A video based method to quantify posture of the head and trunk in sitting

Sánchez, María Beatriz González; Loram, Ian D.; Darby, John; Holmes, Paul; Butler, Penelope B

doi:10.1016/j.gaitpost.2016.10.012

Cited by 24 publications

(28 citation statements)

References 14 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the present study, markers were required to identify a head segment and enable the calculation of head segmental angles. Although a fully automated objective tool should be markerless to avoid the limitations of missing markers when tracking on 2D videos [13], the results of the present study give confidence for future work towards a markerless system. The further development of a tool based on these principles, has the potential to complement clinical assessments currently used (in addition to SATCo) to support the therapeutic interventions in physiotherapy that will contribute to the evidence-based practice decision making.…”

Section: Discussionmentioning

confidence: 68%

Quantification of head motion in children with cerebral palsy when testing segmental trunk control

Loram

Holmes

et al. 2018

2018 IX International Seminar of Biomedical Engineering (SIB)

Self Cite

View full text Add to dashboard Cite

2018)Quantification of head motion in children with cerebral palsy when testing segmental trunk control.Abstract-The development of objective assessment tools to complement subjective evaluations could have a major impact in the evaluation of the consequences of neuromotor disabilities, specifically in the assessment of trunk control in children with cerebral palsy (CP). Current assessments are subjective and typically assess the general control status of a child under unbalancing situations. Only the Segmental Assessment of Trunk Control (SATCo) evaluates static, active and reactive control at six discrete trunk segmental levels and free sitting to identify the topmost trunk segment where the child does not demonstrate control. The characteristics of the SATCo make it an ideal start point to generate an objective measurement tool of control. Motion of the head is one of the main characteristics observed during a SATCo to determine the child's segmental level of control. This pilot study measured head motion during a SATCo to determine if the measurement reflected the clinical observations. Three children with different degrees of neuromotor disability were tested using the SATCo while a video camera recorded sagittal plane movements. A Head segment was defined by markers on the ear tragus and temporal fossa in vertical line with the ear when the head was aligned. Head segmental angles were calculated in relation to a real vertical, and the cumulative displacement was normalised by time (deg·s-1). One segment where external support was provided high on the trunk (Upper-Thoracic, UT) and one where external support was at the pelvis i.e. low on the trunk (Lower-Lumbar, LL) were selected for the analysis. Overall, head motion values were larger for the lower trunk tests (LL) than for the tests at the higher segment (UT) with greater external support provided. The poorer lumbar control of Child 2 and Child 3 was demonstrated in both active (39.9°·s-1, 23.3°·s-1) and reactive tests (37.3°·s-1, 13.8°·s-1). The results also showed how increasing task complexity (static to reactive) and reducing the level of support (UT to LL), increased head motion in the presence of a neuromotor disability. The results presented here corroborated the clinical evaluation of the SATCo. Head motion will have small values when children with CP are tested at a trunk segmental level where control has already been learnt or where external support is provided; children with poorer lumbar control will have an increased head motion when testing is below their trunk segmental level of assured control. This video-based approach represents a step towards the development of an objective tool for the assessment of head/trunk control in children with CP.

show abstract

Section: Discussionmentioning

confidence: 68%

Quantification of head motion in children with cerebral palsy when testing segmental trunk control

Loram

Holmes

et al. 2018

2018 IX International Seminar of Biomedical Engineering (SIB)

Self Cite

View full text Add to dashboard Cite

show abstract

“…To generate results comparable to the existing literature, segment angles were computed according to [7], [9]. Results for the relevant segments (head, neck, UT, MT, LT, UL, LL, pelvis) for our fully automated method recorded comparable performance, 6.4° (MAE), to the current benchmark semiautomated method, 3.8° (MAE) (table 1).…”

Section: B Segment Anglesmentioning

confidence: 76%

“…This same data set had been used in previous research to develop a semi-automated, objective method for quantifying postural control [7]. That previous method reported problems of tracking drift and feature dropout (loss of feature during tracking), due to 3D motion in the 2D image plane, as well as occlusion of point-features [7], [9]. The present study introduced deep learning methods to overcome these problems and provide a fully automatic approach, which predicts the point-features directly from single images (i.e.…”

Section: Discussionmentioning

confidence: 99%

“…Previous work [7], [9] has demonstrated that the two components of 'alignment' and 'contact' can be accurately classified from a clinical assessment and quantified using video recordings. However, the methods used previously were semiautomated requiring manual initialisation of postural point features and necessitating significant human intervention to compensate for tracking drift and occlusion [7], [9].…”

mentioning

confidence: 99%

See 1 more Smart Citation

Fully automated image-based estimation of postural point-features in children with cerebral palsy using deep learning

Cunningham¹,

Sánchez²,

Butler³

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Objective: To automate identification of postural point-features from colour videos of children with neuromotor disability, during clinical assessment. The automatic identification of 13 points of interest (2, 6, 2, 3 points on the head, trunk, pelvis, arm respectively) is required to estimate the location and orientation of head, trunk, and arm segments, from videos of the clinical test "Segmental Assessment of Trunk Control" (SATCo) which is a test of seated postural control. Methods: Three expert operators manually annotated 13 point-features in every fourth image of 177 short (5-10 second) videos (25 Hz) showing 12 children with cerebral palsy (ages: 4.52 ±2.4, male: 9), participating in SATCo testing. Linear interpolation for the remaining images resulted in 30,825 annotated images. Meanpooling and max-pooling convolutional neural networks were trained with cross-validation, giving held-out test results for all children. Results: The point-features were estimated with error 4.40 ±3.75 pixels (mean-pooling), and 4.49 ±4.45 pixels (maxpooling), at approximately 100 images per second. Trunk segment angles (head, neck, 6 thoraco-lumbar-pelvic segments) were estimated with error 6.4° ±2.8° allowing accurate classification (F1 > 80%) of deviation from a reference posture at thresholds up to 3°, 3°, 2° respectively. Contact between arm point features (elbow, wrist) and supporting surface was classified at F1 = 80.5%. Conclusion and Significance: This study demonstrates, for the first time, a technical solution to automate identification of i) a sitting segmental-posture including individual trunk segments, ii) changes away from that posture, and iii) support from the upper limb, required for the clinical SATCo.

show abstract

“…A method for quantifying postural alignment in sitting has been developed that uses a video-based system [8]. The aim of the study reported here was to explore the potential for an automated method to establish use of the upper limb component of the CKC.…”

Section: Introductionmentioning

confidence: 99%

The potential of an automated system to identify the upper limb component of a controlled sitting posture

et al. 2017

Self Cite

View full text Add to dashboard Cite

Full trunk control in sitting is demonstrated only when the head-trunk are aligned and upper limbs remain free of contact from mechanical support. These components represent a Controlled Kinetic Chain and can be evaluated in people with neuromotor disability using the Segmental Assessment of Trunk Control (SATCo) when a therapist provides manual trunk support at different segmental levels. However, the SATCo, as with other clinical assessments of control, is subjective. The SATCo was translated to objective rules relating the position of the hands and elbows to the head-trunk and then tested to determine the extent to which this automated objective method replicated the clinical judgement. Clinical evaluation used video to determine whether the upper limb was free of mechanical support while the objective evaluation used 3D motion capture of the trunk and upper limbs with a classification rule. The agreement between clinical and objective classification was calculated for three conditions of a distance-from-support-surface threshold parameter in five healthy adults and five children with cerebral palsy. The unfitted (zero-threshold values) method replicated the clinical judgement in part (68.26%±15.7, adults, 48.3%±33.9 children). The fitted (level-of-support determined) agreement showed that the process could be refined using trial specific parameters (88.32%±5.3 adults, 89.84%±10.2 children). The fixed-values agreement showed high values when using general group parameters (80.80%±3.1 adults, 74.31%±21.5 children). This objective classification of the upper limb component of trunk control largely captures the clinical evaluation. It provides the first stages in development of a clinically-friendly fully automated method.

show abstract

A video based method to quantify posture of the head and trunk in sitting

Cited by 24 publications

References 14 publications

Quantification of head motion in children with cerebral palsy when testing segmental trunk control

Quantification of head motion in children with cerebral palsy when testing segmental trunk control

Fully automated image-based estimation of postural point-features in children with cerebral palsy using deep learning

The potential of an automated system to identify the upper limb component of a controlled sitting posture

Contact Info

Product

Resources

About