Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking
Background: Falls are common and dangerous for stroke survivors. Current fall risk assessment methods rely on subjective scales. Objective sensor-based methods could improve prediction accuracy. Objective: Develop machine learning models using inertial sensors to objectively classify fall risk in st...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/3/812 |
| _version_ | 1797318234908655616 |
|---|---|
| author | Masoud Abdollahi Ehsan Rashedi Sonia Jahangiri Pranav Madhav Kuber Nasibeh Azadeh-Fard Mary Dombovy |
| author_facet | Masoud Abdollahi Ehsan Rashedi Sonia Jahangiri Pranav Madhav Kuber Nasibeh Azadeh-Fard Mary Dombovy |
| author_sort | Masoud Abdollahi |
| collection | DOAJ |
| description | Background: Falls are common and dangerous for stroke survivors. Current fall risk assessment methods rely on subjective scales. Objective sensor-based methods could improve prediction accuracy. Objective: Develop machine learning models using inertial sensors to objectively classify fall risk in stroke survivors. Determine optimal sensor configurations and clinical test protocols. Methods: 21 stroke survivors performed balance, Timed Up and Go, 10 Meter Walk, and Sit-to-Stand tests with and without dual-tasking. A total of 8 motion sensors captured lower limb and trunk kinematics, and 92 spatiotemporal gait and clinical features were extracted. Supervised models—Support Vector Machine, Logistic Regression, and Random Forest—were implemented to classify high vs. low fall risk. Sensor setups and test combinations were evaluated. Results: The Random Forest model achieved 91% accuracy using dual-task balance sway and Timed Up and Go walk time features. Single thorax sensor models performed similarly to multi-sensor models. Balance and Timed Up and Go best-predicted fall risk. Conclusion: Machine learning models using minimal inertial sensors during clinical assessments can accurately quantify fall risk in stroke survivors. Single thorax sensor setups are effective. Findings demonstrate a feasible objective fall screening approach to assist rehabilitation. |
| first_indexed | 2024-03-08T03:49:28Z |
| format | Article |
| id | doaj.art-18bfe57ad1714f5396b8f51460f133db |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-08T03:49:28Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-18bfe57ad1714f5396b8f51460f133db2024-02-09T15:21:54ZengMDPI AGSensors1424-82202024-01-0124381210.3390/s24030812Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-TaskingMasoud Abdollahi0Ehsan Rashedi1Sonia Jahangiri2Pranav Madhav Kuber3Nasibeh Azadeh-Fard4Mary Dombovy5Department of Industrial and Systems Engineering, Rochester Institute of Technology, Rochester, NY 14623, USADepartment of Industrial and Systems Engineering, Rochester Institute of Technology, Rochester, NY 14623, USADepartment of Industrial and Systems Engineering, Rochester Institute of Technology, Rochester, NY 14623, USADepartment of Industrial and Systems Engineering, Rochester Institute of Technology, Rochester, NY 14623, USADepartment of Industrial and Systems Engineering, Rochester Institute of Technology, Rochester, NY 14623, USADepartment of Rehabilitation and Neurology, Unity Hospital, Rochester, NY 14626, USABackground: Falls are common and dangerous for stroke survivors. Current fall risk assessment methods rely on subjective scales. Objective sensor-based methods could improve prediction accuracy. Objective: Develop machine learning models using inertial sensors to objectively classify fall risk in stroke survivors. Determine optimal sensor configurations and clinical test protocols. Methods: 21 stroke survivors performed balance, Timed Up and Go, 10 Meter Walk, and Sit-to-Stand tests with and without dual-tasking. A total of 8 motion sensors captured lower limb and trunk kinematics, and 92 spatiotemporal gait and clinical features were extracted. Supervised models—Support Vector Machine, Logistic Regression, and Random Forest—were implemented to classify high vs. low fall risk. Sensor setups and test combinations were evaluated. Results: The Random Forest model achieved 91% accuracy using dual-task balance sway and Timed Up and Go walk time features. Single thorax sensor models performed similarly to multi-sensor models. Balance and Timed Up and Go best-predicted fall risk. Conclusion: Machine learning models using minimal inertial sensors during clinical assessments can accurately quantify fall risk in stroke survivors. Single thorax sensor setups are effective. Findings demonstrate a feasible objective fall screening approach to assist rehabilitation.https://www.mdpi.com/1424-8220/24/3/812neurological disorderwearable sensorsmotion analysisfallneuroscienceTUG |
| spellingShingle | Masoud Abdollahi Ehsan Rashedi Sonia Jahangiri Pranav Madhav Kuber Nasibeh Azadeh-Fard Mary Dombovy Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking Sensors neurological disorder wearable sensors motion analysis fall neuroscience TUG |
| title | Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking |
| title_full | Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking |
| title_fullStr | Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking |
| title_full_unstemmed | Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking |
| title_short | Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking |
| title_sort | fall risk assessment in stroke survivors a machine learning model using detailed motion data from common clinical tests and motor cognitive dual tasking |
| topic | neurological disorder wearable sensors motion analysis fall neuroscience TUG |
| url | https://www.mdpi.com/1424-8220/24/3/812 |
| work_keys_str_mv | AT masoudabdollahi fallriskassessmentinstrokesurvivorsamachinelearningmodelusingdetailedmotiondatafromcommonclinicaltestsandmotorcognitivedualtasking AT ehsanrashedi fallriskassessmentinstrokesurvivorsamachinelearningmodelusingdetailedmotiondatafromcommonclinicaltestsandmotorcognitivedualtasking AT soniajahangiri fallriskassessmentinstrokesurvivorsamachinelearningmodelusingdetailedmotiondatafromcommonclinicaltestsandmotorcognitivedualtasking AT pranavmadhavkuber fallriskassessmentinstrokesurvivorsamachinelearningmodelusingdetailedmotiondatafromcommonclinicaltestsandmotorcognitivedualtasking AT nasibehazadehfard fallriskassessmentinstrokesurvivorsamachinelearningmodelusingdetailedmotiondatafromcommonclinicaltestsandmotorcognitivedualtasking AT marydombovy fallriskassessmentinstrokesurvivorsamachinelearningmodelusingdetailedmotiondatafromcommonclinicaltestsandmotorcognitivedualtasking |