Project

Back to overview

Personalized management of low back pain with mHealth: Big Data opportunities, challenges and solutions

Applicant Riener Robert
Number 167302
Funding scheme NRP 75 Big Data
Research institution Mobile Health Systems Lab Department of Health Sciences and Technology ETH Zürich
Institution of higher education ETH Zurich - ETHZ
Main discipline Information Technology
Start/End 01.03.2017 - 30.09.2021
Approved amount 593'798.00
Show all

All Disciplines (5)

Discipline
Information Technology
Rehabilitation
Occupational Medicine, Ergonomy
Health
Methods of Epidemiology and Preventive Medicine

Keywords (13)

Low Back Pain; Occupational Health; Exercise; Exergaming; Adaptive modeling; Serious Games; Translational medicine; Prevention; Wearable Sensors; Mobile Health; Physiotherapy; Crowdsourcing; Big Data

Lay Summary (German)

Lead
Wissenschaftler der ETH Zürich und der Universitätsspitäler Zürich arbeiten mit der Medizingeräteindustrie zusammen, um sich via Smartphone mit Tausenden von Rückenschmerzbetroffenen zu vernetzen. Ziel ist das Krankheitsbild Rückenschmerzen besser zu verstehen und mithilfe mobiler Gesundheitstechnologien neue Lösungsansätze für Therapie und Prävention zu entwickeln.
Lay summary

Eine mHealth-App soll Freiwillige (sogenannten "citizen scientists") wie ein Tagebuch bei der Bewertung und Dokumentation ihrer Rückenschmerzen unterstützen. Auf diese Weise können wichtige Daten einer grossen Gruppe von Personen gesammelt werden, die Auskunft darüber geben, wie häufig Schmerzen auftreten und individuell erlebt werden. Daraus werden wir neue Modelle entwickeln, um verschiedene Behandlungsstrategien (Physiotherapie, Gesundheitserziehung, sensorgestützte Smartphone-Spiele) zur Vorbeugung oder Linderung von Rückenschmerzen für den individuellen Fall computergestützt auszuwählen. Anschliessend werden die im Projekt beteiligten Mediziner untersuchen, welche Behandlungsarten erfolgreich sind und warum.

Nahezu jede in der Schweiz lebende Person leidet im Laufe ihres Lebens mindestens einmal an Rückenschmerzen – einer vielfältigen, komplexen und bis heute noch nicht vollständig verstandenen Erkrankung. Teure chirurgische Eingriffe sollten vermieden werden. Angesichts der vielfältigen Ausprägungen dieser Krankheit ist es für die Ärzte nicht immer leicht, die richtige Behandlungsmethode zu finden. Ein wichtiger Lösungsschritt ist die Beobachtung des Verhaltens vieler verschiedener Menschen, die an Rückenschmerzen leiden oder daran erkranken. Vernetzte digitale Technologien können hierbei gute Dienste leisten.

Ziel des Projekts ist der Einsatz mobiler Gesundheitstechnologien (mHealth-Tools), um Veränderungen bei Personen mit Rückenschmerzen zu sammeln und daraus Software-Modelle zu entwickeln und erproben, die individualisierte Präventionsmassnahmen vorschlagen können. Die entsprechende Datenbasis soll per App in der Klinik und zu Hause erfasst werden. Ausserdem werden wir neue Strategien für die Erfassung und Verwaltung sensibler Gesundheitsdaten und die Implementierung entsprechender Analyse-Software entwickeln.

Direct link to Lay Summary Last update: 26.07.2017

Lay Summary (French)

Lead
Des chercheurs universitaires collaborent avec l’industrie biomédicale pour se connecter à milliers de personnes suisse souffrant de lombalgies via leurs smartphones. Le projet vise à mieux comprendre les douleurs lombaires et à développer de nouvelles solutions préventives et thérapeutiques au moyen des technologies mobiles.
Lay summary

Une application mHealth attrayante permettra aux gens d’évaluer et de documenter leurs douleurs lombaires. Nous pourrons ainsi collecter des données importantes provenant d’un large groupe de volontaires, ce qu’on appelle des "citoyens scientifiques". Cela nous apprendra à quelle fréquence les lombalgies se manifestent et mettra en lumière l’expérience individuelle des personnes touchées. Nous développerons ensuite de nouvelles stratégies expérimentales afin de tester différentes méthodes de prévention et de réduction des lombalgies, telles que la physiothérapie, l’éducation et des jeux mobiles pour la santé. Des médecins travaillant avec des petits groupes d’utilisateurs de l’application évalueront leur. Afin de déterminer quel traitement est efficace et pourquoi, nous allons aussi chercher à savoir comment choisir la méthode la plus adaptée à chaque utilisateur sur la base de sa situation personnelle.Les lombalgies affectent presque tous les individus en Suisse au moins une fois dans leur vie. Les causes de cette maladie complexe ne sont pas entièrement connues. Bien que le recours, coûteux, à la chirurgie ne représente pas une option désirable, la nature variée de cette affection fait que les médecins ont de la peine à proposer un traitement efficace. Une étape clé en vue de résoudre ce problème est de surveiller les comportements individuels lors des épisodes de lombalgie, une approche rendue possible par le recours aux technologies numériques comme les smartphones, les capteurs portables et Internet.L’objectif du projet est d’utiliser des nouvelles techniques de santé mobiles (mHealth) pour déceler les changements au niveau des lombalgies chez les patients en Suisse. Nous développerons des logiciels pour personnaliser la prévention et thérapie des douleurs lombaires en recourant à un large ensemble de données enregistrées au moyen d’applications pour smartphones que possèdent de nombreuses personnes, souffrant ou non de lombalgie, chez elles ou à l’hôpital.


Direct link to Lay Summary Last update: 26.07.2017

Lay Summary (English)

Lead
University researchers in Zurich are collaborating with the medical device industry to connect to thousands of Swiss people with low back pain via their smartphones. The idea is to improve understanding of low back pain and to develop new solutions for therapy and prevention using mobile health technologies.
Lay summary

An engaging mHealth app will enable people to assess and document their low back pain. This will allow us to collect important data from a large group of volunteers – so-called citizen scientists – that will tell us how frequently low back pain occurs and also shed light on people’s individual experience of low back pain. We will then develop new experimental strategies to test different methods of preventing or reducing low back pain, such as physiotherapy, education and mobile games for health. Doctors working with small groups of app users will then evaluate these methods for effectiveness. In addition to determining which treatments work and why, we will also investigate how to select the most suitable method for each user based on his or her personal life situation.

Low back pain affects almost all individuals in Switzerland at least once at some point in their lives. It is a complex illness whose cause is not fully understood. Costly surgery is undesirable, but the variable nature of the illness makes it challenging for doctors to find effective treatment. A key step in solving the problem is to monitor the behaviour of many individuals during episodes of low back pain. Digital technologies such as smartphones, wearable sensors and the internet make this approach possible.

The objective of the project is to use novel mobile health (mHealth) tools for identifying changes in low back pain among people in Switzerland. We will develop software models for personalizing ways to prevent low back pain using a large set of data recorded with smartphone apps from many people with and without low back pain at home and in the clinic. We will also introduce novel strategies for making health data meaningful, developing analysis software and handling this sensitive health data.

Direct link to Lay Summary Last update: 26.07.2017

Responsible applicant and co-applicants

Employees

Project partner

Publications

Publication
Ethics review of big data research: What should stay and what should be reformed?
Ferretti Agata, Ienca Marcello, Sheehan Mark, Blasimme Alessandro, Dove Edward S., Farsides Bobbie, Friesen Phoebe, Kahn Jeff, Karlen Walter, Kleist Peter, Liao S. Matthew, Nebeker Camille, Samuel Gabrielle, Shabani Mahsa, Rivas Velarde Minerva, Vayena Effy (2021), Ethics review of big data research: What should stay and what should be reformed?, in BMC Medical Ethics, 22(1), 51-51.
Exergaming Using Postural Feedback From Wearable Sensors and Exercise Therapy to Improve Postural Balance in People With Nonspecific Low Back Pain: Protocol for a Factorial Pilot Randomized Controlled Trial
Meinke Anita, Peters Rick, Knols Ruud, Karlen Walter, Swanenburg Jaap (2021), Exergaming Using Postural Feedback From Wearable Sensors and Exercise Therapy to Improve Postural Balance in People With Nonspecific Low Back Pain: Protocol for a Factorial Pilot Randomized Controlled Trial, in JMIR Research Protocols, 10(8), e26982-e26982.
Injury risks among elite competitive alpine skiers are underestimated if not registered prospectively, over the entire season and regardless of whether requiring medical attention
Fröhlich Stefan, Helbling Moritz, Fucentese Sandro F., Karlen Walter, Frey Walter O., Spörri Jörg (2021), Injury risks among elite competitive alpine skiers are underestimated if not registered prospectively, over the entire season and regardless of whether requiring medical attention, in Knee Surgery, Sports Traumatology, Arthroscopy, 29(5), 1635-1643.
A Deep Learning Approach to Diagnosing Multiple Sclerosis from Smartphone Data
Schwab Patrick, Karlen Walter (2021), A Deep Learning Approach to Diagnosing Multiple Sclerosis from Smartphone Data, in IEEE Journal of Biomedical and Health Informatics, 25(4), 1284-1291.
Granger-causal Attentive Mixtures of Experts: Learning Important Features with Neural Networks
Schwab Patrick, Miladinovic Djordje, Karlen Walter (2019), Granger-causal Attentive Mixtures of Experts: Learning Important Features with Neural Networks, in AAAI Conference on Artificial Intelligence, AAAI, Honolulu.
PhoneMD: Learning to Diagnose Parkinson's Disease from Smartphone Data
Schwab Patrick, Karlen Walter (2019), PhoneMD: Learning to Diagnose Parkinson's Disease from Smartphone Data, in AAAI Conference on Artificial Intelligence, AAAI, Honolulu.
Developing a smartphone app for tracking digital biomarkers of low back pain
MeinkeAnita, SchwabPatrick, KarlenWalter (2018), Developing a smartphone app for tracking digital biomarkers of low back pain, Swiss Society for Biomedical Engineering, Biel.
Hierarchy of Convolutional Recurrent Attentive Neural Networks.
Schwab Patrick, Khashkhashi Moghaddam, Kareln Walter (2017), Hierarchy of Convolutional Recurrent Attentive Neural Networks., in 10th Annual RECOMB/ISCB Conference on Regulatory & Systems Genomics with DREAM Challenges, New York, USASage Bionetworks, Seattle, USA.
Beat by Beat : Classifying Cardiac Arrhythmias with Recurrent Neural Networks
Schwab Patrick, Scebba Gaetano, Zhang Jia, Delai Marco, Karlen Walter (2017), Beat by Beat : Classifying Cardiac Arrhythmias with Recurrent Neural Networks, in Computing in Cardiology, RennesIEEE, Rennes, France.

Collaboration

Group / person Country
Types of collaboration
Balgrist University Hospital Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure
- Exchange of personnel
- Industry/business/other use-inspired collaboration
University Hospital Zurich Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure
- Exchange of personnel
- Industry/business/other use-inspired collaboration
Hocoma AG Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Exchange of personnel
- Industry/business/other use-inspired collaboration
Rheumaliga Schweiz Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Industry/business/other use-inspired collaboration
SwissSki Switzerland (Europe)
- Industry/business/other use-inspired collaboration

Scientific events

Active participation

Title Type of contribution Title of article or contribution Date Place Persons involved
AAAI Conference on Artificial Intelligence 2020 Talk given at a conference Learning Counterfactual Representations for Estimating Individual Dose-Response Curves 13.03.2020 Philadelphia, United States of America Karlen Walter; Schwab Patrick;
Thirty-third Conference on Neural Information Processing Systems Talk given at a conference CXPlain: Causal Explanations for Model Interpretation under Uncertainty 08.12.2019 Vancouver, Canada Karlen Walter; Schwab Patrick;
AAAI Conference on Artificial Intelligence Talk given at a conference Granger-causal Attentive Mixtures of Experts: Learning Important Features with Neural Networks 27.01.2019 Honoloulu, United States of America Schwab Patrick;
Annual Meeting of the Swiss Society for Biomedical Engineering (SSBE) Poster Developing a smartphone app for tracking digital biomarkers of low back pain 21.08.2018 Biel, Switzerland Meinke Anita; Karlen Walter;
35th International Conference on Machine Learning, {ICML} Talk given at a conference Not to Cry Wolf: Distantly Supervised Multitask Learning in Critical Care 10.07.2018 Stockholm, Sweden Schwab Patrick;
BigHeart Research Seminars Individual talk Point-of-care decision making using mobile sensing and big data 22.06.2018 Singapore, Singapore Karlen Walter;
Gordon Research Conference on Advanced Health Informatics Talk given at a conference Point-of-Care Decision Making Using Mobile Sensing and Big Data Approaches 18.06.2018 Hongkong, Hongkong Karlen Walter;
Advanced Health Informatics, Gordon Research Conference, Convergence: Integrating Informatics with Sensing Imaging and Robotics for Health Poster Design of mHealth tools for personalized management of low back pain 17.06.2018 HongKong, Hongkong Karlen Walter;
10th Annual RECOMB/ISCB Conference on Regulatory & Systems Genomics with DREAM Challenges Talk given at a conference Automated Extraction of Digital Biomarkers for Parkinson's Disease using A Hierarchy of Convolutional Recurrent Attentive Neural Networks 21.11.2017 New York, United States of America Schwab Patrick;
Computing in Cardiology Talk given at a conference Beat by Beat: Classifying Cardiac Arrhythmias with Recurrent Neural Networks 24.09.2017 Rennes, France Schwab Patrick;


Knowledge transfer events

Active participation

Title Type of contribution Date Place Persons involved
Annual Meeting of the Swiss Society of Intensive Care Medicine Talk 19.09.2018 Interlaken, Switzerland Schwab Patrick;
medidays Talk 10.09.2018 Zurich, Switzerland Karlen Walter;
Interpretable and Sample Efficient Deep Learning Talk 19.07.2018 Böblingen, Germany Schwab Patrick;


Communication with the public

Communication Title Media Place Year
Media relations: print media, online media Digital den Rücken fordern NZZ am Sonntag German-speaking Switzerland 2017
Talks/events/exhibitions Scientifica German-speaking Switzerland 2017
Media relations: radio, television Scienze in Vetrina RSI Telegiornale Italian-speaking Switzerland 2017

Associated projects

Number Title Start Funding scheme
136454 Mobile Health Monitoring for Remote Locations of the Developing World 01.12.2011 Fellowships for advanced researchers

Abstract

Low back pain (LBP) is a common condition that has a large socio-economic impact on Switzerland and throughout the European countries. The current treatment strategy frequently involves analgesic medication, potentially followed by costly surgery. There is a strong need to find effective LBP prevention strategies to avoid surgery and to predict which citizens are most at risk in developing severe LBP with the aim to reduce morbidity, and consequently cost. A major challenge in finding such a strategy is the high diversity in LBP occurrence, duration and underlying causes, as well as between the individuals affected, calling for subject specific solutions. Aggregating data, increasing the number of individuals that are monitored before, during, or after LBP episodes, including contextual, behavioural and lifestyle information, and linking them to medical outcomes are the key steps to developing personalized treatment and prevention strategies. The objective for this project is to establish and use a new framework for continuously monitoring health and disease progression with new and innovative mobile computing solutions in a large population of Swiss individuals who suffer from LBP or are at high risk for developing LBP. The framework will also allow for the interaction with individuals to deliver health interventions and simultaneously measure direct and indirect outcomes. We aim to use Big Data aggregated from a combination of mobile apps that promote instrumented self-assessment and exercises performed by individuals at home as well as intensive physiotherapy to develop models for personalizing LBP prevention. This project brings together computer and health sciences from ETH Zurich and application experts from medicine and physiotherapy (The Balgrist and Zurich University Hospital), and the medical device industry (Hocoma AG) to solve technical and clinical challenges. With an innovative combination of crowdsourcing health data and conventional clinical study designs, a new rich data set for modeling LBP disease paths will be generated. Challenges of reliable data collection, data security and privacy concepts in digital health and adaptive modeling for personalization of models will be addressed. This project will offer unexplored Big Data solutions to overcome the barrier of conventional clinical trials that are limited to small datasets from homogenous patient groups. Based on these new tools, new knowledge on management of LBP using personalized approaches will be obtained.
-