Human-Centered Explainable AI at the Edge for eHealth

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    1 Scopus citations

    Abstract

    Explainable Artificial Intelligence (XAI) is a new paradigm of Artificial Intelligence (AI) that is giving different AI/ Machine Learning (ML) models a boost to penetrate sectors where people are thinking about adopting AI. This work focuses on the adoption of XAI in the health sector. It portrays that careful integration of XAI in both cloud and edge could change the whole healthcare industry and make humans more aware of their present health conditions, which is the need of the hour. To demonstrate the same, we have done an experiment based on the prediction of a particular medical condition called "cardiac arrest"in a specific subject group (patients who are 70 years old). Here, based on the explanation provided by the XAI model (e.g., SHAP, LIME) at Cloud and Edge, our system can predict the chances of a "cardiac arrest"for the subject with a valid explanation. This type of model will be the next big upgrade in the healthcare industry in terms of automation and a self-explanatory system that works as a personal health assistant for individuals.

    Original languageBritish English
    Title of host publicationProceedings - 2023 IEEE International Conference on Edge Computing and Communications, EDGE 2023
    EditorsClaudio Ardagna, Feras Awaysheh, Hongyi Bian, Carl K. Chang, Rong N. Chang, Flavia Delicato, Nirmit Desai, Jing Fan, Geoffrey C. Fox, Andrzej Goscinski, Zhi Jin, Anna Kobusinska, Omer Rana
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages227-232
    Number of pages6
    ISBN (Electronic)9798350304831
    DOIs
    StatePublished - 2023
    Event7th IEEE International Conference on Edge Computing and Communications, EDGE 2023 - Hybrid, Chicago, United States
    Duration: 2 Jul 20238 Jul 2023

    Publication series

    NameProceedings - IEEE International Conference on Edge Computing
    Volume2023-July
    ISSN (Print)2767-9918

    Conference

    Conference7th IEEE International Conference on Edge Computing and Communications, EDGE 2023
    Country/TerritoryUnited States
    CityHybrid, Chicago
    Period2/07/238/07/23

    Keywords

    • Edge
    • eHealth
    • Interpretability
    • IoMT
    • Machine Learning
    • XAI

    Fingerprint

    Dive into the research topics of 'Human-Centered Explainable AI at the Edge for eHealth'. Together they form a unique fingerprint.

    Cite this