even surgeries, resulting in increased costs, reduced healthcare quality, and wasted medical resources or lives for both patients and public hospitals. To address this issue, we propose the MedCore iOS application, which aims to connect hospitals in Saudi Arabia, allowing physicians, patients, laboratory specialists, and hospitals to access information easily and transparently. Furthermore, MedCore leverages the power of AI and data availability by serving as a diagnostic tool for physicians. We trained this diagnostic model on 4,962 patients with 140 unique symptoms using decision trees, and optimized it via grid and random searches. The optimized model achieved 91% accuracy and 92% F1-score. Combining this model with a physician's own prediction can increase their decision certainty or provide an informative experience, ultimately enhancing healthcare outcomes in Saudi Arabia.
Published in: 2024 International Conference on Smart Applications, Communications and Networking (SmartNets)
Date of Conference: 28-30 May 2024
Date Added to IEEE Xplore: 05 July 2024
ISBN Information: