Recruiting · registry status

Predicting freezing while walking

Researchers are collecting movement data to develop computer methods that might anticipate freezing episodes in Parkinson's.

Learning by observing or collecting information · Study reference: DRKS00035559

Plain-language introduction written with AI from the registry; not independently checked by a clinician. Read the original details below ↓

This registry record was last updated over six months ago, or its update date is missing. Recruitment may have changed. Confirm with the team before making plans.

Open the official registry record ↗

Follow this study

Looking for volunteers

Start (reported date): 2026-03-20

Planned dates can move. A study finishing does not tell us when a paper will be published.

No results summary has been confirmed in the registry records we imported. See connected papers below; we keep checking after recruitment ends.

Changes we have recorded

2026-10-11 — recruiting

These are dates we observed a change, not necessarily the dates it happened.

Papers connected to this study

No connected paper has been found yet. The tracker checks the growing library for study identifiers and registry-linked publications.

Who can join?

Age range not listed · Ask the team who can join

These are starting points, not the full rules. The research team can tell you whether the study is right for your situation.

Read all the rules for taking part

Sex eligibility reported by registry: all

Inclusion criteria: - Clinical diagnosis of PD (ICD- 10: G20) - Modified Hoehn & Yahr Stage I to IV - Self-reported FOG severity of at least one FOG episode per month, irrespective of FOG occurring ON- or OFF-medication. - Shoe sizes from 36 – 47 (Due to Smart Insoles) Exclusion criteria: - Requires adult representation - A fall frequency of more than once a week. - Acute pain related subjective influence on walking pattern - Inability to walk 10m and 1 minute without a walking stick/rollator/walker indoors. - >135kg body weight (due to smart insoles) - Wounds or fractures on the feet (due to smart insoles) - Inability of the client to wear closed shoes in combination with the smart insoles (due to smart insoles)

Full study name & original research details

Official study title

PreFOG: Prediction of freezing of gait in people with Parkinson's Disease using machine learning methods - PreFOG

Short title used by the registry

PreFOG: Prediction of freezing of gait in people with Parkinson's Disease using machine learning methods

Original description

Intervention: Group 1: Carrying out data collection with people with PD and FoG. A maximum of 25 people with PD will be recruited over a period of 1 year through Parkinson self support groups in Lower Austria. After written consent (informed consent) has been obtained, participants will be asked to complete several clinical tests, while the wear smart insoles and are videotaped. Personal and identifiable items, including jewelry and tattoos, will be removed or covered. During the tests, participants will be accompanied by an experienced physical therapist for safety reasons. The following tests are performed: –Timed up and Go (TuG) test. In this test, participants have to stand up from a chair, walk 3 meters, then turn 180 degrees, walk back and sit down on the chair again. The test is performed under three different conditions: normal, while carrying a glass of water in a plastic cup and during a cognitive task (naming animals that live on land or water) (Cedervall et al., 2020). In addition, turning is performed in both directions, which results in a total of 6 test runs per person. –10 meter walking test. The participants are asked to walk 10 meters, with the central 3 meters leading through a narrow corridor. This corridor will be equipped with mattresses on each side. Participants are asked to start from a standing position and walk in a straight line through the narrow corridor, stopping at a 10 meter mark. –1 minute walking test. Participants are asked to walk for 1 minute without stopping. If a person feels tired, a break can be taken. –1 minute walking test with simultaneous speaking. Participants are asked to walk for 1 minute and have a conversation with a person which walks by their side without interruption. Participants can tak Primary outcome: The aim of this study is to investigate whether it is possible to predict Freezing of Gait by employing machine learning techniques using multimodal data composed of foot pressure and inertial sensor data. Secondary outcome: How comprehensible is the German translation of the Characterizing Freezing of Gait Questionnaire (C-FoG-D)?

Conditions reported: G20;Parkinson disease

Registry records for this study

Records are joined using registration identifiers. Titles alone do not establish that two studies are the same.

German Clinical Trials Register · DRKS00035559 ↗ — Recruiting; WHO ICTRP processed 2026-06-01; checked 2026-10-14

Study type: Observational
Interventions: Group 1: Carrying out data collection with people with PD and FoG. A maximum of 25 people with PD will be recruited over a period of 1 year through Parkinson self support groups in Lower Austria. After written consent (informed consent) has been obtained, participants will be asked to complete several clinical tests, while the wear smart insoles and are videotaped. Personal and identifiable items, including jewelry and tattoos, will be removed or covered. During the tests, participants will be accompanied by an experienced physical therapist for safety reasons. The following tests are performed: –Timed up and Go (TuG) test. In this test, participants have to stand up from a chair, walk 3 meters, then turn 180 degrees, walk back and sit down on the chair again. The test is performed under three different conditions: normal, while carrying a glass of water in a plastic cup and during a cognitive task (naming animals that live on land or water) (Cedervall et al., 2020). In addition, turning is performed in both directions, which results in a total of 6 test runs per person. –10 meter walking test. The participants are asked to walk 10 meters, with the central 3 meters leading through a narrow corridor. This corridor will be equipped with mattresses on each side. Participants are asked to start from a standing position and walk in a straight line through the narrow corridor, stopping at a 10 meter mark. –1 minute walking test. Participants are asked to walk for 1 minute without stopping. If a person feels tired, a break can be taken. –1 minute walking test with simultaneous speaking. Participants are asked to walk for 1 minute and have a conversation with a person which walks by their side without interruption. Participants can tak
Phases: Not reported
Sponsor: IMC University of Applied Sciences Krems
Start date reported by registry: 2026-03-20 (reported)

Registry updated: Not reported · Status last verified by the registry submitter: Not reported

Registry records retrieved 2026-10-14 (UTC). Individual records may have older updates. Recruitment and eligibility must be confirmed with the study team.

Contact the research team

Public study contacts supplied to the registry. Ask whether recruitment is still open and what participation involves.

Agnes Wilhelm · +43 2732 802194 · agnes.wilhelm@imc.ac.at

Study locations

Site status can differ from overall study status. “Status not reported” means local availability needs confirmation. Remote participation and travel arrangements must be checked with the team.

Facility not reported

Austria

Status not reported

The original descriptions and participation rules come from the registry. Participation is voluntary and does not guarantee benefit.