Empathic conversational agents for real-time monitoring and co-facilitation of patient-centered healthcare

被引:26
作者
Adikari, Achini [1 ]
de Silva, Daswin [1 ]
Moraliyage, Harsha [1 ]
Alahakoon, Damminda [1 ]
Wong, Jiahui [2 ]
Gancarz, Mathew [2 ]
Chackochan, Suja [2 ]
Park, Bomi [3 ]
Heo, Rachel [3 ]
Leung, Yvonne [2 ,3 ]
机构
[1] La Trobe Univ, Ctr Data Analyt & Cognit, Bundoora, Vic 3083, Australia
[2] De Souza Inst, Toronto, ON, Canada
[3] Univ Toronto, Fac Med, Dept Psychiat, Toronto, ON, Canada
来源
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE | 2022年 / 126卷
关键词
Chatbots; Conversational agents; Patient-centered care; Cancer care; Empathic AI; Human-centric AI; Patient emotions; Group emotions; Real-time monitoring; Co-facilitation; Markov models; Artificial intelligence; EMOTIONS; SUPPORT;
D O I
10.1016/j.future.2021.08.015
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Healthcare systems across the world are transitioning into patient-centered healthcare models to ensure improved health outcomes, increased operational efficiencies and respectful patient engagement. Digital health technologies are at the forefront of this transition in facilitating a role for the patient in the clinical dimensions of the healthcare trajectory, from diagnosis and interventions to treatment and recovery. Despite this prevalence in the clinical space, the non-clinical needs of patient mental health and wellbeing are frequently overlooked by contemporary patient-centered healthcare models. Conversational agents (or chatbots) are digital dialogue systems that are widespread and widely used in sequential information provision and information acquisition tasks. Given the intimate nature of this human-machine interaction, conversational agents can be effectively utilized to support and sustain patient mental health and wellbeing. In this paper, we propose an empathic conversational agent framework based on an ensemble of natural language processing techniques and artificial intelligence algorithms for real-time monitoring and co-facilitation of patient-centered healthcare for improved mental health and wellbeing outcomes. The technical contributions of this framework are; detection of patient emotions, prediction of patient emotion transitions, detection of group emotions, formulation of patient behavioral metrics, and resource recommendations based on patient concerns. The architectural contributions of the framework are intelligent communication channels that stream empathic conversational elements and resource recommendations for the multi-user conversations and co-facilitation updates for the human healthcare provider interface. The framework was empirically evaluated on a benchmark dataset and further validated based on a clinical protocol designed for its application in an online support group setting for cancer patients and caregivers in Canada. The results of these experiments confirm the effectiveness of this framework, its contributory role and practical value in realizing a patient-centered healthcare model for improved mental health and wellbeing outcomes. (C) 2021 Elsevier B.V. All rights reserved.
引用
收藏
页码:318 / 329
页数:12
相关论文
共 53 条
  • [1] Understanding Citizens' Emotional Pulse in a Smart City Using Artificial Intelligence
    Adikari, Achini
    Alahakoon, Damminda
    [J]. IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2021, 17 (04) : 2743 - 2751
  • [2] A self structuring artificial intelligence framework for deep emotions modeling and analysis on the social web
    Adikari, Achini
    Gamage, Gihan
    de Silva, Daswin
    Mills, Nishan
    Wong, Sze-Meng Jojo
    Alahakoon, Damminda
    [J]. FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE, 2021, 116 : 302 - 315
  • [3] Can online support groups address psychological morbidity of cancer patients? An artificial intelligence based investigation of prostate cancer trajectories
    Adikari, Achini
    de Silva, Daswin
    Ranasinghe, Weranja K. B.
    Bandaragoda, Tharindu
    Alahakoon, Oshadi
    Persad, Raj
    Lawrentschuk, Nathan
    Alahakoon, Damminda
    Bolton, Damien
    [J]. PLOS ONE, 2020, 15 (03):
  • [4] Adikari A, 2019, IEEE INTL CONF IND I, P183, DOI [10.1109/indin41052.2019.8972196, 10.1109/INDIN41052.2019.8972196]
  • [5] An information-theoretic perspective of tf-idf measures
    Aizawa, A
    [J]. INFORMATION PROCESSING & MANAGEMENT, 2003, 39 (01) : 45 - 65
  • [6] Survey on Intelligent Chatbots: State-of-the-Art and Future Research Directions
    Almansor, Ebtesam H.
    Hussain, Farookh Khadeer
    [J]. COMPLEX, INTELLIGENT, AND SOFTWARE INTENSIVE SYSTEMS (CISIS 2019), 2020, 993 : 534 - 543
  • [7] Cohesion in online environments
    Altebarmakian, Maria
    Alterman, Richard
    [J]. INTERNATIONAL JOURNAL OF COMPUTER-SUPPORTED COLLABORATIVE LEARNING, 2019, 14 (04) : 443 - 465
  • [8] [Anonymous], 2021, GLOVE GLOBAL VECTORS
  • [9] [Anonymous], 2021, STATE OF THE ART OPE
  • [10] [Anonymous], 2019, CUSTOMER SUPPORT TWI