Predicting Fall Risk in Stroke Patients Using a Machine Learning Model and Multi-Sensor Data

Development and Validation of a Machine Learning-based Model to Predict a High-risk Group for Falls Using Multi-sensor Signals in Stroke Patients

The study assesses a machine learning model developed to predict fall risk among stroke patients using multi-sensor signals. This prospective, multicenter, open-label, sponsor-initiated confirmatory trial aims to validate the safety and efficacy of the model which utilizes electromyography (EMG) signals to categorize patients into high-risk or low-risk fall categories. The innovative approach hopes to offer a predictive tool that enhances preventative strategies in clinical settings, potentially reducing fall-related injuries in stroke survivors.

Study Overview

• Status: Recruiting
• Conditions: Stroke; Fall
• Intervention / Treatment: Device: EMG Analysis Software

Detailed Description

Objective: The primary objective is to develop and validate a machine learning-based model that uses multi-sensor (EMG) signals to identify stroke patients at high risk of falls. This model aims to improve on traditional fall risk assessments which rely heavily on physical assessments and patient history.

Study Design: This is a prospective, multicenter, open-label, confirmatory clinical trial. It involves collecting EMG data from stroke patients and applying machine learning techniques to predict fall risk. The study will compare the predictive accuracy of the machine learning model against conventional fall risk assessment tools.

Methods:

1. Participants:

   • Sample Size: 80 stroke patients and 10 healthy adults to establish baseline EMG readings.

2. Interventions:

   • Participants will undergo EMG signal collection from key lower limb muscles while performing standardized movements.

3. Outcome Measures:

   • Primary Outcome: Sensitivity and specificity of the machine learning model in predicting high-risk fall patients.
   • Secondary Outcomes: Comparison of the machine learning model's predictive performance with traditional fall risk assessment tools (e.g., Berg Balance Scale).

Data Collection:

• EMG sensors will be attached to the patients' muscles of the lower limbs. Sensors will record muscle activity during movement, which will then be analyzed using the machine learning model.
• The predictive model will be trained using features extracted from the EMG signals, and its performance will be validated against actual fall incidents reported during the follow-up period.

Statistical Analysis:

• The machine learning model's efficacy will be measured through its sensitivity (ability to correctly identify high-risk patients) and specificity (ability to correctly identify low-risk patients).
• Receiver Operating Characteristic (ROC) curves and Area Under the Curve (AUC) statistics will be used to assess model performance.

• Study Type: Observational
• Enrollment (Estimated): 90

Contacts and Locations

• Study Contact: Name: JungHyun Kim, prof; Phone Number: 82+1088632341; Email: kiking0@naver.com

Study Locations

• Korea, Republic of
  • Jongno
    • Seoul, Jongno, Korea, Republic of, 03080
      • Recruiting
      • Seoul National University Hospital
      • Contact: junghyun kim, Ph. D.; Phone Number: 82+1021740890

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Probability Sample

Study Population

The study aims to enroll approximately 80 stroke patients and 10 healthy adults to facilitate a comprehensive analysis of the EMG-based machine learning model's effectiveness.

Description

Stroke Participants

Inclusion Criteria:

• 19 years and older
• the onset of the stroke is less than 3months ago
• Lower extremity weakness due to stroke (MMT =< 4 grade)
• Cognitive ability to follow commands

Exclusion Criteria:

• stroke recurrence
• other neurological abnormalities (e.g. parkinson's disease).
• severely impaired cognition
• serious and complex medical conditions(e.g. active cancer)
• cardiac pacemaker or other implanted electronic system

Health Participants

Inclusion Criteria:

• 19 years and older
• Individuals who fully understand the necessity of the study and have voluntarily consented to participate as subjects

Exclusion Criteria:

• other neurological abnormalities (e.g. parkinson's disease).
• severely impaired cognition
• serious and complex medical conditions(e.g. active cancer)
• cardiac pacemaker or other implanted electronic system

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Sensitivity of the Machine Learning Model	The primary outcome measure is the sensitivity of the machine learning model, which refers to its ability to correctly identify patients who are at high risk of falls. Sensitivity is defined as the proportion of actual positives that are correctly identified.	At the time of the single visit

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Specificity of the Machine Learning Model	Specificity measures the proportion of actual negatives that are correctly identified.	At the time of the single visit

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Area Under the Receiver Operating Characteristic Curve	This is a performance measurement for classification problems at various threshold settings. ROC is a probability curve, and AUC represents the degree or measure of separability. It tells how much the model is capable of distinguishing between classes.	At the time of the single visit
Matthews Correlation Coefficient	The MCC is used in machine learning as a measure of the quality of binary classifications. It takes into account true and false positives and negatives and is generally regarded as a balanced measure which can be used even if the classes are of very different sizes.	At the time of the single visit

Collaborators and Investigators

• Sponsor: Seoul National University Hospital
• Collaborators: Ministry of Trade, Industry & Energy, Republic of Korea
• Investigators: Principal Investigator: Woo Hyung Lee, prof, Seoul National University Hospital; Study Director: Byung-Mo Oh, prof, Seoul National University Hospital; Study Director: Han Gil Seo, prof, Seoul National University Hospital; Study Director: Sung Eun Hyun, prof, Seoul National University Hospital; Study Director: Hyunmi Oh, prof, National Traffic Injury Rehabilitation Hospital; Study Director: Sumin Oh, B.S., National Traffic Injury Rehabilitation Hospital; Study Director: SO YEON JEON, B.S., Seoul National University Hospital

Study record dates

Study Major Dates

• Study Start (Actual): May 20, 2024
• Primary Completion (Actual): March 12, 2025
• Study Completion (Estimated): April 28, 2026

Study Registration Dates

• First Submitted: April 15, 2024
• First Submitted That Met QC Criteria: April 17, 2024
• First Posted (Actual): April 23, 2024

Study Record Updates

• Last Update Posted (Actual): June 2, 2025
• Last Update Submitted That Met QC Criteria: May 30, 2025
• Last Verified: May 1, 2025

More Information

Terms related to this study

• Keywords: Electromyography; Predict model; Machin leanning
• Additional Relevant MeSH Terms: Cerebrovascular Disorders; Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Vascular Diseases; Cardiovascular Diseases; Stroke
• Other Study ID Numbers: 0720242110; 20240012366 (Other Identifier: Ministry of Food and Drug Safety)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

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