Deep Learning-based Scene Understanding Model for Assistive System Related to Alzheimer’s Patients  
Author Ke Xu


Co-Author(s) Suranjan Panigrahi


Abstract The ability to automatically identify the indoor room is a very important feature of the assistive living system for Alzheimer patients. By accurately identifying the patient’s whereabouts, the smart home system could effectively estimate their potential action and provide help accordingly, or send alert to the caregivers in an emergent situation. In this paper, we developed a deep learning-based model for indoor scene understanding was developed and evaluated. This model is envisioned as a part of the future assistive system for Alzheimer’s patients. The overall accuracies of the model ranged from 88% to 95%.


Keywords Assistive System, Alzheimer’s Patients, Scene Understanding, Deep Learning
    Article #:  DSIS19-74
Proceedings of ISSAT International Conference on Data Science & Intelligent Systems
August 1-3, 2019 - Las Vegas, NV, U.S.A.