HELENA: An intelligent digital assistant based on a Lifelong Health User Model
Introduction
The implementation of information and communication technology in the medical field is commonly referred to as eHealth (or digital health), which consists of using information technology for the benefit of human health (El Benny et al., 2021, Srivastava et al., 2015). Applications for monitoring patients usually adopt wearable smart sensors and devices to monitor vital parameters such as heart rate, blood pressure, blood saturation, and body temperature. These devices are often connected to the Internet or mobile platforms via Bluetooth technology (Sebestyen et al., 2014, Zhang et al., 2021). This puts the world of eHealth within reach of the end-users, who can gain a huge advantage in dealing with daily activities that affect their health.
In this user-centered vision of eHealth, digital assistants can play an important role. For instance, voice-based digital assistants, such as Siri and Alexa, are rapidly becoming adopted by consumers (Brill, Munoz, & Miller, 2019). Recent studies have shown high user satisfaction when using conversational devices and personal digital assistants in eHealth, but the perception of their limited reliability is still a problem (Abdul-Kader and Woods, 2015, Omrani et al., 2022).
It is crucial to provide reliable eHealth services based on explainable models that support decision-making. It is necessary to design systems that support the physician in making informed and timely decisions and provide an empathetic, easy-to-use, reliable interface to improve user experience and increase trust.
The contributions of this work are:
- •
the conceptualization of LHUM, a Lifelong Health User Model, which is a comprehensive clinical user profile fed with data coming from the interaction with the users or from other systems. LHUM could be exported in standard formats and shared among physicians treating the patient;
- •
the design and implementation of HELENA, a healthcare digital assistant exploiting LHUM to provide personalized services. HELENA allows users to monitor personal health status through a text-based interface (i.e., a chatbot), as well as a classic graphical interface with buttons and labels;
- •
the definition of a privacy-aware and transparent process for lifelong monitoring of the patient’s health status;
- •
an experimental session to assess user satisfaction with HELENA.
Section snippets
Related work
Emerging technologies are reshaping the healthcare sector in several ways: how consumers access it, how and which providers deliver it, and what outcomes must be achieved. Such ecosystems could be enabled by a combination of: (i) A holistic user model to integrate information that today is fragmented across different information systems (Musto, Narducci, et al., 2020); (ii) Advanced data analytics and artificial intelligence personalization engines to generate insights for patients and their
Lifelong Health User Model
User profiling is nowadays a common practice over the web, which began with seminal work by Kobsa, Koenemann, and Pohl (2001), who proposed a keyword-based representation of the web user. Since then, many strategies have been proposed to exploit the “digital footprints” that people leave on the web (i.e., social networks, web services, purchases, preferences, searches, etc.) to build a user profile (Polignano, Basile, et al., 2017, Polignano, Gemmis, et al., 2017, Polignano, Narducci, et al.,
HELENA: An intelligent digital assistant for smart healthcare
This section presents the architecture and implementation details of HELENA, a digital assistant that allows users to store, access, share, and update information regarding their health profile.
Experimental evaluation
We carried out a user study to assess both the effectiveness and usability of the platform. To this purpose, we addressed the following three research questions:
- •
RQ1: completeness and effectiveness of the conceptual model. Which facets of the LHUM do the users consider more relevant? Is the user profile complete and effective?
- •
RQ2: usability (i.e., Efficiency, Emotional Feeling, Functionality, Control, Simplicity of Use, Global Perception) of the two user interfaces, i.e., graphical and
Implications and limits of the work
The HELENA platform differs significantly from our previous project HealthAssistantBot (Polignano et al., 2020), a simple Telegram chatbot offering basic functionalities for managing demographic information and treatments, and health information monitoring. HELENA was designed to overcome the lack, in the health domain, of a comprehensive and maintainable profile. For this purpose, we propose LHUM (Lifelong Health User Model), a health model that covers all possible aspects of daily patient
Conclusions and future work
This work proposed HELENA, a platform to manage users’ health information stored in a lifelong health user model (LHUM). The main aim of our work was to design a comprehensive medical user profile filled with data from heterogeneous sources, that are easy to manage and share, and that can be used in the long-term. The main advantage of this kind of profile is to store health information in a single digital place for a lifetime. LHUM is composed of eight aspects that focus only on medical
CRediT authorship contribution statement
Marco Polignano: Methodology, Software, Writing – review & editing. Pasquale Lops: Conceptualization. Marco de Gemmis: Formal analysis. Giovanni Semeraro: Supervision, Data curation.
Acknowledgments
This work is supported by the project “AmICA: Assistenza olistica Intelligente per l’aCtive Ageing in ecosistemi indoor e outdoor” n°T1-MZ-09, Traiettoria 1 del Piano operativo salute: “Active & Healthy Ageing - Tecnologie per l’invecchiamento attivo e l’assistenza domiciliare”, Ministero della Salute Italiano. The work of Marco Polignano has been also supported by Apulia Region, Italy through the project ”Un Assistente Dialogante Intelligente per il Monitoraggio Remoto di Pazienti ” (Grant n.
Marco Polignano is Assistant Professor at the Department of Informatics, University of Bari, Italy, in the SWAP (Semantic Web Access and Personalization) research group. He received a Ph.D. in Computer Science and Mathematics in 2018, at the same university, with the thesis titled ”An affect-aware computational model for supporting decision-making through recommender systems”. He was a program committee member for many international conferences, including IJCAI, ECAI, IUI, AIVR, WWW. He was a
References (48)
- et al.
An affective circumplex model of neural systems subserving valence, arousal, and cognitive overlay during the appraisal of emotional faces
Neuropsychologia
(2008) - et al.
A systematic review of intelligent assistants
Future Generation Computer Systems
(2022) Machine learning for medical diagnosis: History, state of the art and perspective
Artificial Intelligence in Medicine
(2001)- et al.
To trust or not to trust? An assessment of trust in AI-based systems: Concerns, ethics and contexts
Technological Forecasting and Social Change
(2022) - et al.
Towards emotion-aware recommender systems: An affective coherence model based on emotion-driven behaviors
Expert Systems with Applications
(2021) - et al.
Application of cognitive computing in healthcare, cybersecurity, big data and IoT: A literature review
Information Processing and Management
(2022) - et al.
Survey on chatbot design techniques in speech conversation systems
International Journal of Advanced Computer Science and Applications
(2015) - et al.
User-centered design
Color theory and its application in art and design, vol. 19
(2013)- et al.
Alzheimer assistant: A mobile application using machine learning
Romanian Journal of Information Technology and Automatic Control
(2020)
A case study of usability testing–the SUMI evaluation approach of the EducaNext portal
WSEAS Transactions on Information Science and Applications
Deep learning and medical diagnosis: A review of literature
Multimodal Technologies and Interaction
Siri, Alexa, and other digital assistants: A study of customer satisfaction with artificial intelligence applications
Journal of Marketing Management
Sonars: A social networks-based algorithm for social recommender systems
Real world user model: Evolution of user modeling triggered by advances in wearable and ubiquitous computing
Information Systems Frontiers
ENRICHME: Perception and interaction of an assistive robot for the elderly at home
International Journal of Social Robotics
A tool of conversation: Chatbot
International Journal of Computer Sciences and Engineering
TensioBot: A chatbot assistant for self-managed in-house blood pressure checking
Journal of Medical Systems
Basic emotions
Application of the eHealth literacy model in digital health interventions: Scoping review
Journal of Medical Internet Research
Google android: An emerging software platform for mobile devices
International Journal on Computer Science and Engineering
SUMI: The software usability measurement inventory
British Journal of Educational Technology
Personalised hypermedia presentation techniques for improving online customer relationships
The Knowledge Engineering Review
Separation of concerns in model-driven development
IEEE Software
Cited by (1)
Human Understanding Capabilities of Symbiotic AI systems in FAIR (Future Artificial Intelligence Research)
2023, CEUR Workshop Proceedings
Marco Polignano is Assistant Professor at the Department of Informatics, University of Bari, Italy, in the SWAP (Semantic Web Access and Personalization) research group. He received a Ph.D. in Computer Science and Mathematics in 2018, at the same university, with the thesis titled ”An affect-aware computational model for supporting decision-making through recommender systems”. He was a program committee member for many international conferences, including IJCAI, ECAI, IUI, AIVR, WWW. He was a local organizing committee member for the Ai*iA 2017 conference and organizer of the Evalita 2018 challenge — ABSITA about the aspect-based sentiment analysis, Evalita 2020 ATE_ABSITA, UMAP 2020-2022 ExUm workshop about user modeling and personalization. In 2016 and 2018, he was a Marie Skłodowska-Curie Research and Innovation Staff Exchange (MSCA-RISE) fellow, involved in the project N. 691071, titled ”Seo-Dwarf: Semantic EO Data Web Alert and Retrieval Framework”. His research interests are Information Filtering, Recommender Systems, Natural Language Processing, Cognitive computing.
Pasquale Lops is Associate Professor at the Department of Computer Science, University of Bari Aldo Moro, Italy. He received the Ph.D. in Computer Science from the University of Bari in 2005 with a dissertation on “Hybrid Recommendation Techniques based on User Profiles”. His research interests include recommender systems and user modelling, with a specific focus on the adoption of techniques for semantic content representation. He authored over 200 articles, and he is one of the authors of the textbook ”Semantics in Adaptive and Personalized Systems: Methods, Tools and Applications”, published by Springer. He was Area Chair of User Modelling for Recommender Systems at UMAP 2016, and co-organized more than 20 workshops related to user modeling and recommender systems. He gave a tutorial on “Semantics-Aware Techniques for Social Media Analysis, User Modeling, and Recommender Systems” at UMAP 2016 and 2017, he was a speaker at two editions of the ACM Summer School on Recommender Systems. He was a keynote speaker at the 1st Workshop on New Trends in Content- based Recommender Systems (CBRecSys) at RecSys 2014. Finally, he gave the interview “Beyond TF- IDF” in the Coursera MOOC on Recommender Systems.
Marco de Gemmis is Associate Professor at the Department of Computer Science, University of Bari Aldo Moro, Italy, where he received his PhD in Computer Science in 2005. His primary research interests include content-based recommender systems, natural language processing, information retrieval, text mining, and in general personalized information filtering. He authored over 100 scientific articles published in international journals and collections, proceedings of international conferences and workshops, and book chapters. He was program committee member for international conferences, including: ACM Recommender Systems; User Modeling, Adaptation and Personalization (UMAP), and served as a reviewer for international journals. He was invited speaker at several universities, including: University of Roma 3, University of Basque Country San Sebastian, University of Cagliari, University of Milano-Bicocca, University of Naples Federico II, and at Workshop on Semantics-Enabled Recommender Systems at ICDM 2016. He was Marie Curie Fellow in the SEO-DWARF project, funded by the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 691071.
Giovanni Semeraro is full professor of computer science at University of Bari Aldo Moro, Italy, where he teaches “Intelligent Information Access and Natural Language Processing”, and “Programming languages”. He leads the Semantic Web Access and Personalization (SWAP) “Antonio Bello” research group. In 2015 he was selected for an IBM Faculty award on Cognitive Computing for the project “Deep Learning to boost Cognitive Question Answering”. He was one of the founders of AILC (Italian Association for Computational Linguistics) and on the Board of Directors till 2018. From 2006 to 2011 he was on the Board of Directors of AI*IA (Italian Association for Artificial Intelligence). His research interests include machine learning; AI and language games; recommender systems; user modelling; intelligent information mining, retrieval, and filtering; semantics and social computing; natural language processing; the semantic web; personalization. He has been the principal investigator of University of Bari in several European, national, and regional projects. He is author of more than 400 publications in international journals, conference and workshop proceedings, as well as of 3 books, including the textbook ”Semantics in Adaptive and Personalized Systems: Methods, Tools and Applications” published by Springer.