1
|
Haberfehlner H, Roth Z, Vanmechelen I, Buizer AI, Jeroen Vermeulen R, Koy A, Aerts JM, Hallez H, Monbaliu E. A Novel Video-Based Methodology for Automated Classification of Dystonia and Choreoathetosis in Dyskinetic Cerebral Palsy During a Lower Extremity Task. Neurorehabil Neural Repair 2024; 38:479-492. [PMID: 38842031 DOI: 10.1177/15459683241257522] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 06/07/2024]
Abstract
BACKGROUND Movement disorders in children and adolescents with dyskinetic cerebral palsy (CP) are commonly assessed from video recordings, however scoring is time-consuming and expert knowledge is required for an appropriate assessment. OBJECTIVE To explore a machine learning approach for automated classification of amplitude and duration of distal leg dystonia and choreoathetosis within short video sequences. METHODS Available videos of a heel-toe tapping task were preprocessed to optimize key point extraction using markerless motion analysis. Postprocessed key point data were passed to a time series classification ensemble algorithm to classify dystonia and choreoathetosis duration and amplitude classes (scores 0, 1, 2, 3, and 4), respectively. As ground truth clinical scoring of dystonia and choreoathetosis by the Dyskinesia Impairment Scale was used. Multiclass performance metrics as well as metrics for summarized scores: absence (score 0) and presence (score 1-4) were determined. RESULTS Thirty-three participants were included: 29 with dyskinetic CP and 4 typically developing, age 14 years:6 months ± 5 years:15 months. The multiclass accuracy results for dystonia were 77% for duration and 68% for amplitude; for choreoathetosis 30% for duration and 38% for amplitude. The metrics for score 0 versus score 1 to 4 revealed an accuracy of 81% for dystonia duration, 77% for dystonia amplitude, 53% for choreoathetosis duration and amplitude. CONCLUSIONS This methodology study yielded encouraging results in distinguishing between presence and absence of dystonia, but not for choreoathetosis. A larger dataset is required for models to accurately represent distinct classes/scores. This study presents a novel methodology of automated assessment of movement disorders solely from video data.
Collapse
Affiliation(s)
- Helga Haberfehlner
- Department of Rehabilitation Sciences, KU Leuven Bruges, Bruges, Belgium
- Department of Rehabilitation Sciences, KU Leuven, Leuven, Belgium
- Amsterdam UMC, Vrije Universiteit Amsterdam, Department of Rehabilitation Medicine, Amsterdam, The Netherlands
- Amsterdam Movement Sciences, Rehabilitation & Development, Amsterdam, The Netherlands
| | - Zachary Roth
- Department of Rehabilitation Sciences, KU Leuven Bruges, Bruges, Belgium
- Department of Rehabilitation Sciences, KU Leuven, Leuven, Belgium
| | - Inti Vanmechelen
- Department of Rehabilitation Sciences, KU Leuven Bruges, Bruges, Belgium
- Department of Rehabilitation Sciences, KU Leuven, Leuven, Belgium
| | - Annemieke I Buizer
- Amsterdam UMC, Vrije Universiteit Amsterdam, Department of Rehabilitation Medicine, Amsterdam, The Netherlands
- Amsterdam Movement Sciences, Rehabilitation & Development, Amsterdam, The Netherlands
- Amsterdam UMC, Emma Children's Hospital, Amsterdam, The Netherlands
| | | | - Anne Koy
- Department of Pediatrics, University of Cologne, Faculty of Medicine and University Hospital Cologne, Cologne, Germany
| | - Jean-Marie Aerts
- Department of Computer Science, Mechatronics Research Group (M-Group), KU Leuven Bruges, Distrinet, Bruges, Belgium
| | - Hans Hallez
- Department of Biosystems, Division of Animal and Human Health Engineering, Measure, Model and Manage Bioresponse (M3-BIORES), KU Leuven, Leuven, Belgium
| | - Elegast Monbaliu
- Department of Rehabilitation Sciences, KU Leuven Bruges, Bruges, Belgium
- Department of Rehabilitation Sciences, KU Leuven, Leuven, Belgium
| |
Collapse
|
2
|
Suresh N, Garg D, Pandey S, Malhotra RK, Majumdar R, Mukherjee SB, Sharma S. Spectrum of Movement Disorders and Correlation with Functional Status in Children with Cerebral Palsy. Indian J Pediatr 2022; 89:333-338. [PMID: 34097232 DOI: 10.1007/s12098-021-03785-7] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/04/2020] [Accepted: 04/22/2021] [Indexed: 11/29/2022]
Abstract
OBJECTIVES To detail the spectrum of movement disorders (MD) among children with cerebral palsy (CP) and assess impact on functional status. METHODS In this cross-sectional study, children with CP were recruited and examined for various MDs. Tone abnormality was assessed using Hypertonia Assessment Tool (HAT), functional status using Gross Motor Function Classification System Expanded and Revised (GMFCS E&R), Manual Ability Classification System (MACS), and Communication Function Classification System (CFCS). These scores were classified into mild-moderate (level I-III)/severe (level IV-V) categories. RESULTS A total of 113 children (mean age 4.9 ± 3.4 y, 66.4% boys) were enrolled. MDs were noted in 52 (46%) children; the most frequent were dystonia (28%), chorea (14%), choreoathetosis (8%). Of 64 children with quadriparetic CP, 27 (42.2%) demonstrated MDs. Of 19 children with hemiparetic CP, 2 (10.5%) had MDs. Of 16 children with dyskinetic CP, 15 (93%) had MDs. Children with dyskinetic CP had significantly higher frequency of MDs (p = 0.001). There was no difference in occurrence of all MDs or dystonia aloneamongst the two categories (mild-moderate/severe) of GMFCS E&R levels, CFCS levels or MACS levels. CONCLUSION Although diverse MDs occur frequently in CP, these do not correlate with the broad functional status of the child. The study is limited by small sample size.
Collapse
Affiliation(s)
- Narayanaswamy Suresh
- Department of Pediatrics (Neurology Division), Lady Hardinge Medical College and Associated Hospitals, New Delhi, India
| | - Divyani Garg
- Department of Neurology, Lady Hardinge Medical College and Associated Hospitals, New Delhi, India
| | - Sanjay Pandey
- Department of Neurology, Govind Ballabh Pant Institute of Post Graduate Medical Education and Research, New Delhi, India
| | - Rajeev Kumar Malhotra
- Delhi Cancer Registry, Department of Surgical Oncology, Dr. B.R.A. Institute Rotary Cancer Hospital, All India Institute of Medical Sciences, New Delhi, India
| | - Ritu Majumdar
- Department of Physical Medicine and Rehabilitation, Lady Hardinge Medical College and Associated Hospitals, New Delhi, India
| | - Sharmila B Mukherjee
- Department of Pediatrics (Neurology Division), Lady Hardinge Medical College and Associated Hospitals, New Delhi, India
| | - Suvasini Sharma
- Department of Pediatrics (Neurology Division), Lady Hardinge Medical College and Associated Hospitals, New Delhi, India.
| |
Collapse
|
3
|
Bekteshi S, Nica IG, Gakopoulos S, Konings M, Maes R, Cuyvers B, Aerts JM, Hallez H, Monbaliu E. Exercise load and physical activity intensity in relation to dystonia and choreoathetosis during powered wheelchair mobility in children and youth with dyskinetic cerebral palsy. Disabil Rehabil 2021; 44:4794-4805. [PMID: 33970729 DOI: 10.1080/09638288.2021.1921064] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/21/2022]
Abstract
PURPOSE To explore the relation between exercise load, physical activity intensity, and movement disorders during powered wheelchair (PW) mobility in people with severe dyskinetic cerebral palsy (DCP). METHODS Ten participants with DCP, 6-21 years old, users of a head/foot steering system were included. Dystonia and choreoathetosis were assessed using the Dyskinesia Impairment Mobility Scale (DIMS), heart rate (HR) was used to assess the exercise load of the tasks on the participants, and the accelerometry-based activity index (AI) to measure the physical activity intensity and energy expenditure during mobility task performance. RESULTS Neck- and distal arm dystonia showed significant correlations with HR (0.64 < rs < 0.77; 0.009 < p < 0.048), whereas neck- and proximal arm choreoathetosis with AI (0.64 < rs < 0.76, 0.011 < p < 0.044). Total-body AI was strongly correlated to the AI of the arms (0.66 < rs < 0.90, < 0.001 < p < 0.038), but not to the AI of the head. CONCLUSIONS During PW mobility tasks, dystonia is associated to exercise load and choreoathetosis to physical activity intensity and energy expenditure. Findings highlight the difficulties in measuring exercise load and activity intensity in PW users with DCP due to the involuntary hypertonic and/or hyperkinetic hallmark of the movement disorders. Nevertheless, a relaxed surrounding with minimal distractions during PW training may increase learning efficiency. Future studies with a bigger sample size are highly recommended to fully establish the relationship between the variables and to allow generalizability of results.Implications for rehabilitationDystonia is positively related to heart rate during powered mobility, which may be explained by the hypertonic hallmark of dystonia causing an increase in exercise load.Choreoathetosis is positively related to the physical activity index during powered mobility where the hyperkinetic hallmark of choreoathetosis may lead to an increase in physical activity intensity and energy expenditure.Arm overflow movements are the component which contribute the most to total-body activity index, thus, minimizing these movements may lower the overall energy expenditure during powered mobility.Mobility training in a relaxed surrounding with minimal distractions and minimized arm overflow movements may lead to a less-demanding powered wheelchair mobility experience and increased learning efficiency.
Collapse
Affiliation(s)
- Saranda Bekteshi
- KU Leuven, Bruges Campus, Department of Rehabilitation Sciences, Research Group for Neurorehabilitation, Bruges, Belgium
| | - Ioana Gabriela Nica
- KU Leuven, Department of Biosystems, Division of Animal and Human Health Engineering, Measure, Model and Manage Bioresponse (M3-BIORES), Leuven, Belgium
| | - Sotirios Gakopoulos
- KU Leuven, Bruges Campus, Department of Computer Science, Mechatronics Research Group, Bruges, Belgium
| | - Marco Konings
- KU Leuven, Bruges Campus, Department of Rehabilitation Sciences, Research Group for Neurorehabilitation, Bruges, Belgium
| | - Rozanne Maes
- KU Leuven, Department of Biosystems, Division of Animal and Human Health Engineering, Measure, Model and Manage Bioresponse (M3-BIORES), Leuven, Belgium
| | - Benoit Cuyvers
- KU Leuven, Department of Biosystems, Division of Animal and Human Health Engineering, Measure, Model and Manage Bioresponse (M3-BIORES), Leuven, Belgium
| | - Jean-Marie Aerts
- KU Leuven, Department of Biosystems, Division of Animal and Human Health Engineering, Measure, Model and Manage Bioresponse (M3-BIORES), Leuven, Belgium
| | - Hans Hallez
- KU Leuven, Bruges Campus, Department of Computer Science, Mechatronics Research Group, Bruges, Belgium
| | - Elegast Monbaliu
- KU Leuven, Bruges Campus, Department of Rehabilitation Sciences, Research Group for Neurorehabilitation, Bruges, Belgium
| |
Collapse
|
4
|
Development of the Dyskinesia Impairment Mobility Scale to Measure Presence and Severity of Dystonia and Choreoathetosis during Powered Mobility in Dyskinetic Cerebral Palsy. APPLIED SCIENCES-BASEL 2019. [DOI: 10.3390/app9173481] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
The majority of individuals with dyskinetic cerebral palsy cannot use powered mobility with a joystick, due to the lack of manual abilities by the severe presence of dystonia and choreoathetosis. Reliable measurements of these movement disorders is indispensable for good evaluation towards evidence–based insights during powered mobility. This study aimed to develop and assess the Dyskinesia Impairment Mobility Scale (DIMS), a video–based tool to measure presence and severity of dystonia and choreoathetosis during powered mobility. DIMS was measured for the neck and arms region during five mobility tasks. Interrater reliability, test–retest reliability, internal consistency and concurrent validity of the DIMS were assessed. Interrater reliability coefficients ranged between 0.68 and 0.87 for the total DIMS, and the dystonia and choreoathetosis subscales. Test–retest reliability was moderate to excellent (range 0.51–0.93) while Cronbach’s alpha was good (range 0.69–0.81) for the total scale and subscale scores. Concurrent validity showed during mobility tasks significant correlations with rest postures in the arm region, and with requested but voluntary activity in the neck region. The DIMS reliably measures the presence and severity of the movement disorders during powered mobility, increasing insights into the underlying mechanisms of independent mobility. This scale may therefore be a promising tool to evaluate mobility training.
Collapse
|