Personal Informatics Systems and the Integration of Data from Novel Sensor Technologies, 2021, Dudley Ph.D. Thesis

Discussion in 'General ME/CFS news' started by Dolphin, Oct 2, 2021.

  1. Dolphin

    Dolphin Senior Member (Voting Rights)

    Messages:
    5,792
    https://researchportal.bath.ac.uk/e...-systems-and-the-integration-of-data-from-nov

    https://researchportal.bath.ac.uk/files/225871655/CillianDudleyThesis.pdf


    Personal Informatics Systems and the Integration of Data from Novel Sensor Technologies


    Student thesis: Doctoral Thesis › PhD

    Abstract

    Personal Informatics (PI) systems are used by people to record, track and analyse data about themselves.

    The increasing availability and affordability of sensing technologies, which is likely to continue into the future, provides people with the ability to record and analyse an ever increasing amount of data about themselves.

    Primarily, research relating to Personal Informatics seeks to understand how people are using currently available, widely adopted sensing technologies, or how specific, mainstream sensing technologies might be used to solve a specific problem.

    Currently, there is little understanding of how people attempt to make sense of data from newly emerging sensing technologies; those which do not have widespread adoption, or whose data is less understood by the general population.

    Given the rapid expansion of sensing capabilities, the research described in this thesis provides an important understanding of how people use emerging sensing technologies to learn more about themselves, as well as the challenges and opportunities these technology present in relation to self-understanding.

    Throughout our research we used a NeuroSky brain-computer interface as a technology probe, representing a novel sensing technology, and developed a feature-rich multifaceted PI system for allowing users to engage with data. Our findings stem from qualitative analysis of participant interviews in one exploratory study and two in-the-wild studies of these novel sensing technologies.

    Additionally, we explore aspects of trust in novel sensing technologies.

    We present quantitative and qualitative analysis of participant responses from a lab-based study designed to explore how perceptions of trust are shaped.

    We present several design considerations and challenges for developing PI systems that integrate novel sensing technologies.

    These considerations and challenges are based around two groups of users with distinct tracking behaviours: those tracking for documentary purposes - to gather data that documents their lives - and those with a specific life-aspect that motivates data tracking, such as those living with chronic health conditions or symptoms.

    Specifically, we explore the impact of novel sensing technologies for those with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), prolonged fatigue, chronic fatigue (CF), and idiopathic chronic fatigue (ICF).

    We additionally present four design considerations for the development of features which enable users to interrogate their data, using automatically generated insights and predefined analyses.

    We highlight that trust in devices may be shaped by users' preconceived initial levels of trust in a device, rather than on specific feedback from the device or visualisations of the data that it generates.

    Date of Award 8 Sep 2021
    Original language English
    Awarding Institution
    • University of Bath
    Supervisor Simon Jones (Supervisor) & Eamonn O'Neill (Supervisor)
    Keywords
    • personal informatics
    • quantified self
    • human-computer interaction
    • eHealth
    Cite this
    • Standard
    Personal Informatics Systems and the Integration of Data from Novel Sensor Technologies
    Dudley, C. (Author). 8 Sep 2021
    Student thesis: Doctoral Thesis › PhD
     
    Simon M, Barry, MSEsperanza and 4 others like this.
  2. MSEsperanza

    MSEsperanza Senior Member (Voting Rights)

    Messages:
    2,947
    Location:
    betwixt and between
    Last edited: Oct 3, 2021
    Barry, Sean and Peter Trewhitt like this.
  3. Trish

    Trish Moderator Staff Member

    Messages:
    55,414
    Location:
    UK
    I have looked through the section:
    Chapter 5 Making Sense of Novel Sensor Data Within The Context of Fatigue
    Starting on page 126.
    https://purehost.bath.ac.uk/ws/portalfiles/portal/225871655/CillianDudleyThesis.pdf

    There were 9 participants, most with ME/CFS, a few with chronic fatigue.
    They were provided with the EEG and fitbit kit, delivered to their homes, and used them for 3 weeks, with sessions with the researcher via Skype for technical support and feedback.

    The general feeling seems to be that the EEG was a complete washout, not giving them any useful information, and supposedly interpreting their EEG patterns in terms of mood and cognitive activity levels that didn't in many case bear any relation to reality.

    The Fitbit heart rate monitoring, resting heart rate over time, and step monitoring over time were seen by some as useful for pacing, but they recommended much longer time spans to get a clear picture of changes over a suggested 2 years rather than a few weeks.

    They make it clear that target setting wasn't appropriate in ME/CFS as it can lead to overdoing activity and setbacks.

    This is a study of how users find wearable technology, whether it's useable, interpretable and useful in guiding behaviour. The focus was therefore more on things like design of interfaces, interpretablity, time scales, immediate accessiblity of data as well as longer term interpretation, and the importance to users that there's research to back up the validity and reliability of the data they are getting. In that sense, I think it's a useful study for designers of wearables.

    In particular the EEG based one seems to me to be a complete non-starter, providing misinterpretation of data that can't measure what it purports to be measuring and that just confuses users and would horribly mislead any psychotherapist who got hold of such data about a client. This I think makes it very questionable on ethical grounds.
     
    ringding, Keela Too, Lilas and 6 others like this.
  4. ringding

    ringding Senior Member (Voting Rights)

    Messages:
    415
    Location:
    Bristol, UK
    I completely missed this at the time. Thanks for linking to the thread @Trish .
     
    Trish and Peter Trewhitt like this.

Share This Page