Accurate Monitoring of 24-h Real-World Movement Behavior in People with Cerebral Palsy Is Possible Using Multiple Wearable Sensors and Deep Learning
Monitoring and quantifying movement behavior is crucial for improving the health of individuals with cerebral palsy (CP). We have modeled and trained an image-based Convolutional Neural Network (CNN) to recognize specific movement classifiers relevant to individuals with CP. This study evaluates CNN...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/22/9045 |
| _version_ | 1797457823459704832 |
|---|---|
| author | Ivana Bardino Novosel Anina Ritterband-Rosenbaum Georgios Zampoukis Jens Bo Nielsen Jakob Lorentzen |
| author_facet | Ivana Bardino Novosel Anina Ritterband-Rosenbaum Georgios Zampoukis Jens Bo Nielsen Jakob Lorentzen |
| author_sort | Ivana Bardino Novosel |
| collection | DOAJ |
| description | Monitoring and quantifying movement behavior is crucial for improving the health of individuals with cerebral palsy (CP). We have modeled and trained an image-based Convolutional Neural Network (CNN) to recognize specific movement classifiers relevant to individuals with CP. This study evaluates CNN’s performance and determines the feasibility of 24-h recordings. Seven sensors provided accelerometer and gyroscope data from 14 typically developed adults during videotaped physical activity. The performance of the CNN was assessed against test data and human video annotation. For feasibility testing, one typically developed adult and one adult with CP wore sensors for 24 h. The CNN demonstrated exceptional performance against test data, with a mean accuracy of 99.7%. Its general true positives (TP) and true negatives (TN) were 1.00. Against human annotators, performance was high, with mean accuracy at 83.4%, TP 0.84, and TN 0.83. Twenty-four-hour recordings were successful without data loss or adverse events. Participants wore sensors for the full wear time, and the data output were credible. We conclude that monitoring real-world movement behavior in individuals with CP is possible with multiple wearable sensors and CNN. This is of great value for identifying functional decline and informing new interventions, leading to improved outcomes. |
| first_indexed | 2024-03-09T16:28:26Z |
| format | Article |
| id | doaj.art-6323e5d65bcf4a90b1f4689c519c5f1c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T16:28:26Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-6323e5d65bcf4a90b1f4689c519c5f1c2023-11-24T15:05:08ZengMDPI AGSensors1424-82202023-11-012322904510.3390/s23229045Accurate Monitoring of 24-h Real-World Movement Behavior in People with Cerebral Palsy Is Possible Using Multiple Wearable Sensors and Deep LearningIvana Bardino Novosel0Anina Ritterband-Rosenbaum1Georgios Zampoukis2Jens Bo Nielsen3Jakob Lorentzen4Department of Pediatric Neurology 5003, University Hospital Copenhagen, Rigshospitalet, 2100 Copenhagen, DenmarkThe Elsass Foundation, 2920 Charlottenlund, DenmarkDepartment of Neuroscience, Faculty of Health and Medical Sciences, The Panum Institute, Copenhagen University, 2200 Copenhagen, DenmarkDepartment of Neuroscience, Faculty of Health and Medical Sciences, The Panum Institute, Copenhagen University, 2200 Copenhagen, DenmarkDepartment of Pediatric Neurology 5003, University Hospital Copenhagen, Rigshospitalet, 2100 Copenhagen, DenmarkMonitoring and quantifying movement behavior is crucial for improving the health of individuals with cerebral palsy (CP). We have modeled and trained an image-based Convolutional Neural Network (CNN) to recognize specific movement classifiers relevant to individuals with CP. This study evaluates CNN’s performance and determines the feasibility of 24-h recordings. Seven sensors provided accelerometer and gyroscope data from 14 typically developed adults during videotaped physical activity. The performance of the CNN was assessed against test data and human video annotation. For feasibility testing, one typically developed adult and one adult with CP wore sensors for 24 h. The CNN demonstrated exceptional performance against test data, with a mean accuracy of 99.7%. Its general true positives (TP) and true negatives (TN) were 1.00. Against human annotators, performance was high, with mean accuracy at 83.4%, TP 0.84, and TN 0.83. Twenty-four-hour recordings were successful without data loss or adverse events. Participants wore sensors for the full wear time, and the data output were credible. We conclude that monitoring real-world movement behavior in individuals with CP is possible with multiple wearable sensors and CNN. This is of great value for identifying functional decline and informing new interventions, leading to improved outcomes.https://www.mdpi.com/1424-8220/23/22/9045cerebral palsymovement behaviorwearable sensorsdeep learningmonitoring |
| spellingShingle | Ivana Bardino Novosel Anina Ritterband-Rosenbaum Georgios Zampoukis Jens Bo Nielsen Jakob Lorentzen Accurate Monitoring of 24-h Real-World Movement Behavior in People with Cerebral Palsy Is Possible Using Multiple Wearable Sensors and Deep Learning Sensors cerebral palsy movement behavior wearable sensors deep learning monitoring |
| title | Accurate Monitoring of 24-h Real-World Movement Behavior in People with Cerebral Palsy Is Possible Using Multiple Wearable Sensors and Deep Learning |
| title_full | Accurate Monitoring of 24-h Real-World Movement Behavior in People with Cerebral Palsy Is Possible Using Multiple Wearable Sensors and Deep Learning |
| title_fullStr | Accurate Monitoring of 24-h Real-World Movement Behavior in People with Cerebral Palsy Is Possible Using Multiple Wearable Sensors and Deep Learning |
| title_full_unstemmed | Accurate Monitoring of 24-h Real-World Movement Behavior in People with Cerebral Palsy Is Possible Using Multiple Wearable Sensors and Deep Learning |
| title_short | Accurate Monitoring of 24-h Real-World Movement Behavior in People with Cerebral Palsy Is Possible Using Multiple Wearable Sensors and Deep Learning |
| title_sort | accurate monitoring of 24 h real world movement behavior in people with cerebral palsy is possible using multiple wearable sensors and deep learning |
| topic | cerebral palsy movement behavior wearable sensors deep learning monitoring |
| url | https://www.mdpi.com/1424-8220/23/22/9045 |
| work_keys_str_mv | AT ivanabardinonovosel accuratemonitoringof24hrealworldmovementbehaviorinpeoplewithcerebralpalsyispossibleusingmultiplewearablesensorsanddeeplearning AT aninaritterbandrosenbaum accuratemonitoringof24hrealworldmovementbehaviorinpeoplewithcerebralpalsyispossibleusingmultiplewearablesensorsanddeeplearning AT georgioszampoukis accuratemonitoringof24hrealworldmovementbehaviorinpeoplewithcerebralpalsyispossibleusingmultiplewearablesensorsanddeeplearning AT jensbonielsen accuratemonitoringof24hrealworldmovementbehaviorinpeoplewithcerebralpalsyispossibleusingmultiplewearablesensorsanddeeplearning AT jakoblorentzen accuratemonitoringof24hrealworldmovementbehaviorinpeoplewithcerebralpalsyispossibleusingmultiplewearablesensorsanddeeplearning |