A KNOWLEDGE DRIVEN EXPERT SYSTEM FOR EARLY DEMENTIA SCREENING AND DIAGNOSIS

Abstract

Dementia is a neurological disorder that progresses and has a significant impact on memory, cognitive functions, and daily activities. Early detection is essential for enhancing patient outcomes and optimizing care strategies. Nevertheless, traditional diagnostic methods can be lengthy, require specialized knowledge, and often miss crucial early indicators. This research introduces a knowledge-driven expert system aimed at facilitating the early screening and diagnosis of dementia. The objective of this study is to enhance access to early screening, support healthcare professionals in their decision-making processes, and ultimately improve patient management and overall quality of life. The system uses a meticulously organized knowledge-base and inference mechanisms to emulate expert reasoning, which enables it to assess patient symptoms and provide initial evaluations. Comprehensive validation against real-world clinical datasets showcases the system's impressive performance, with the interactive expert system achieving a peak accuracy of 93%, and a precision of 0.97 surpassing the traditional statistical models. The system's rule-based architecture is not only a step towards improving diagnostic accuracy, but it also enables critical interpretability, fostering trust among clinicians, and allowing for smooth integration into current healthcare practices.