Dementia Care Using AI: Real-time Patient Trajectory Monitoring System
Vikash Pr, S Dharsan, Madhu Shree Aravindan, Godfrey Winster S · 2023
Dementia, characterised by cognitive decline, has emerged as a significant global health concern, with a new case diagnosed every 3 seconds worldwide. This study addresses the challenges faced by dementia carers through an innovative approach that combines Progressive Web App (PWA) technology with machine learning and GPS trajectory analysis, customised to aid in care. The proposed method identifies and addresses gaps in existing research, aiming to improve the lives of individuals in the early stages of dementia and their carers by offering real-time anomaly detection and alerts based on patient trajectory data. The methodology encompasses several pivotal stages: Data acquisition and transmission involve collecting and relaying patient trajectory data to the cloud; the isolation forest technique, a sophisticated anomaly detection method, identifies deviations from normal behaviour within patient trajectories. Real-time anomaly detection and alert generation provide swift notification to carers in the event of abnormal patient movements. The approach also encompasses carer engagement and response dynamics. Carers play a crucial role in determining the veracity of detected anomalies and responding accordingly. Based on carer feedback, the system adapts and updates its algorithm for symptom detection and threshold setting. In conclusion, the integrated strategy presented in this study offers an innovative and effective approach to enhancing the quality of care for individuals with dementia and their carers, paving the way for enhanced patient monitoring and carer engagement.