1st Workshop on Tools and Algorithms for Mental Health and Wellbeing, Pain, and Distress (MHWPD)
http://mhw.media.mit.edu, mhw@media.mit.edu
Abstract:
We organize a workshop in the field of affective health computing, focusing on detection and intervention techniques for mental health and well-being, pain and distress. We invite contributions from researchers with multidisciplinary expertise (computer science, engineering, psychology and medicine), both in academia and industry, in the following domain:
Distress - e.g. pain, panic, confusion, itching - in patients with restricted communicative verbal abilities such as neonates and children, somnolent patients and patients with dementia is difficult to diagnose. For example, the subjectively experienced pain may be partly or even completely unrelated to the somatic pathology of tissue damage and other disorders. Therefore, the clinically used methods of distress assessment do not allow for objective and robust measurement, and physicians must rely on the patient’s report regarding the quality and intensity of the distress. Common tools are verbal scales, which are restricted to patients with normal mental abilities. However, there are procedures for distress assessment available for people with verbal and/or cognitive impairments and scales for pain assessment in people who are sedated and require ventilation. Overall, these diagnostic methods have limited reliability, validity or are very time consuming. If valid measurement of distress is not possible, treating the negative affect may lead to cardiac stress in risk patients and over- or under-usage of medical treatment. There are several efforts to create an automatic system for recognizing distress through different kind of modalities and machine learning techniques.
Mental health and wellbeing are one of the most challenging issues of the modern society. For instance, depression is growing worldwide: by 2020 one suicide will happen every 20 seconds, and by 2030 it will be the #1 disease burden. Other typical causes of poor mental health and wellbeing are high levels of stress and anxiety, sleep deprivation, and loneliness due to impoverished social communication. Also, chronic mental illnesses such as schizophrenia, and neurodevelopmental disorders such as autism, if not monitored and treated timely, can lead to further degradation of the person’s mental health and wellbeing. Most of existing methods and algorithms for monitoring and providing feedback to individual’s mental health and wellbeing have been built/evaluate using (limited) data captured in highly constrained settings (e.g., labs). This can limit the applicability and reliability of such tools and algorithms when applied in every-day situations. The goals of this part of the workshop are (i) to explore practical research challenges and opportunities for designing new methods, algorithms and applications for affect/mood measurement/prediction in every-day life or clinical settings, and (ii) to introduce novel target tools and algorithms and discuss the directions on how the design of these should be tackled in the future.
The special focus will be on (but it is not limited to):
Session1: Automated recognition of pain & distress and treatment
Session2: Practical Tools and Algorithms for Mental Health and Wellbeing
Workshop Program
14:00 Welcome remarks
14:05 Paper Presentations - pain and distress (12min
presentation + 3 min Questions & Answers)
Automatic Recognition of Pain, Anxiety, Engagement and Tiredness for Virtual Rehabilitation from Stroke: A Marginalization Approach
Jesus J Rivas; Lorena Palafox; Jorge Hernandez Franco; Carmen Lara; Nadia Berthouze; Felipe Orihuela-Espina; Luis E. Succar
Analysis of Facial Expressiveness During Experimentally Induced Heat Pain
Philipp Werner; Ayoub Al-Hamadi; Steffen Walter
Multi-task Neural Networks for Personalized Pain Recognition from Physiological Signals
Daniel Lopez Martinez; Rosalind Picard
15:00 Paper Presentations - mental health and wellbeing (12min presentation + 3 min Questions & Answers)
Analysis of Phonetic Markedness and Gestural Effort Measures for Acoustic Speech-Based Depression Classification
Brian Stasak
Differential Performance of Automatic Speech-Based Depression Classification Across Smartphones
Brian Stasak
15:30 Break
16:00 Continue
Detection of Universal Cross-Cultural Depression Indicators from the Physiological Signals of Observers
Josephine F Plested; Tom Gedeon; Xuanying Zhu; Abhinav Dhall; Roland Goecke
Robot Models of Mental Disorders
Matthew Lewis; Lola Canamero
16:30 Keynote speech: Professor Santosh Kumar (The University of Memphis, Director NIH Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K))
Title: mSelf – Mobile Sensor Big Data Challenges in Monitoring and Improving Health, Wellness, and Performance
Abstract: The increasing availability of mobile sensors that allow collection of raw sensor data, along with mobile big data software platforms that allow labeled collection, curation, modeling, and visualization of such data for development and validation of new markers and sensor-triggered interventions, is opening up exciting new research directions. They include novel sensor systems for self-tracking of health, wellness, and performance. This talk will introduce examples of such sensors and mobile sensor big data computing platforms developed by multi-university Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K), to highlight the mobile computing, data modeling, and big data computing research challenges emerging in such mobile sensor systems. It will also expose the audience to unique validation challenges facing new sensor-based markers and sensor-triggered interventions that must work for each individual in their natural field environment and deliver high-utility with minimal burden. The talk will conclude with a vision of mobile self (mSelf) and summarize multidisciplinary research needed to successfully realize such a vision.
Bio: Santosh Kumar is a Lillian & Morrie Moss Chair of Excellence Professor in Computer Science at the University of Memphis. His research focuses on using mobile sensors for self-monitoring of health, wellness, and performance. He and his students have developed computational models to infer human health and behaviors such as stress, conversation, smoking, craving, and cocaine use from wearable sensor data. He leads several large multidisciplinary projects in mobile sensors funded by National Institutes of Health (NIH), National Science Foundation (NSF), and IARPA that involve 25 investigators in computing, engineering, behavioral science, and medicine from 17 universities. Santosh was named one of America’s ten most brilliant scientists under the age of 38 by Popular Science in 2010. In 2015, he was named Tennessee’s first chair of excellence in Computer Science.
17:10 Panel Discussion with presenters, organizers and public
17:25 Closing Remarks
Program Committee
Ayoub Al-Hamadi (Magdeburg)
Adriano Andrade (Uberlandia)
Elisabeth Andre (Augsburg)
Min Hane Aung (Cornell)
Sascha Gruss (Ulm)
Roland Goecke (Canberra)
Jongwha Kim (Seoul)
Mriam Kunz (Groningen)
Friedhelm Schwenker (Ulm)
Amir Muaremi (Stanford)
Pablo Paredes (Stanford)
Anton Batliner (Munchen)
Saeed Abdullah (Cornell)
Daniel McDuff (Microsoft Research)
Submission Format
Long (up to 8 pages), Short papers (up to 4 pages) (ACII paper format, double blind review)
Workshop proceedings will be published by IEEE Xplore.
Submission Website
Please create a CMT account at: https://cmt3.research.microsoft.com/ACII2017/ and select: Create a new submission (Practical Tools and Algorithms for Mental ...)
Important Dates
Workshop paper submission deadline: July 20, 2017 [Extended!!]
Acceptance notification: August 11, 2017
Camera Ready Papers Due: August 18, 2017
Date of workshop: October 23, 2017
Location
San Antonio, TX
Organizers
Akane Sano (MIT)
Steffen Walter (Ulm, Germany)
Ognjen (Oggi) Rudovic (MIT)
Nadia Bianchi-Berthouze (UCLIC - UCL Interaction Centre, UK)
Björn Schuller (Imperial College London/University of Passau)
Rosalind W. Picard (MIT)