Development of Digital Diagnostic Devices for Parkinson's Disease

In this project, ocular motor, pupil and gait data in people with Parkinson's disease (PD) will be collected in order to develop machine learning models for the diagnosis and monitoring of PD. With this, the investigators aim to advance the state of the art in PD diagnosis and monitoring. By integrating the principles of machine learning with high-quality sensor data, more accurate and earlier diagnosis could potentially be achieved. Ocular motor and pupil data will be collected with the standard clinical examination and with neos, a medical device approved for objective ocular motor and pupil measurement. Gait will be collected using an IMU sensor and GaitQ senti, a consumer device that allows for an objective and continuous remote gait monitoring.

Study Overview

• Status: Recruiting
• Conditions: Parkinson Disease
• Intervention / Treatment: Diagnostic test: oculomotor and pupillary function assessment

Detailed Description

Parkinson's disease (PD) is one of the most common neurodegenerative diseases worldwide, affecting 1% of the population older than 65.

Currently, PD diagnosis is based on history, clinical assessments, and neurological examination.

The most widely used criteria for diagnosis are the Movement Disorder Society (MDS) criteria and instrument (i.e. The MDS-UPDRS). Further information may be gained from people's subjective description of their symptoms and/or via some short walking tests, such as 3-meter Timed Up and Go (TUG) performed as a snapshot in the clinic. However, people's symptoms vary through and between days and subjective descriptions rely on their memory and observations at home. These recollections can be unreliable or lack enough detail (particularly when the person has cognitive impairment). Therefore, current PD diagnosis criteria are highly dependent on the person and on the diagnosing physician. This subjectivity may lead to a variability in the diagnosis. Furthermore, these clinical assessments are unable to accurately track disease progression over time, making it difficult to provide personalized care. Additionally, manual examinations lack precise measurement instruments, resulting in a low precision of observed measurements and the inability to detect early-stage, subclinical signs. An objective diagnosis based on quantitative data rather than subjective interpretation of clinical findings is important2. Therefore, an early and accurate diagnosis of PD, as well as accurate disease progression monitoring, are still important challenges in PD.

Several oculo-visual abnormalities have been described in PD. Studies report an abnormal ocular motor function in 75-87.5% of people with PD (3,4). These dysfunctions may precede or follow motor symptoms and thus, the evaluation of ocular motor function may provide valuable information regarding early disease detection or disease progression (5). The most commonly reported ocular motor dysfunctions are impairments in saccades, smooth pursuit, and vergence (3,4,6).

Gait impairments are among the most common and disabling symptoms of PD (29). Gait impairments include freezing of gait (FOG), an inability to initiate or maintain normal walking patterns, often resulting in a stochastic stop/start gait, and festinating gait (FSG), which is a shortening of stride length with elevated step frequency, resulting in fast, shuffling steps. Both FOG and FSG contribute to an increased risk of falls (and fall-related injuries) in people with PD relative to the wider elderly population. Objective, and continuous remote gait monitoring would be highly important in people with PD, to objectively track gait impairments in real-time, and potentially contribute to objectively track disease progression, which may lead to personalized care for individuals with PD.

In this project, ocular motor, pupil and gait data in people with Parkinson's disease (PD) will be collected in order to develop machine learning models for the diagnosis and monitoring of PD. With this, the investigators aim to advance the state of the art in PD diagnosis and monitoring. By integrating the principles of machine learning with high-quality sensor data, more accurate and earlier diagnosis could potentially be achieved. Ocular motor and pupil data will be collected with the standard clinical examination and with neos, a medical device approved for objective ocular motor and pupil measurement. Gait will be collected using an IMU sensor and GaitQ senti, a consumer device that allows for an objective and continuous remote gait monitoring.

The primary objective of this project is to collect ocular motor, pupil and gait data from people with PD in order to develop and compare machine learning models for diagnosing and monitoring PD.

Secondary objectives are:

• Correlate ocular motor, pupil and gait parameters with several clinical parameters, including the MDS-UPDRS.
• Collect real-world evidence (RWE) data regarding health economics parameters to address the individual and combined properties, effects, and/or impacts of the deployed health technologies.
• By analysing the data collected, we also aim to contribute to the scientific understanding of PD, potentially uncovering new insights into disease patterns, progression, and response to treatments.

• Study Type: Observational
• Enrollment (Estimated): 100

Contacts and Locations

• Study Contact: Name: Ana Coito, Ph.D.; Phone Number: +41 (0)31 589 67 92; Email: ana.coito@machinemd.com
• Study Contact Backup: Name: Pia Massatsch, Ph.D.; Email: pia.massatsch@machinemd.com

Study Locations

• Switzerland
  • Canton of Zurich
    • Zurich, Canton of Zurich, Switzerland, 8091
      • Recruiting
      • University Hospital of Zürich
      • Contact: Konrad Weber, Prof. Dr. med.; Phone Number: +41 44 255 55 49; Email: Konrad.Weber@usz.ch
      • Contact: Erika Han, MD; Email: erika.han@usz.ch
      • Principal Investigator: Konrad Weber, Prof. Dr. med.
      • Sub-Investigator: Erika Han, MD

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

At least 100 patients with typical PD and at least 12 patients with atypical Parkinson's.

Description

Inclusion Criteria:

• Diagnosis of Parkinson's disease or of another parkinsonian syndrome (atypical Parkinson's)
• Refractive error between -6 and +4 diopters, on both eyes
• Informed consent by participant documented per signature
• Able to self-report history of daily gait freezing and/or festination
• Able to walk unsupported or using an aid for at least 5 minutes and if over 69 used to carrying out this level of exercise

Exclusion Criteria:

• Other known neurological diseases
• Current medication/drugs that could potentially influence performance in ocular motor tasks and/or compliance in the judgement of the investigator (e.g. benzodiazepines, alcohol, stimulants, or recreational drugs) - except Parkinson's medications
• Incapacity to understand and comply with the examination (e.g. due to advanced cognitive decline, failure to comply with easy experimental instructions and tasks)
• Any injury or disorder that may affect eye movement measurements or balance (other than Parkinson's or referring primary condition)
• Any skin conditions or broken skin in the calf and behind the knee area
• Lack of access or limited connectivity to WiFi in home setting

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
people with Parkinson's disease	Diagnostic test: oculomotor and pupillary function assessment; This is an exploratory open-label single-centre research project intended to collect data of PD patients in order to develop machine learning models for helping in the diagnosis and monitoring of PD. Each patient will have an initial visit and then a second visit after two weeks. The estimated duration of the study visit 1 is 3 hours, not taking into account the time for patient information and informed consent. The estimated duration of the study visit 2 is 2 hours. During these two visits, patients will undergo an MDS-UPDRS assessment, a neos examination, a standard manual ocular motor and pupil function examination, and gait assessment simultaneously with GaitQ senti and an IMU sensor placed on their back. In the two weeks separating the two visits, patient's gait will be monitored daily at home with GaitQ senti, where they will perform a daily TUG test comprising of 15 m walk, 5 sit to stand, and 5-minute walk. Healthy individuals data will be included from a previous study.
Atypical parkinsonism	Diagnostic test: oculomotor and pupillary function assessment; This is an exploratory open-label single-centre research project intended to collect data of PD patients in order to develop machine learning models for helping in the diagnosis and monitoring of PD. Each patient will have an initial visit and then a second visit after two weeks. The estimated duration of the study visit 1 is 3 hours, not taking into account the time for patient information and informed consent. The estimated duration of the study visit 2 is 2 hours. During these two visits, patients will undergo an MDS-UPDRS assessment, a neos examination, a standard manual ocular motor and pupil function examination, and gait assessment simultaneously with GaitQ senti and an IMU sensor placed on their back. In the two weeks separating the two visits, patient's gait will be monitored daily at home with GaitQ senti, where they will perform a daily TUG test comprising of 15 m walk, 5 sit to stand, and 5-minute walk. Healthy individuals data will be included from a previous study.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Development of machine learning models for diagnosing and monitoring of PD	The machine learning algorithms will be trained based on a clinical dataset of 50 PD patients, healthy individuals (data from another study), and 12 additional patients with other parkinsonian disorders. This dataset consists of ocular motor and pupil data provided by neos, ocular motor and pupil assessment provided by the standard clinical examination, gait data provided by GaitQ senti (placed on the patient's leg), gait data provided by an IMU sensor placed on the patient's back, demographic information (age, sex, ethnicity, eye colour), clinical information (disease stage, disease duration, age of onset of disease, medication, MDS-UPDRS score).	1 year

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Correlation with clinical parameters	The secondary outcome is correlation between ocular motor parameters with clinical parameters (disease stage, disease duration, age of onset of disease, medication, MDS-UPDRS score).	1 year

Collaborators and Investigators

• Sponsor: machineMD AG
• Collaborators: University of Zurich; University of Exeter; University Hospital, Zürich; gaitQ Limited
• Investigators: Principal Investigator: Konrad Weber, Prof. Dr. med., University of Zurich

Study record dates

Study Major Dates

• Study Start (Actual): October 1, 2024
• Primary Completion (Estimated): June 30, 2026
• Study Completion (Estimated): June 30, 2026

Study Registration Dates

• First Submitted: October 24, 2024
• First Submitted That Met QC Criteria: October 28, 2024
• First Posted (Actual): October 29, 2024

Study Record Updates

• Last Update Posted (Estimated): October 7, 2025
• Last Update Submitted That Met QC Criteria: October 2, 2025
• Last Verified: October 1, 2025

More Information

Terms related to this study

• Keywords: gait; Parkinson's disease; eye movements; pupil; virtual reality headset; eye tracking; oculomotor; ocular motor
• Additional Relevant MeSH Terms: Synucleinopathies; Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Neurodegenerative Diseases; Movement Disorders; Parkinsonian Disorders; Basal Ganglia Diseases; Parkinson Disease
• Other Study ID Numbers: HRO-002_BIP-PD; 109.813 INT-LS (Other Grant/Funding Number: Innosuisse)

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No