INdividual SIGnature mHealth Technology

A multidisciplinary project to design and develop mobile, wearable and physical computing technologies, to gather hard to get data, from hard to reach groups and places, in realistic settings and in real time.

Case Study 1 – System

 Towards an Ecological Model of Self-harm – System

Developing an eco-technological system to research people’s experiences and perspectives through smart-phone based digital diary methods. The digital diary gives participants the flexibility of deciding when to record their thoughts as well as how; text, audio, photos or video. The digital diary is hosted and presented as a secure ‘blog’. These data are enriched with further sources that are combined to build the ecological model. These sources include:

– Customisable and remotely updatable in-app questionnaires

– Location tracking using the smartphone app

– Sleep length/quality and daytime activity achieved by integrating commoditised wearable computing (Jawbone ‘Up’)

– Heart beat activity to provide heart rate variability over given timeframe and heat rate at given instances using bespoke heart beat data logger developed for this project.

Case Study 1 – App

Towards an Ecological Model of Self-harm – iPhone app

The iPhone App has two principal roles:

Digital Diary

Participants are asked to make a diary entry in the morning and evening. Each diary entry is preceded by a series of questions that have conditionality depending on the time of day and the ongoing participant response. The questions are configured as check-box, likert scale sliders or free text entry. The questions are generated from an XML file, which enables easy updating and modification of content.

The answers to these questions are uploaded to a secure server for later analysis.

Participants conclude the diary entry with a free text account of events, thoughts and feelings since the last diary entry. This account can be substituted or embellished with audio, photographs and video.

The account is uploaded to a private server and presented to the participant as a ‘blog’.

Location tracking

The iPhone app tracks the location of the participant. The granularity of this tracking can be adjusted from ‘significant location change’ (approx 500m) to maximum GPS resolution (approx) 5m. The location tracking data is uploaded to a secure server for later analysis each time the participant uploads a diary entry.

Location tracking can be disabled within the app by the participant, allowing privacy when desired.

Case Study 1 – Heart Beat Data Logger

Towards an Ecological Model of Self-harm – Heat Beat Data Logger

There are strong indications in the literature that identify heart rate variability (HRV) as a reliable biomarker for stress. Obtaining reliable data for assessing levels of HRV over an extended period (3 weeks in this project), outside of the laboratory in naturalistic settings and using non-invasive methods is a non-trivial task. Researchers have no commercially available technologies to get these data under these conditions.

A bespoke heart beat data logger was developed for this project. The data logger writes a time-stamp to a micro SD card each time the participant’s heart beats. Each time-stamp is analogous to the ‘R’ beat of the participant and can be aligned in time and space with the other collected data in subsequent analyses. The R-R intervals in the resultant data can be used to calculate HRV at any given point and for any given time period of the participant in the study. These data can also be used to calculate heart rate.

The heart beat data logger uses a conventional un-coded chest strap. This sensing method enables low artefact data to be collected in a low-invasive and power efficient way. The R beat pulses from the chest strap are received by the logger, time-stamped and written to the micro SD card.

The heart beat data logger is charged from a USB level charger or a USB port. It can run continuously for approx. 72 hours between charges and can continue to log data during a charge cycle.

Case Study 1 – Data Visualisation

Towards an Ecological Model of Self-harm – Data Visualisation

A number of heterogenous datasets are collected during the study:

1.Participant diaries

– Answers to given questions, with conditional branching, on likert scales and check lists
– Free text
– Audio recordings
– Video recordings
– Photographs

2. Participant location tracking with set granularity ranging from significant location change (approx. 500m) to small location change (max GPS resolution approx. 5m).

3. Activity data collected from a Jawbone ‘UP’.

4. Sleep length and quality data collected from a Jawbone ‘UP’.

5. Heart beat (R beat) timestamps.

Bespoke data visualisation tools were developed for this project to align the data-sets in time and space for each participant. The visualisations are interactive, customisable and expandable. They enable the data to be explored in a case by case, episode driven manner, resulting in insights into significant data features, concomitant features in each case and recurring concomitants across cases.

These episodic insights then feed into multi-level modelling analysis.

Contact Information

If you are interested in any aspect of this work and ongoing projects then please contact us. We are actively seeking collaborators, partners and networks.

Dr. Lisa Marzano – Applied psychology and mental health
Dept. of Psychology Middlesex University

Dr. Andy Bardill – Innovation management, product/interaction design and technology development
Director of redLoop:the mdx design and innovation centre

Dr. Bob Fields – Human-computer interaction and computer-supported cooperative work
Dept. of Computer Science Midlesex University

Dr. Kate Herd – User experience and product/interaction design
Assoc. Director of redLoop:the mdx design and innovation centre