School of Science, Engineering and Information Technology

Smartphone Usage Index (SPUI): Non-invasive smartphone use monitoring to assess cognitive impairment

Project Title:

Smartphone Usage Index (SPUI): Non-invasive smartphone use monitoring to assess cognitive impairment.

Supervisor(s):

Dr Giles Oatley, A/Prof Andrew Stranieri, Dr David Stratton, Prof Sidney Morris

Contact person and email address:

Dr Giles Oatley, g.oatley@federation.edu.au

A brief description of the project:

Current cognitive assessment tools require a level of involvement from the participant that makes them intrusive and difficult to repeat regularly. For example, the Montreal Cognitive Assessment (MoCA - Nasreddine et al., 2005), widely used to distinguish mild cognitive impairment from healthy ageing requires participants to connect letters and numbers, draw a clock, draw a cube and recall words and numbers in sequence.  MoCA takes many minutes to administer and repeated administration for assessments over time is challenging for health care professionals in many health care settings (Wu, Wang, Cheng, & Chien, 2019).

Moreover, the association of cognitive assessment with motor skills, is detrimental and renders such assessment invalid where motor skills have declined in advance of mental skills . We distinguish between and assess two very different perspectives, the “micro”, dexterity with the phone (mobile device use cases), and the “macro”, actually using the phone for specific “life use cases” (as distinct to phone use cases). It is felt that mobile device use cases are very likely to change, as the technology rapidly evolves, but that life use cases are what you use the device for, which are not going to change – contacting people, event-related (social, food-related, family, health) and so on. The “life use cases” were thought to be the ones that would show degradation in use/completion when impairment develops, e.g. withdrawal from social events/interaction, indicated by less use of phone, as opposed to just slower or clumsier use of the same phone.

Ismail, Rajji and Shulman (2010) surveys a number of brief cognitive impairment tools. While the brief tools correlate well with longer assessments of cognitions, they are almost invariably very coarse in their gradation of cognition. Further, they must be administered by a healthcare professional on one patient at a time making routine, regular assessments difficult. (Wu et al., 2019) also notes that they have been designed for Western patients and are not well suited to others. Consequently assessment of cognitive ability was not a routine part of sickness assessment for Aged Care residents or hospital patients and motivated the emergence of a new Australian guidelines that recommend frequent assessment for early intervention (Laver et al., 2016).

Lumsden et al (2016) review literature on the use of computer games to conclude that careful application of gamification can lead to valid cognitive assessments.  Zhou et al (2018) address the need to integrate the assessment of motor skills with cognitive assessments by using wearable ankle movement sensors to advance a more holistic assessment of frailty. Moore, Swendsen and Depp, (2017) review literature on self administered mobile cognitive assessment tools to conclude that these are promising new ways to assess cognitive decline. These are typically smartphone Apps, periodically invoked, that present the user with an on-screen task to perform where performance metrics are associated with one aspect of cognitive decline.

The current study takes the recent initiatives toward alternate ways to assess cognition, one step further with the suggestion that normal smartphone use can be associated with cognitive assessments. The hypothesis is that, as smartphones become more and more widely adopted, a feasible indicator of cognitive ability would changes in the way a person's smartphone use changes. This might be characterised as a Smartphone Usage Index (SPUI) that correlates with validated cognitive assessment tools.

The SPUI would be implemented in a way that is non-invasive in that it is not determined  by having the patient play a game or use an App but executes behind the scenes, gathering data on the use of the smartphone.  As such, cognitive assessments are performed continuously, non-invasively without usage-related issues because there is no user interface, without inconveniencing the patient and without requiring healthcare professional resources.