International Workshop on
Affective Computing for
Mental Wellbeing
Challenges, Opportunities, and Promising Synergies
The prevalence of mental illness is globally on the rise, leading to high individual and societal burdens. Many people in need do not seek professional help, the ones that do are confronted with limited availability of practitioners, and despite considerable progress, the success rate of therapies remains limited. While more and more digital health solutions are emerging, their adoption and reported success are still low.
Affective computing can support several important processes in enhancing and maintaining wellbeing, such as supporting self-awareness by tracking states and behavior using multimodal cues related to verbal and non-verbal communication, physiology, and activities. At the same time, clinicians can use such objective measures to complement traditional questionnaires and improve diagnosis, as well as in developing personalized interventions based on individuals' affective states and their unique needs and preferences. XR solutions or m-health apps can be enhanced by becoming more responsive and adaptive. Last but not least, such technology can make mental health services more accessible, especially to those facing barriers such as geographic distance, stigma, or lack of resources.
This workshop brings together researchers in Affective Computing (AC), clinicians in the emerging area of digital mental health and digital psychiatry, developers from industry, and policymakers to discuss what aspects of digital mental health apps and tools can most benefit from AC technologies and existing technologies already incorporating AC, such as embodied conversational agents and affective virtual agents, and affect-adaptive human-machine interaction. Since advances in AC and AI in mental health also give rise to important ethical concerns, this workshop aims to identify and address these emerging issues. Topics include the therapeutic relationship with technology and synthetic relationships, more broadly, artificial empathy, transparency (deception), safety, and affective privacy, among others.
This workshop seeks original contributions, including but not limited to:
We invite submissions in the following formats:
All submissions should be in pdf format following the ACII submission guidelines , using the conference Latex template . Each paper will be sent to at least two expert reviewers and will have one of the organizers assigned as editor. Papers are to be submitted via ACII’s EasyChair platform by selecting the “Workshop: Affective Computing for Mental Wellbeing” track.
Accepted submissions will be published as ACII workshop proceedings. At least one author must register for the conference (any registration type) and present. A single registration may cover up to three papers.
Selected submissions will be considered for publication in a special issue of JMIR Mental Health.
All deadlines are set at 23:59 PDT (GMT-7)
| Submission | |
| Notification to authors | 9 June 2023 |
| Camera-ready | 1 August 2023 |
| Workshop | 10 September 2023 |
| 09:00 | Welcome note |
| 09:10 | Keynote
Multimodal Machine Learning and Human Centered Computing for Mental Health and Wellbeing Akane Sano |
| 09:55 | Coffee Break |
| 10:15 | Keynote Challenges and Opportunities for Speech-based Mental Health Analysis Nicholas Cummins |
| Paper presentations | 11:00 | Context-aware EEG-based perceived stress recognition based on emotion transition paradigm Jiyao Liu, Lang He, Zhiwei Chen, Ziyi Chen, Yu Hao, and Dongmei Jiang | 11:15 | BERSting at the screams: Recognition of shouting and distress from mobile phone recordings Paige Tuttosi and Angelica Lim |
| 11:30 | Investigating self-supervised learning for predicting stress and stressors from passive sensing Harish Haresamudram, Jina Suh, Javier Hernandez, Jenna Butler, Ahad Chaudhry, Longqi Yang, Koustuv Saha, and Mary Czerwinski |
| 11:45 | Multimodal fusion improves BERT-based online suicidal ideation detection Noah Jones, Erik Kastman, Kelly Zuromski, Daniel Kessler, Matthew Nock and Rosalind Picard |
| 12:00 | Lunch break |
| 13:00 | Keynote Beyond Telehealth: Advances in Digital Mental Health Research and Practical Clinical Considerations for Smartphone Apps in Care John Torous |
| Paper presentations | |
| 13:45 | You go first: The effects of self-disclosure reciprocity in human-chatbot interactions Emmelyn Croes, Marjolijn Antheunis, and Linwei He |
| 14:00 | Mindfulness based stress reduction: A randomised trial of a virtual human, teletherapy, and a chatbot Mariam Karhiy, Mark Sagar, Mike Antoni, Kate Loveys, and Elizabeth Broadbent |
| 14:15 | Investigating psychological and physiological effects of forest walking: A machine learning approach Bhargavi Mahesh, Andreas Seiderer, Michael Dietz, Elisabeth Andre, Joachim Rathmann, Jonathan Simon, Christoph Beck, and Yekta Said Can |
| 14:30 | Social performance rating during social skills training in adults with autism spectrum disorder and schizophrenia Kana Miyamoto, Hiroki Tanaka, Jennifer Hamet Bagnou, Elise Prigent, Celine Clavel, Jean-Claude Martin, and Satoshi Nakamura |
| 14:45 | Towards successful deployment of wellbeing sensing technologies: Identifying misalignments across contextual boundaries Jina Suh, Javier Hernandez Rivera, Koustuv Saha, Kathy Dixon, Mehrab Bin Morshed, Esther Howe, Anna Kawakami, and Mary Czerwinski |
| 15:00 | Coffee Break |
| 15:30 | Keynote Why Ethics is More Important than Ever for Affective Computing in Behavioral and Public Health David Luxton |
| 16:15 | Group discussion What are the near-future actions we must take to ensure the ethical and effective application of affective commuting in behavioral health care? |
Multimodal Machine Learning and Human Centered Computing for Mental Health and Wellbeing
Abstract. Mobile health systems with sensors and computing enable non-disruptive monitoring of daily life behaviors and responses, aiding real-time interventions. Combining diverse and multimodal measurements like clinical and remote sensing data holds potential to predict and manage health issues. Despite progress in psychiatry and other fields, challenges persist, and overcoming challenges is crucial. In this talk, I will address these challenges and showcase progress and future directions for measuring, predicting, and supporting mental health and wellbeing. Specifically, I will highlight the significance of robust machine learning models that harness unlabeled data and data augmentation, the potential of leveraging social graph networks, and the development of adaptive and diverse sensing and interpretable feedback systems.
Bio Akane Sano is an Assistant Professor at Rice University, Department of Electrical Computer Engineering, Computer Science, and Bioengineering. She directs the Computational Wellbeing Group and is a member of Rice Digital Health Initiative. Her research includes data science, machine learning, and human-centered intelligent systems for health and wellbeing and spans the field of affective computing, ubiquitous and wearable computing, and biobehavioral sensing and analysis/modeling. She has been developing tools, algorithms, and systems to measure, forecast, understand, and improve health and wellbeing using multimodal data from mobile and wearable devices in daily life settings, and clinical assessment. She received her Ph.D. at the Massachusetts Institute of Technology and her M.Eng. and B.Eng. at Keio University, Japan. Her recent awards include the NSF Career Award, the Best of IEEE Transactions on Affective Computing 2021, the Best Paper Award at IEEE BHI 2019 conference, and the Best Paper Award at the NIPS 2016 Workshop on Machine Learning for Health.
Challenges and Opportunities for Speech-based Mental Health Analysis
Abstract. Speech is a unique and rich health signal: no other signal contains its singular combination of cognitive, neuromuscular and physiological information. However, its highly personal and complex nature also means that there are several significant challenges to overcome to build a reliable, useful and ethical tool suitable for widespread use in mental health research. This presentation will first describe how our voice is a tacit communicator of our health. Then, it will present an overview of current state-of-the-art in the prediction of mental health using speech. It will end by highlighting future challenges in relation to the translation of speech analysis into clinic practise.
Bio. Nicholas (Nick) Cummins is a lecturer in AI for speech analysis for health at the Department of Biostatistics and Health Informatics at King’s College London. Nick’s current research interests include speech processing, affective computing and multisensory signal analysis. He is fascinated by the application of machine learning techniques to improve our understanding of different health conditions and mental health disorders in particular. Nick is actively involved in a range of research projects including RADAR-AD and the DARPA Computational Cultural Understanding (CCU) programme.
Nick was awarded his PhD in electrical engineering from UNSW Australia in February 2016 for his thesis ‘Automatic assessment of depression from speech: paralinguistic analysis, modelling and machine learning’. After completing his PhD, he was a postdoctoral researcher at the Chair of Complex and Intelligent Systems at the University of Passau, Germany. Most recently, he was a habilitation candidate at the Chair of Embedded Intelligence for Health Care and Wellbeing at the University of Augsburg, also in Germany. During his time in Germany, he was involved in the DE-ENIGMA, RADAR-CNS, TAPAS and sustAGE Horizon 2020 projects. He also wrote and delivered courses in speech pathology, deep learning and intelligent signal analysis in medicine.
Beyond Telehealth: Advances in Digital Mental Health Research and Practical Clinical Considerations for Smartphone Apps in Care
Abstract. As the use of telepsychiatry in mental health via video/phone visits soars, it is important to also consider how asynchronous telepsychiatry tools like smartphone apps can also advance care. This talk will focus on the evolving field of smartphone digital phenotyping and consider the potential of real-time data capture via smartphones, methods necessary to analyze such data, and practical clinical applications of these tools. Looking at the evolving smartphone mental health ecosystem, the talk will also cover the topic of app evaluation and supporting research for making informed choices related to smartphone apps for use in research or patient care.
Bio. John Torous, MD MBI is director of the digital psychiatry division, in the Department of Psychiatry at Beth Israel Deaconess Medical Center (BDIMC), a Harvard Medical School affiliated teaching hospital, where he also serves as a staff psychiatrist and assistant professor. He has a background in electrical and computer sciences and received an undergraduate degree in the field from UC Berkeley before attending medical school at UC San Diego. He completed his psychiatry residency, fellowship in clinical informatics, and master's degree in biomedical informatics at Harvard. Dr. Torous is active in investigating the potential of mobile mental health technologies for psychiatry and has published over 250 peer reviewed articles and 5 book chapters on the topic. He directs the Digital Psychiatry Clinic at BIDMC which seeks to improve access to and quality of mental health care through augmenting treatment with digital innovations. Dr. Torous serves as editor-in-chief for the journal JMIR Mental Health, web editor for JAMA Psychiatry, and currently chairs the American Psychiatric Association’s Health IT Committee.
Why Ethics is More Important than Ever for Affective Computing in Behavioral and Public Health
Abstract. Affective computing is vital to the future of behavioral and public health. For example, affective computing detects and quantifies emotional states and behaviors, predicts behavior through multimodal data analysis, and drives appropriate health interventions. Affective computing enhances human interaction and engagement with machines, including virtual care providers. Affective computing is also an emerging feature of generative AI with possible applications in healthcare. This talk addresses emerging ethics topics specific to affective computing in behavioral and public health, focusing on risks lacking the needed attention to date. Topics include the therapeutic relationship with technology, synthetic relationships, and artificial empathy. The issues of transparency (deception), safety, and affective privacy, are also discussed, along with other topics that present significant risks to these fields and society. Specific considerations for the type of technology are discussed, including digital mental health apps, web-based tools, embodied conversational agents, and affective virtual agents.
Bio. David D. Luxton, Ph.D., M.S., is a nationally recognized expert, consultant, and trainer in suicide prevention, telehealth, and innovative technologies in behavioral healthcare. He is Affiliate Professor in the Department of Psychiatry and Behavioral Sciences at the University of Washington School of Medicine in Seattle and founder of Luxton Labs, LLC.. Dr. Luxton previously served as a Research Health Scientist and a Research Psychologist and Program Manager for the U.S. Department of Defense. Dr. Luxton's research and writing focus on the development and evaluation of innovative technologies in healthcare with specializations in artificial intelligence and ethics, telehealth, military and veterans’ health, and forensic psychology. He has served on numerous state-level and national workgroups and committees, including for the U.S. Department of Defense, Department of Veterans Affairs, and IEEE. He’s authored five books and more than 100 scientific articles and helped develop state and national-level guidelines and standards in artificial intelligence, telehealth, and forensic psychology. Dr. Luxton is also a licensed clinical psychologist and veteran of the United States Air Force.
