A ChatGPT-enabled Intelligent Assistive Technology for Homecare to Support Activities of Daily Living for People with Mild Dementia

被引:0
作者
Chen, Xiaowei [1 ]
Fan, Guoliang [1 ]
Roberts, Emily [2 ]
Howell, Steven, Jr. [1 ]
机构
[1] Oklahoma State Univ, Sch Elect & Comp Engn, Stillwater, OK 74078 USA
[2] Oklahoma State Univ, Dept Design & Merchandising, Stillwater, OK 74078 USA
来源
2024 4TH INTERNATIONAL CONFERENCE ON HUMAN-MACHINE INTERACTION, ICHMI 2024 | 2024年
基金
美国国家卫生研究院;
关键词
ChatGPT; transfer learning; people with mild dementia; intelligent assistive technologies; audio/visual prompting; indoor localization; instrumental activities of daily living;
D O I
10.1145/3678429.3678430
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
People with dementia (PwD) face significant cognitive challenges that hinder their ability to perform daily tasks, referred to as activities of daily living (ADLs), and are particularly impeded when it comes to the more intricate instrumental ADLs (IADLs). Lacking sufficient support, individuals with dementia are at a greater risk of losing their independence and becoming more vulnerable. A range of intelligent assistive technologies (IATs) have been proposed to help individuals with mild dementia perform certain ADLs on their own. However, most current IATs offer limited support for IADLs to PwD, primarily due to the technologies inability to comprehend the unique communication needs of PwD and a lack of user-specific customization. This paper presents a new proposed homecare assistive tool for people with mild dementia, called CATcare (Cognitive Assistive Technology Care). Our CATcare system, utilizing ChatGPT, offers a comprehensive understanding of PwD. Easily operated and configured via smartphone or smart glasses, it allows caregivers to tailor it to the unique IADL requirements of each care recipient. Additionally, it offers step-by-step guidance, prompts, and timely feedback to help PwD complete their desired tasks. Our research leverages the capabilities of transfer learning from AI models in indoor localization, object detection, and Natural Language Processing (NLP) and shows the potential and promise to develop a customizable and personalizable CATcare tool. This CATcare system is intended to improve the quality of life of PwD and reduce the burden of their caregivers.
引用
收藏
页码:1 / 7
页数:7
相关论文
共 38 条
  • [1] Alam Ridwan, 2018, 2018 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI), P170, DOI 10.1109/BHI.2018.8333396
  • [2] Alessa AM, 2023, Arxiv, DOI arXiv:2304.09866
  • [3] Alyafeai Z, 2020, Arxiv, DOI arXiv:2007.04239
  • [4] [Anonymous], 2018, Mobile object detection using tensorflow lite and transfer learning
  • [5] Banovic Silva, 2018, Mater Sociomed, V30, P221, DOI 10.5455/msm.2018.30.221-224
  • [6] Chen Xiaowei, 2023 IEEE 23 INT C B
  • [7] Promoting Reminiscences with Virtual Reality Headsets: A Pilot Study with People with Dementia
    Coelho, Tiago
    Marques, Catia
    Moreira, Daniela
    Soares, Maria
    Portugal, Paula
    Marques, Antonio
    Ferreira, Ana Rita
    Martins, Sonia
    Fernandes, Lia
    [J]. INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH, 2020, 17 (24) : 1 - 13
  • [8] Desai Shital, 2022, Designing Interactions for Music and Sound, P151
  • [9] The experience of using prompting technology from the perspective of people with Dementia and their primary carers
    Evans, N.
    Boyd, H.
    Harris, N.
    Noonan, K.
    Ingram, T.
    Jarvis, A.
    Ridgers, J.
    Cheston, R.
    [J]. AGING & MENTAL HEALTH, 2021, 25 (08) : 1433 - 1441