Skip to content

Analysis of facial expression parameters recorded by video on a mobile device for the diagnosis of Parkinson's disease for people living in rural and remote communities. A comparative study between people with early to mid-stage Parkinson's disease and age-matched healthy volunteers.

A comparative clinical study of analysis of facial expressions recorded by video for the diagnosis of Parkinson's disease for people living in rural and remote communities. A comparative study between people with early to mid-stage Parkinson's disease and age-matched healthy volunteers.

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
ANZCTR
Registry ID
ACTRN12626000636358
Enrollment
82
Registered
2026-05-26
Start date
2024-05-17
Completion date
2026-11-20
Last updated
2026-06-01

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

None listed

Brief summary

Individuals suffering Parkinson's disease (PD) often exhibit the symptom of hypomimia, known as facial masking. This symptom is characterised as difficulty in changing facial expression in response to different emotions and is evident as a secondary sign of PD in the early stages of the disease process. Analysis of facial expressions recorded via video is a non-invasive and promising diagnostic tool for the early detection of PD. This study proposes to use expression analysis recorded as a video utilising computer vision and machine learning techniques as a non-invasive method to identify Parkinson's disease. Data collection for the PhD project was completed December 2024 for this project with a final reported submitted and approved by RMIT HREC 24/10/2024. Data collection continues as part of the broader project which includes: ACTRN12623000220662 A comparative clinical study of a handwriting app for the diagnosis of Parkinson's disease for people living in rural and remote communities. A comparative study between people with early to mid-stage Parkinson's disease and age-matched healthy volunteers ACTRN12623000219684 A comparative clinical study of a voice recording app for the diagnosis of Parkinson's disease for people living in rural and remote communities. A comparative study between people with early to mid-stage Parkinson's disease and age-matched healthy volunteers.

Interventions

Analysis of facial expressions recorded via video is a non-invasive and promising diagnostic tool for the early detection of PD. This study proposes to use expression analysis recorded as a video utilising computer vision and machine learning techniques as a non-invasive method to identify Parkinson's disease. Up to eight facial feature tasks will be video recorded. The researcher will record videos in person of people with Parkinson's disease performing 8 or less facial tasks and upload the vid

Analysis of facial expressions recorded via video is a non-invasive and promising diagnostic tool for the early detection of PD. This study proposes to use expression analysis recorded as a video utilising computer vision and machine learning techniques as a non-invasive method to identify Parkinson's disease. Up to eight facial feature tasks will be video recorded. The researcher will record videos in person of people with Parkinson's disease performing 8 or less facial tasks and upload the video records to Google Drive, whereas the healthy people will conduct the task themselves online and upload the videos through Google Forms. This study will accept patients diagnosed with Parkinson's disease via Goulburn Valley Health, Dandenong Neurology, and the Monash Health clinic. In association with Parkinson's Australia the research team will recruit People living with PD (PLwPD) who are living in the community but not attending one of the clinics that are associated with current recruitment sites. In this case participants will either complete the facial expression tasks in a community setting or in their own home. Up to 8 of the following tests will be conducted: • the syllable “pa” as quickly as possible in a single breath (PA); • the syllable “pataka” as quickly as possible in a single breath (PATAKA); • five repetitions of pretending to blow a candle (BLOW); • five repetitions of pretending to kiss a baby (KISS); • five repetitions of maximum opening of the jaw (OPEN); • five repetitions of pretending to smile with tight lips (SPREAD), • five repetitions of making a big smile (BIGSMILE); • five repetitions of raising the eyebrows (BROW). The duration of the recording session is less than 5 minutes and, when people living with PD (PLwPD) participate in the trial from their own home, only one facial expression is required - three repetitions of making a big smile. The approximate time for completing this task is less than one minute. However, the task needs to be completed twice a day, once in the morning after taking PD medication and once in the afternoon, six or more hours after taking the PD medication. The trial continues over a 4 week period and participants are requested to complete the task every day over this period.

Sponsors

RMIT University
Lead SponsorUniversity

Eligibility

Sex/Gender
All
Age
18 Years to No maximum
Healthy volunteers
No

Inclusion criteria

Participants who have been diagnosed with mild to moderate Parkinson's disease and age matched healthy volunteers.

Exclusion criteria

Cognitive impairment as evidenced by the Montreal Cognitive Assessment (MoCA) which will be completed by registered health professionals of participating clinics; history of stroke; unable to provide informed consent. healthy volunteers self report that they do not suffer with a cognitive impairment, have a history of stroke, and are able to provide informed consent.

Outcome results

None listed

Source: ANZCTR · Data processed: Jun 11, 2026