Pre-Conference Workshops

Pre-Conference Workshops are included in your ICAMPAM Registration fee.

Compositional Data Analysis – Hands on Demonstration

Sebastian Chastin and Dr. Elisabeth Winkler

Glasgow Caledonian University, Glasgow, UK

This will be an interactive workshop during which the participants will learn the foundation of compositional analysis and go through a set of practical examples giving then the opportunity to apply the techniques on realistic examples. Participants will be provided with quizzes, a workbook, example data and software resources. The workshop will guide the participants through a full data analysis process through a series of exercise and practical demonstrations, starting with how to prepare and process the data. This will be followed by how to analyse the day activity composition in terms of time spent in different behaviours and produce statistical and graphical summary and finally conduct linear modelling of the relationship between the activity composition, health outcomes and determinants.

NIH Gait Lab Tour and Demo

Diane Damiano and Christopher Stanley

National Institutes of Health Clinical Center, Maryland, USA

The Functional & Applied Biomechanics Section in the Rehabilitation Medicine Department of the National Institutes of Health is a state-of-the-art intramural research laboratory located within the largest research hospital in the world (the NIH Clinical Center).  The tour will highlight some of our research activities that focus on the quantification and remediation of functional human movements in individuals with movement disorders mainly as a result of neurological injuries such as cerebral palsy or traumatic brain injury.  To this end, we combine biomechanical instrumentation including but not limited to a 10 camera passive infrared motion capture system, force plates, instrumented treadmill, wireless EMG system, and a dynamic body-weight support system with emerging mobile functional brain imaging technologies (EEG and fNIRS) to investigate the neural mechanisms of normal and abnormal motor coordination. We further design and test novel interventions to improve motor capabilities such as our pediatric wearable lower extremity exoskeleton to train children over time to improve their gait patterns and efficiency. We will also demonstrate the capability of commercially available Virtual Reality gaming systems for accurate, accessible and affordable motion tracking.

Functional Data Analysis for Wearables: Methods and Applications

Vadim Zipunnikov, Jeff Goldsmith and Ciprian Crainiceanu

Johns Hopkins Bloomberg School of Public Health, Baltimore, USA

The main goal of this workshop is to present an overview of the functional data analysis methods for modeling physical activity data, review their strengths and limitations, demonstrate their implementation in R packages “refund” and “mgcv” and applications in large epidemiological studies such as Head Start Program, National Health and Nutrition Examination Survey, Baltimore Longitudinal Study of Aging, and Women’s Health Initiative.

Sensor Methods Collaboratory

Richard Troiano

National Cancer Institute, Maryland, USA

The purpose of the SMC pre-conference workshop is to reinvigorate the efforts of the SMC and generate potential next steps in moving the SMC forward. Dr. Richard Troiano will present an overview of the SMC’s mission and initial plans. Additionally, SMC group leads will provide information regarding the content and potential of individual SMC working groups. The speakers will then facilitate a discussion intended to encourage brainstorming as well as address three important issues. First, as the NHANES 2011-2014 data are prepared for public release, what is the preferred method to process the data? Second, what strategies can be implemented to incentivize and facilitate data sharing and collaboration among researchers focused on ambulatory monitoring? Third, how can large, richly characterized data files be used to develop or benchmark analytical methods (algorithm development)?

Enhancing the ambulatory assessment tool box

Ulrich Ebner-Priemer, Karlsruhe Institute of Technology, Germany
Hans Bussmann, Erasmus MC – University Medical Center, Rotterdam, The Netherlands

To better understand Physical Behaviour in daily life, information about the context of the performed physical behaviour is sometimes essential. In this sense, it is helpful to define context as broad as possible, covering nature and the built environment, the social environment including social interactions, as well as affective and motivational states. Fortunately, the advent of mobile smartphone technology enables an effortless monitoring of a variety of context information in daily life. Combining the assessment of Physical Behaviour with location via GPS, physiological effort via ECG, social interactions via tracking communication behaviour of the smartphone, and motivational states via e-diaries, might lead to an increased insight into everyday life Physical Behaviour.

In the proposed workshop we will exemplarily report studies combining the assessment of Physical Behaviour and context information. We will report on how to assess a multitude of parameters, including location, green space, electrocardiogram, communication behaviour, online behaviour, cognitive status, affective and motivational. In addition, we will introduce the concept of “triggered” diaries. E-diaries are triggered by physical activity, physiology, location or urban context to get additional context information during episodes of interest. All in all, we will work on expanding our tool-boxes to get insight into daily life behaviour.

Grasping Physical Activity: Using 3D printers to visualise physical activity

Melitta McNarry, Applied Sports, Technology, Exercise and Medicine Research Centre, Swansea University, UK
Kelly Mackintosh
, College of Engineering, Swansea University, UK

Current physical activity (PA) guidelines recommend that children should engage in at least 60 minutes moderate-to-vigorous physical activity (MVPA) every day, yet few children meet these guidelines. A frequently cited barrier leading to this short-fall in PA is that guidelines are difficult to measure, interpret and apply. Recent research has not only developed a novel device to accurately measure PA, but allows real-time visualisation. Moreover, the integration of 3D printing facilitates the creation of a tangible output, which provides children with a novel and exciting way to conceptualise their PA levels in accordance with government guidelines. However, it goes far beyond just PA; a 3D printed model can provide a wealth of information regarding numerous specific movements and how well executed they are.

We would like to conduct an interactive session to present the real-time monitoring and visualisation aspects of our work. This session will involve participants wearing the monitors to collect live data that is simultaneously displayed on the screen and subsequently printed in the form of a 3D shape. The greater the range of movements, the more interesting the shape! This will then be used to discuss key public health messages regarding PA and our current research. Furthermore, the process that went into developing the coding for the models and the choice of model shapes will be discussed. This is a contemporary issue that everyone has an opinion on; the innovative approach will capture the imagination of the participants as they try to generate unique personalised 3D shapes that can be kept as a souvenir. Those interested in Engineering, Physiology, Physical Activity, Health, Psychology and Computer Science will be intrigued by this holistic approach.

Data privacy - from ethics to compliance and trust

Joss Langford, Technical Director, Activinsights, Kimbolton, UK
Mark Lizar
, CEO, Open Consent Group
Tom Torkar
, Partner, Michelmores

The process of running free-living intervention studies, the expanding reach of European data privacy law and the desire to make more use of consumer wearables are driving us beyond the remit of ethics alone into the areas of compliance and trust.

We increasingly need a deep technical understanding of how personal data is being collected and stored in order to be compliant with the consumer regulatory regimes and to talk sensitively with subjects about how their data is being used. The aims of the workshop are:

  • To develop a forum for the discussion of commercial/consumer privacy, security and identity issues within ISMPB.
  • To establish the current state of understanding and appetite for learning on the topic of data privacy.
  • To inform ISMPB members, workshop and conference attendees of the current status of the law and industry practice in the area of global data protection with particular focus on Europe and US.
  • To inform ISMPB members, workshop and conference attendees of the upcoming changes in European law (General Data Protection Regulations, May 2018) and the likely impacts for them. For example: consent, processing notices, privacy-by-design, privacy impact assessments, right to be forgotten, data portability and system architecture design.
  • To discuss the changing subject/consumer attitudes to the collection and use of their data in research, products and services.
AlphaBET: Consensus meeting on the development of a taxonomy of 24 hour activity cycle data.

Jorunn Helbostad, NTNU

Sebastien Chastin, Glasgow Caledonian University, Ghent University

Malcolm Granat, Salford University

Richard Troiano, NIH

Mark Tremblay, CHEO

 

The AlphaBET project aims to develop a consensus common taxonomy (naming and cataloging) for classification, harmonisation and storage of the 24 hour activity cycle data recorded with objective sensors. It is an open science project hosted on the Open Science Framework (https://osf.io/vsauf/) and initiated by a steering committee of 12 experts representing different fields of research dealing with sensor data and researchers in scientific ontology in an attempt to pull together multiple initiatives aiming at the same goal (Sebastien Chastin,  Jorunn Helbostad, Mark Stephen Tremblay,  Barbara Ainsworth, Paul Jarle Mork, Lynn Rochester, Ulf Schwarz, Johannes Bussmann, Malcolm Granat, Rick Troiano, Neville Owen, Jacqueline Kerr)

The steering committee established a protocol and initial framework.

The goals of this workshop are

  1. to present this initial work to the wider community
  2. to engage the wider community in an informed and structured discussion
  3. to give a voice to the wider community and benefit from the wider expertise
  4. to advance the development and produce a consensus taxonomy through a concept mapping exercise with the ICAMPAM community

The workshop will follow a set protocol, combining live consensus making via electronic concept mapping with discussions around short expert statement.

Objective Measures of Free-Living Physical Behaviour: What can they tell us about physical capacity in persons with mobility limiting conditions?

Douglas Maxwell, Founder and CEO, PAL Technologies, Glasgow, UK
Andy Kerr, Biomedical Engineering Department, PAL Technologies
Nicholas Smith, Associate, Biomedical Engineering Department, PAL Technologies
David Loudon, Software Engineer and Data Visualisation Expert, PAL Technologies

The main goal of this workshop is to discuss how we can best derive person-centred outcomes that reflect physical capacity from body-worn senor data. From our study of physical behaviours we know people have basic free-living physical demands. We can look at these in terms of ability (for example rising from a chair) and connect this with participation (how often a person stands up). In terms of mobility, we know people make short duration stepping bouts around the house for self-care and domestic activities but community activities require longer bouts of stepping. We have calculated a Walking Breaks Index (WBI) to compare upright periods with short and long stepping bouts and have used this index to quantify impairment in persons with mobility limiting conditions. Similarly, we have demonstrated how the time from standing up to stepping varies in clinical populations. We will look at how measure like the WBI and time to first step can be used as a measure of physical capacity and how they can be generated from wearable sensors in a range of clinical populations including people with stroke and vascular disease (claudication) We will explore the advantages of combining sensor signals (accelerometers, gyroscopes, magnetometers and barometers) to provide contextual information about behaviours. We will also explore novel ways of visualising physical behaviour information can both complement and inform analysis for both clinician and patient.

Participants will take away an appreciation of how the measurement of physical behaviour can provide objective outcomes related to physical capacity from patients in a free-living environment and how appropriate visualisation of these measures can provide an insight into and record of a patient’s recovery profile.

Novel analytics, signal processing and exercise science

Gareth Stratton, A-STEM Research Centre Swansea University, UK

Huw Summers, Swansea University, UK

Rowan Brown, Associate Professor, Swansea University, UK

Michael Rosenberg, Associate Professor, Swansea University, UK

In this workshop delegates will be exposed to novel analytics hitherto used in cell biology and animal behaviour both individually and in groups. Further insight into how these analytics can be integrated into hand held devices and apps.

While accelerometry is the de facto standard for objectively measuring physical activity, emerging analytical approaches used in other disciplines may offer new insight into traditional analyses. The number of sensors and analytics in physical activity measurement in humans has grown significantly. Yet the approaches used to analyse and characterise the range of signals has lagged behind. There are rich data available from the signals captured by sensors that can be used to quantify and qualify physical activity. The problem is creating manageable outputs from large amounts of data. This workshop will demonstrate how analytics hitherto used in animal and cell behaviour can be used to process physical activity data for individuals and cluster these analyses to gain insight into group behaviour.

Pooling of daily activity data across borders and cultures

Tamara Harris, National Institute on Aging, Intramural Program
Eric Shiroma
, National Institute on Aging, Intramrual Research Program
Charles Matthews
, National Cancer Institute, Intramural Research Program
Annemarie Koster
, Maastricht University, Dept. of Social Medicine
Sari Stenholm
, Dept of Preventive Medicine, Tempe University

Accelerometer data may be a strong candidate for pooled research projects.  However, difference in devices used, wear location and data collection and cleaning methods present challenges to combining data across studies.  This workshop is an exploration of a novel method for pooling accelerometry data, using daily activity patterns.  The first goal is to ring researchers from a variety of population studies in Europe, Asia, and the U.S. to compare data on daily activity arrayed as total counts by hour.  These data will be generated prior to the workshop and submitted to the organizers for further analysis.  Individual  study data will be reviewed at the workshop.  The second goal is to discuss statistical and analytical challenges to combining data across these studies to assess the consistency of the pattern of daily activity.  The third goal is to present these methods and share them with the attendees of the workshop.  This workshop is designed to be interactive.  There will be dedicated time for discussion about submitted data, analytical approaches and future projects.