Machine Learning for Mobile Health
Workshop at NeurIPS 2020
Submission Deadline: 16 Oct AOE (Anywhere On Earth)
Author Notification: 30 Oct
Workshop: 12 Dec, 10am - 5:30pm EST
Link to Gather.Town: https://neurips.gather.town/app/fSShkGpRlyZHkNZk/ML4MH
Please note that the gather.town is linked to NeurIPS, so you must have an active NeurIPS active in order to enter.
Link to Neurips website for pre-recorded & live zoom sessions: https://neurips.cc/virtual/2020/protected/workshop_16135.html
Mobile health (mHealth) technologies have transformed the mode and quality of clinical research. Wearable sensors and mobile phones provide real-time data streams that support automated clinical decision making, allowing researchers and clinicians to provide ecological and in-the-moment support to individuals in need. Mobile health technologies are used across various health applications such as improving HIV medication adherence, increasing activity levels, and reinforcing abstinence in addictions, among others. The development of mobile health technologies, however, has progressed at a faster pace than the science and methodology to evaluate their validity and efficacy.
Current mHealth technologies are limited in their ability to understand how adverse health behaviors develop, how to predict them, and how to encourage healthy behaviors. In order for mHealth to progress and have expanded impact, the field needs to facilitate collaboration among machine learning researchers, statisticians, mobile sensing researchers, human-computer interaction researchers, and clinicians. Techniques from multiple fields must be brought to bear on the substantive problems facing this interdisciplinary field: experimental design, causal inference, multi-modal complex data analytics, representation learning, reinforcement learning, deep learning, transfer learning, data visualization, and clinical integration.
This workshop will assemble researchers from the key areas in this interdisciplinary space necessary to better address the challenges currently facing the widespread use of mobile health technologies.
We invite contributions of up to 4-page extended abstracts (not counting references or optional supplementary materials) from a broad range of areas including but not limited to:
Machine Learning Methods for mHealth
Detection or prediction of adverse health behaviors
Predictions for wearable devices or other longitudinal data
Transfer learning in mHealth
Deep learning in mHealth
Decision making and reinforcement learning
Just-in-time adaptive interventions and sequential decision making
Application of (partially observable) Markov decision processes to mHealth
Experimental design for mobile health such as micro-randomized trials
Methods for effective communication of information to both users and clinicians
Application of mobile health in the real-time monitoring, diagnosis, and treatment of disease
Mobile health technology that uses sensor data, delivers interventions, or integrates with clinical practice
Patient-facing clinical applications to disease areas such as chronic disease and mental health
Clinical integration of information from wearable devices
Digital tools to integrate social determinants into patient and population health management
Extended abstracts should be submitted via CMT: https://cmt3.research.microsoft.com/ML4MH2020
Please use the standard NeurIPS style files (e.g. single column format): https://nips.cc/Conferences/2020/PaperInformation/StyleFiles. Review will be double blind, so please anonymize author identities along with clear references to an author's prior work.
Please note that the workshop is non-archival, so we welcome any submissions that may be under review at another conference or journal, as well as work that has been recently accepted elsewhere. We will request that accepted submissions be posted on this website before the start of the workshop, as long as that does not violate any rules regarding preprints for other venues.
If you have any questions please contact the workshop organizers at ML.firstname.lastname@example.org .
All Times are listed in EST. The session will run from 10AM EST to 5:30PM EST
10:00 - 10:10 AM: Introduction
Morning Session: Mobile Health in Action
10:10 - 10:30 AM: Matthew Nock
10:30 - 10:50 AM: Lee Hartsell
10:50 - 11:10 AM: Ally Salim Jr, "AI for Decision Support in Low Resource Areas"
Decision making is one of those extremely complex things that humans can do with relative ease most of the time. Healthcare providers do this hundreds and thousands of times per day, and do an amazing job given their various levels of expertise and the resources available to them. The Elsa Health Assistant is a set of tools and technologies that leverage advances in Artificial Intelligence and causal modeling to augment the capacity of lower cadre healthcare providers and support optimal and consistent decision making. Here we will share the challenges, failures and successes of the technologies and the team.
11:25 - 11:35 AM: Bret Nestor, "Using Wearables for Influenza-Like Illness Detection: The importance of design"
11:35 - 11:45 AM: Jeffrey Chan, "Representing and Denoising Wearable ECG Recordings"
11:45 - 11:55 AM: Qu Tang, "Towards Personal Hand Hygiene Detection in Free-living Using Wearable Devices "
12:00 - 12:10 PM: Q&A with Spotlight Speakers
12:10 - 12:40 PM: Panel Discussion with Invited Panelists
12:40- 1:20 PM: Poster Session
1:20 - 2:00 PM: Lunch/Networking Break (in Gather.Town or on your own)
Afternoon Session: ML methods in Mobile Health
2:00 - 2:20 PM: Susan Murphy, "Assessing Personalization in Digital Health"
Reinforcement Learning provides an attractive suite of online learning methods for personalizing interventions in a Digital Health. However after an reinforcement learning algorithm has been run in a clinical study, how do we assess whether personalization occurred? We might find users for whom it appears that the algorithm has indeed learned in which contexts the user is more responsive to a particular intervention. But could this have happened completely by chance? We discuss some first approaches to addressing these questions.
2:20 - 2:40 PM: Tanzeem Choudhury, "Closing the sensing-to-intervention loop for behavioral health"
Mobile and ubiquitous computing research has led to new techniques for cheaply, accurately, and continuously collecting data on human behavior that include detailed measurements of physical activities, mobility, social interactions, mood, sleep quality and more. Continuous and unobtrusive sensing of behaviors has tremendous potential to support the lifelong management of mental health by: (1) acting as an early warning system to detect changes in mental well-being, (2) delivering personalized interventions to patients when and where they need them, and (3) significantly accelerating patient and physician understanding of changes in mental health in real-time. In this talk, I will give some examples of our work on turning sensor-enabled mobile devices into well-being monitors and instruments for administering real-time/real-place interventions.
2:40 - 3:00 PM: Tim Althoff, "Language-based Behavior and Interventions in Mobile Health"
Mobile health seeks to provide in-the-moment support to individuals in need. In this talk, I will discuss the challenges associated with behavior and interventions that are based on language. Language is high-dimensional and complex, but is a critical component of many health and support interactions. Specifically, I will describe how we can measure empathy in mental health peer support and how we can give feedback in order to empower peer supporters to increase expressed levels of empathy, using large-scale neural transformer architectures and reinforcement learning.
3:15 - 3:25 PM: Kathy Li, "A generative, predictive model for menstrual cycle lengths that accounts for potential self-tracking artifacts in mobile health data"
3:25 - 3:35 PM: Ayse Selin Cakmak, "Using Convolutional Variational Autoencoders to Predict Post-Trauma Health Outcomes from Actigraphy Data"
3:35 - 3:45 PM: Marianne Menictas, "Fast Physical Activity Suggestions: Efficient Hyperparameter Learning in Mobile Health"
3:50 - 4:00 PM: Q&A with Spotlight Speakers
4:00 - 4:30 PM: Panel Discussion with Invited Panelists
4:30 - 5:15 PM: Poster Session
5:15 - 5:30 PM: Concluding Remarks (in Gather.Town)
A major goal of this workshop is to assemble a diverse set of researchers spanning interdisciplinary fields related to mobile health. The invited speakers are:
Susan Murphy (Harvard University, Statistics) Susan Murphy is Professor of Statistics at Harvard University, Radcliffe Alumnae Professor at the Radcliffe Institute, Harvard University, and Professor of Computer Science at the Harvard John A. Paulson School of Engineering and Applied Sciences. Her lab works on clinical trial designs and online learning algorithms for developing personalized mobile health interventions. She is a 2013 MacArthur Fellow, a member of the National Academy of Sciences and the National Academy of Medicine, both of the US National Academies.
Matthew Nock (Harvard University, Psychology) Application of mobile health and machine learning technologies to understanding, predicting, and preventing self-harm and suicidal behaviors.
Lee Hartsell (Duke University, Clinical Neuroimmunologist) Application of mobile health and machine learning techniques to multiple sclerosis.
Tanzeem Choudhury (Cornell Tech, Computing and Information Science) Tanzeem Choudhury is a Professor of Computing and Information Sciences at Cornell Tech where she holds the Roger and Joelle Burnell Chair in Integrated Health and Technology and a co-founder of HealthRhythms Inc, a company whose mission is to add the layer of behavioral health into all of healthcare. At Cornell, she directs the People-Aware Computing group and leads the Precision Behavioral Health Initiative. Tanzeem received her PhD from the Media Laboratory at MIT. She has been awarded the MIT Technology Review TR35 award, NSF CAREER award, TED Fellowship, Kavli Fellowship, ACM Distinguished Membership, and Ubiquitous Computing 10-year Impact Award. For more information, please visit: http://pbh.tech.cornell.edu
Tim Althoff (University of Washington, Computer Science) Tim Althoff is an assistant professor in the Paul G. Allen School of Computer Science & Engineering at the University of Washington. His research advances computational methods that leverage large-scale behavioral data to extract actionable insights about our lives, health and happiness through combining techniques from data science, social network analysis, and natural language processing. Tim holds Ph.D. and M.S. degrees from the Computer Science Department at Stanford University, where he worked with Jure Leskovec. Prior to his PhD, Tim obtained M.S. and B.S. degrees from the University of Kaiserslautern, Germany. He has received several fellowships and awards including the SAP Stanford Graduate Fellowship, Fulbright scholarship, German Academic Exchange Service scholarship, the German National Merit Foundation scholarship, a Best Paper Award by the International Medical Informatics Association, and the SIGKDD Dissertation Award 2019. Tim’s research has been covered internationally by news outlets including BBC, CNN, The Economist, The Wall Street Journal, and The New York Times.
Ally Salim Jr (Founder and CEO - Inspired Ideas) Ally Salim Jr is the Founder and CEO of Inspired Ideas, a company in Tanzania committed to improving lives by leveraging emerging technologies for social good. As an Artificial Intelligence enthusiast and practitioner, Ally has built a number of projects utilizing emerging technologies across a range of sectors, including healthcare, law, and education. He is the creator of Elsa Health, a set of technologies and tools that utilize artificial intelligence and expert knowledge to improve health outcomes in East Africa. He has expertise in software development, system architecture, mobile and web systems development, mathematics, and machine learning. He is passionate about empowering others through teaching and mentoring and appreciates the success of home-grown solutions for the African continent.
1: "Towards Personal Hand Hygiene Detection in Free-living Using Wearable Devices" Qu Tang (Northeastern University); Aditya Ponnada (Northeastern University); Stephen Intille (Northeastern University)
2: "Using Convolutional Variational Autoencoders to Predict Post-Trauma Health Outcomes from Actigraphy Data" Ayse S Cakmak (Georgia Institute of Technology); Nina Thigpen (Google X); Garrett Honke (Google X); Erick Perez Alday (Emory University); Ali Bahrami Rad (Emory University); Rebecca Adaimi (Google X); Chia-Jung Chang (Google X); Qiao Li (Emory University); Pramod Gupta (Google X); Thomas Neylan (University of California, San Francisco); Samuel McLean (School of Medicine, University of North Carolina); Gari Clifford (Emory University)
3: "A generative, predictive model for menstrual cycle lengths that accounts for potential self-tracking artifacts in mobile health data" Kathy Li (Columbia University); Iñigo Urteaga (Columbia University); Amanda Shea (Clue by BioWink); Virginia Vitzthum (Indiana University); Chris H Wiggins (Columbia University); Noémie Elhadad (Columbia); (supplement)
4: "Fast Physical Activity Suggestions: Efficient Hyperparameter Learning in Mobile Health" Marianne Menictas (Harvard University); Sabina J Tomkins (Stanford University); Susan Murphy (Harvard University)
5: "Representing and Denoising Wearable ECG Recordings" Jeffrey Chan (UC Berkeley); Andy Miller (Apple); Emily Fox (Apple)
6: "Using Wearables for Influenza-Like Illness Detection: The importance of design" Jaryd Hunder (The Hospital for Sick Children); Bret Nestor (University of Toronto); Raghu Kainkaryam (Evidation); Erik Drysdale (SickKids); Jeff Inglis (UCSB); Allison Shapiro (Evidation); Sujay Nagaraj (University of Toronto); Marzyeh Ghassemi (University of Toronto, Vector Institute); Luca Foschini (Evidation); Anna Goldenberg (University of Toronto)
7: "Multimodal Privacy-preserving Mood Prediction from Mobile Data: A Preliminary Study" Terrance Liu (Carnegie Mellon University); Paul Pu Liang (Carnegie Mellon University); Michal Muszynski (Carnegie Mellon University); Ryo Ishii (Carnegie Mellon University); David Brent (University of Pittsburgh); Randy P Auerbach (Columbia University); Nick Allen (University of Oregon); Louis-Philippe Morency (Carnegie Mellon University)
8: "Learning Generalizable Physiological Representations from Large-scale Wearable Data" Dimitris Spathis (University of Cambridge); Ignacio Perez-Pozuelo (University of Cambridge); Soren Brage (University of Cambridge); Nicholas J. Wareham (University of Cambridge); Cecilia Mascolo (University of Cambridge)
9: "The case of point-of-care diagnosis in fetal echocardiography" Arijit Patra (University of Oxford)
10: "Exploring Contrastive Learning in Human Activity Recognition for Healthcare" Chi Ian Tang (University of Cambridge); Ignacio Perez-Pozuelo (University of Cambridge); Dimitris Spathis (University of Cambridge); Cecilia Mascolo (University of Cambridge)
11: "Passive detection of behavioral shifts for suicide attempt prevention" Pablo Moreno-Munoz (Universidad Carlos III Madrid); Lorena Romero-Medrano (Universidad Carlos III de Madrid); Angela Moreno (Universidad Carlos III de Madrid); Jesús Herrera-López (Universidad Carlos III de Madrid); Enrique Baca-García (Hospital Universitario Fundación Jiménez Díaz); Antonio Artés Rodríguez (Universidad Carlos III de Madrid)
12: "Unstructured Primary Outcome in Randomized Controlled Trials" Bruno Jedynak (Portland state university); Daniel Taylor rodriguez (Portland State University); David Lovitz (Portland State University); Hiroko Dodge (University of Michigan); Chao-Yi Wu (OHSU); Nora Mattek (OHSU); Jeffrey Kaye (OHSU)
13: "TrueImage: A Machine Learning Algorithm to Improve the Quality of Telehealth Photos" Kailas Vodrahalli (Stanford University); Roxana Daneshjou (Stanford University); Roberto Novoa (Stanford University); Albert Chiou (Stanford University); Justin Ko (Stanford University); James Zou (Stanford University)
14: "Bayesian Hierarchical Vector Autoregression Models for Health Label and Mobile Sensor-based Behavioral Feature Prediction" Ethan N Lyon (Rice University); Akane Sano (Rice University); Luis H Victor (Rice University)
15: "Deep Transfer Learning for Automated Diagnosis of Skin Lesions from Photographs" Doyoon Kim (Cleveland High School); Emma Rocheteau (University of Cambridge)
16: "DeepCompliance: Predicting prompt-level EMA compliance" Supriya Nagesh (Georgia Institute of Technology); Alexander F Moreno (Georgia Institute of Technolog); Stephanie Carpenter (University of Michigan); Jamie Roslyn T Yap (University of Michigan); Soujanya Chatterjee (University of Memphis); Steven Lizotte (University of Utah); Neng Wan (University of Utah); Santosh Kumar (University of Memphis); Cho Lam (University of Utah); David Wetter (University of Utah); Inbal Nahum-Shani (University of Michigan); James Rehg (Georgia Institute of Technology)
17: "Personalized Machine Learning Models for Noninvasive Glucose Prediction Using Wearables" Brinnae M Bent (Duke University); Jessilyn Dunn (Duke University)
18: "Proximity Sensing: Modeling and Understanding Noisy RSSI-BLE Signals and Other Mobile Sensor Data for Digital Contact Tracing" Sheshank Shankar (PathCheck Foundation); Rishank Kanaparti ( PathCheck Foundation); Ayush Chopra (OpenFoodFacts); Rohan Sukumaran (PathCheck Foundation); Parth Patwa (PathCheck Foundation); Sunny Kang (PathCheck); Abhishek Singh (MIT); Kevin McPherson (PathCheck Foundation); Ramesh Raskar (MIT)
19: "Bayesian Active Learning for Wearable Stress and Affect Detection" Abhijith Ragav (Solarillion Foundation); Gautham Krishna Gudur (Ericsson)
20: "Federated Learning with Heterogeneous Labels and Models for Mobile Activity Monitoring" Gautham Krishna Gudur (Ericsson); Satheesh K Perepu (Ericsson)
21: "WiSense: WiFi Proximity Detection for Digital Contact Tracing" Mikhail Dmitrienko (PathCheck Foundation); Abhishek Singh (MIT Media Lab); Patrick Erichsen (PathCheck Foundation); Ramesh Raskar (MIT Media Lab)
22: "Deep Sequence Learning for Accurate Gestational Age Estimation from a $25 Doppler Device" Nasim Katebi (Emory University); Reza Sameni (Emory University); Gari Clifford (Emory University)
Walter Dempsey, University of Michigan (email@example.com). Walter Dempsey is an Assistant Professor of Biostatistics at University of Michigan. He completed his PhD at the University of Chicago, and was a postdoctoral fellow at Harvard University. He develops data analytic methods and experimental designs to inform sequential decision making in mobile health as well as Bayesian nonparametric methods for state-space and network data.
Nicholas Foti, Apple (firstname.lastname@example.org) Nick Foti is a research scientist at Apple. He completed his PhD at Dartmouth College, and was a postdoctoral fellow and then research scientist at the University of Washington. His research focuses on utilizing machine learning and statistical methods for health applications.
Joseph Futoma, Apple (email@example.com) Joseph Futoma is a research scientist at Apple. He completed his PhD at Duke University and was a postdoctoral fellow at Harvard University. He works in the intersection of machine learning, reinforcement learning, and healthcare, and has developed methods leveraging clinical time series data in both inpatient and outpatient settings.
Katherine Heller, Duke University & Google AI (firstname.lastname@example.org) is a research scientist at Google AI, and an Assistant Professor of Statistical Science at Duke University. She completed her PhD at the Gatsby Unit, at UCL, and was a postdoctoral fellow at the University of Cambridge and MIT. She develops Bayesian time series methods for making better medical predictions, both inside and outside a hospital setting.
Yian Ma, University of California, San Diego (email@example.com). Yian Ma is an assistant professor at Halıcıoğlu Data Science Institute, UC San Diego. He works on scalable inference methods and their theoretical guarantees, with a focus on time series data and sequential decision making. He has been developing new Bayesian inference algorithms for uncertainty quantification as well as deriving computational and statistical guarantees for them. Prior to his appointment at UCSD, he worked as a post-doctoral fellow at UC Berkeley. He obtained his Ph.D. degree at University of Washington and his bachelor’s degree at Shanghai Jiao Tong University.
Marianne Njifon, Google AI (firstname.lastname@example.org) Marianne Njifon is an AI Resident at Google Brain. She completed a master’s degree in Machine intelligence from the African Masters of Machine Intelligence (AMMI) center of Rwanda. Prior to this, she obtained a master’s degree in Nonlinear Optics from the University of Yaounde I, and a master’s degree in Mathematical sciences from the African Masters for Mathematical Sciences (AIMS) center of Cameroon. She is currently working on frequency analysis of neural network architectures and has a lot of interests in applications in health and computational biology.
Jiewru Shi, University of Michigan (email@example.com) Jieru Shi is a PhD student in biostatistics at the University of Michigan. She works on statistical methods related to mobile health. Her research is currently focused on causal inferences in micro-randomized trials.
Kelly Zhang, Harvard University (firstname.lastname@example.org). Kelly Zhang is a PhD student in computer science at Harvard University. She works on reinforcement learning methods for real world problems, particularly mobile health. Her research is currently focused on statistical inference methods on adaptively collected data (e.g. data collected with a bandit algorithm).