Vision-based Assessment of Joint Extensibility in Ehlers Danlos Syndrome
Study Details
Study Description
Brief Summary
Ehlers Danlos Syndrome (EDS) is a heterogenous group of genetic disorders with 13 identified subtypes. Hypermobile EDS (hEDS), although the most common subtype of EDS, does not yet have an identified genetic mutation for diagnostic confirmation. Generalized joint hypermobility (GJH) is one of the hallmark features of hEDS. The scoring system used in measurement of GJH was described by Beighton. The Beighton score is calculated using a dichotomous scoring system to assess the extensibility of nine joints. Each joint is scored as either hypermobile (score = 1) or not hypermobile (score = 0). The total score (Beighton score) can vary between a minimum of 0 and a maximum of 9, with higher scores indicating greater joint laxity.
While there is moderate validity and inter-rater variability in using the Beighton score, there continue to be several challenges with its widespread and consistent application by clinicians. Some of the barriers reported in the literature include:
- In open, non-standardized systems there can be significant variation in the method to perform these joint extensibility tests including assessing baseline measurements, ii) Determining consistent and standard measurement tools/methodology e.g. goniometer use can vary widely iii) Assessing the reliability of the cut off values and, iv) Performing full assessment prior to informing patients of possible classification of GJH positivity (low specificity and low positive predictive).
Inappropriate implementation of tests to assess GJH results in inaccurate identification of GJH and potentially unintended negative consequences of making the wrong diagnosis of EDS. The objective of this study is to create a more robust and valid method of joint mobility measurement and reduce error in the screening of EDS through use of a smartphone-based machine learning application systems for measurement of joint extensibility.
The project will:
- Create a smart-phone enabled visual imaging app to assess the measurement of joint extensibility, ii) Assess the feasibility of using the smart-phone app in a clinical setting to screen potential EDS patients, iii) Determine the validity of the application in comparison to in person clinical assessment in a tertiary care academic EDS program. If successful, the smart-phone application could help standardize the care of potential EDS patients in an efficient and cost-effective manner.
Condition or Disease | Intervention/Treatment | Phase |
---|---|---|
|
Detailed Description
1.0 Background and Motivation
1.1 Ehlers-Danlos Syndrome
Ehlers-Danlos Syndrome (EDS) is a heterogeneous group of genetic disorders characterized by abnormal formation and/or function of collagen, fibrillin, and elastin in the body, and typically manifests as weak joints that sub luxate and/or dislocate, stretchy and/or fragile skin, organ/systems dysfunction, and significant widespread pain. Historical data estimates that EDS affects approximately 1 in 5,000 to 1 in 20,000 individuals; although recent reports suggest it could be as common as 1 in 200.
As of 2017, there are 13 recognized types of EDS, including 12 genotypes that can be confirmed molecular analyses as well as Hypermobile EDS (hEDS), for which the associated genes have not yet been identified. Hypermobility of joints in a generalized fashion is a predominant physical feature in EDS. Although hEDS is the most common forms of EDS, the combination of vague, non-specific, and overlapping symptoms makes distinguishing between hEDS and other complex chronic illnesses challenging. Presently, hEDS is diagnosed based on 2017 clinical criteria by clinical interview, medical history, and a physical exam. This clinical criterion includes assessment of generalized joint hypermobility as a mandatory feature.
However, because most of the phenotypic characteristics of hEDS are subjectively qualified, and there are no objective validated tests that offer objective verification for either of these syndromes, it is not uncommon for clinicians to reach conflicting conclusions about a patients' diagnosis.
1.2 Beighton Exam
One of the key clinical tools for assessing joint hypermobility commonly associated with EDS is the Beighton exam. This exam consists of the Beighton maneuvers, which include the apposition of the thumb to forearm, extension of the fifth finger beyond 90 degrees, extension of knees and elbows beyond 10 degrees, and forward flexion. Each joint is scored as either hypermobile (score = 1) or not hypermobile (score = 0). The total score, referred to as the Beighton score, thus varies from a minimum of 0 to a maximum of 9, with higher scores indicating a larger number of hypermobile joints. The Beighton score is an important tool for EDS diagnosis; individuals with a Beighton score greater than 4 who also experience chronic pain are diagnosed with generalized hypermobility spectrum disorders (G-HSD).
While the Beighton score is a validated measure of joint hypermobility, there continue to be several challenges with its widespread and consistent application by clinicians. In a busy clinical setting, physicians often perform only a single maneuver of thumb to forearm apposition as a rapid method of hypermobility assessment introducing a significant diagnostic error and misdiagnosis. Further, the Beighton score is only clinically validated when assessed using a goniometer. However, the use of a goniometer is not standard practice for all clinicians, leading to higher inter-rater variability in Beighton score reporting.
As part of the standard of care at the GoodHope EDS clinic at Toronto General Hospital, clinicians conduct a formal Beighton assessment, measuring the range of motion of each joint with a goniometer. In addition to the Beighton maneuvers, the range of motion of the ankles and shoulders is similarly assessed using a goniometer. The assessment of the range of motion of the ankles and shoulders is not part of the formal Beighton exam, but is part of the standard of care at the GoodHope EDS clinic and is used to provide further diagnostic information. In this study, the investigators will refer to the formal Beighton exam plus the examination of the ankles and shoulders as the "extended Beighton exam".
1.3 Opportunities for Automated Beighton Score Assessment
The investigators propose the development of a data analysis system that can automatically and consistently determine joint range of motion and the Beighton score from video data. The goal of this system is to improve patient care by standardizing the application of the Beighton assessment across settings and clinicians. A future use of this objective Beighton score prediction application is to serve as a screening tool to help identify the patients that should be seen at a specialized EDS clinic most urgently.
To facilitate easy use of the tool in new settings, the investigators propose the use of video data. Cameras are readily available in many consumer devices such as phones, tablets, and webcams, thus allowing for easy adoption of the tool. The investigators envision that a future iteration of this system could be deployed as a phone or web-based application, allowing primary care physicians or patients to upload videos and receive a standardized Beighton score to aid in the referral process.
2.0 Objective
The objective of this study is to develop and validate a computer-vision based system for the assessment of the Beighton score of joint hypermobility. In addition to the joints assessed as part of the Beighton exam (ie. the thumb, fifth finger, elbows, knees, and spine), the investigators will also assess the range of motion of the ankles and shoulders using our computer-vision based approach. To facilitate this, the investigators aim to collect a dataset of videos of 225 patients undergoing a standard physical examination of joint hypermobility at the GoodHope EDS Clinic.
3.0 Research Design
3.1 Overview
This single-center observational study will collect cross-sectional data of participants during a standard physical intake examination at the specialized EDS clinic at the Toronto General Hospital, University Health Network. The purpose of this study is to collect video and limited clinical data to facilitate the development and evaluation of an objective Beighton score assessment system.
3.2 Setting
This study will be conducted at the GoodHope EDS Clinic at Toronto General Hospital. The clinic is a tertiary care facility that services the Ontario population. The clinic largely acts as a center for diagnosing EDS facilitating diagnostic testing, coordinating referrals, assisting with the development of patient care plans, and conducting research. The clinic's primary care team is responsible for the initial assessment which includes a clinical interview, a physical exam, and several patient-completed questionnaires.
The clinical interview inquires into the patient's presenting condition and includes an overview and assessment of joint hypermobility includes the Beighton score and 5-point questionnaire (5PQ). The questionnaires patients are asked to complete center around qualifying pain, fatigue, quality of life, and mental health.
The data for this study will be collected during a subset of the assessments performed as part of the standard of care at the GoodHope EDS Clinic. The specific data that will be collected is outlined in section 4.
4.0 Data Collection
This is a cross-sectional observational study in which the investigators will collect a minimal set of clinical information (age, sex, and range of motion of each joint as measured using a goniometer), as well as videos of participants undergoing an assessment of joint hypermobility during a physical examination performed as part of the usual standard of care at the EDS clinic.
4.1 Video Recordings
4.1.1 Rationale
To evaluate our proposed vision-based tool, a video dataset with a representative sample of the target population is needed. The investigators will record only the section of the physical exam when the participant is undergoing the assessment of joint hypermobility. These videos will be used to develop and then evaluate the proposed tool.
4.1.2 Data Collection Procedure
The investigators will use an UHN-owned consumer-grade mobile phone camera to record the videos. The phone will be encrypted, password protected, and its use will be restricted to study team members. The phone will not be connected to any wireless internet or cellular networks, and videos will be transferred to a secure location on the UHN network through a USB cable after each day of data collection.
The mobile phone will be mounted on a tripod and will be positioned far enough away from the participant, so it does not make contact with them or impede their range of motion. A member of the research team will enable video recording for the duration of the joint hypermobility exam. Video recording will be stopped and restarted for each joint. The position of the camera will be adjusted as needed between the assessment of each joint.
The participant will be asked to perform each maneuver for assessing joint hypermobility twice. In the first repetition, the participant will be the only person in the video frame and will be recorded using the camera. In the second repetition, the clinician will measure the range of motion at each joint using a goniometer. This repetition will not be recorded with the camera. Based on our previous experience, pose-estimation libraries sometimes have issues in predicting joint positions when there are multiple people close together in the video. Recording a video with just the participant will allow for the evaluation of the tool in its intended setting where the clinician is not performing the assessment.
4.2 Clinical Data Items
4.2.1 Rationale
In addition to the video recordings, a limited subset of clinical data will be collected for use in this study. The items that will be collected and their rationale are described below:
-
Range of motion: The range of motion at each joint is measured by the clinician using a goniometer as part of the standard Beighton assessment. The range of motion is also measured with a goniometer for the additional joints (ankles and shoulders) assessed during the physical examination. The range of motion will be measured with the JAMAR Plus+ clinical-grade digital goniometer. These measurements will be used to assess any discrepancies between the vision-based system and the clinician. These data will allow us to directly answer our first research question regarding the ability of our vision-based system to track the positions of joints in potentially hyperextended positions.
-
Beighton score (per joint and overall): The clinician's determination of joint laxity for each joint assessed as part of the Beighton exam, as well as the overall score is needed for training our system, as well as for evaluation of its performance. The Beighton score is collected for the thumbs, fifth fingers, elbows, knees, and spine. These data will be used to address our second and third research questions which examine the overall and per-joint performance of our proposed system.
-
Age and Sex: Joint laxity varies with age and sex so collecting this demographic information will help us assess how well our system performs across individuals from different demographics.
4.2.2 Data Collection Procedure
The clinical data will be collected during the in-person physical examination at the GoodHope EDS Clinic. The participant will be identified by their alpha-numeric study ID and the clinical data will be collected in a spreadsheet on a UHN computer.
Study Design
Arms and Interventions
Arm | Intervention/Treatment |
---|---|
New patients at the GoodHope EDS clinic at Toronto General Hospital All patients seen in the EDS clinic are eligible for inclusion, regardless of their presenting diagnosis or the results of their assessments. |
Other: No intervention, additional video data collection only
No intervention will be used. Consenting participants will have video recordings taken during their exam of joint hypermobility which will be analyzed at a later time
|
Outcome Measures
Primary Outcome Measures
- Comparison of agreement in predicted angle by pose-estimation library [4 months]
The performance of the developed machine learning models for predicting the range of motion will be analyzed by the pose-estimation library used. This analysis will be performed on the subset of the data collected during the first 2 months of data collection. This information will be used to select the pose-estimation libraries to proceed with when refining the machine learning models.
- Comparison of agreement in predicted angle by joint [1 year]
The performance of the developed machine learning models for predicting the range of motion at each joint (spine, knee, ankle, elbow, shoulder, thumb, fifth finger) will be analyzed independently for each joint. This will provide insight with respect to which joints the system is more accurate at predicting from video.
- Assess the accuracy of range of motion prediction using vision-based data [1 year]
Machine learning models trained on videos of individuals performing the joint hypermobility maneuvers will be developed. Their performance will be compared to the range of motion measured by an expert clinician using a goniometer.
Eligibility Criteria
Criteria
Inclusion Criteria:
- All patients seen in the GoodHope EDS clinic at Toronto General are eligible for inclusion, regardless of their presenting diagnosis or the results of their assessments
Exclusion Criteria:
- Patients who do not consent to participate will not be included (participants may withdraw consent at any time)
Contacts and Locations
Locations
Site | City | State | Country | Postal Code | |
---|---|---|---|---|---|
1 | GoodHope EDS - Toronto General Hospital | Toronto | Ontario | Canada | M5G 2C4 |
Sponsors and Collaborators
- University Health Network, Toronto
Investigators
- Principal Investigator: Nimish Mittal, MD, GoodHope Ehlers Danlos Syndrome Clinic, Toronto General Hospital
Study Documents (Full-Text)
None provided.More Information
Publications
- Andriluka M, Pishchulin L, Gehler P, Schiele B. 2D Human Pose Estimation: New Benchmark and State of the Art Analysis. In: 2014 IEEE Conference on Computer Vision and Pattern Recognition. 2014. p. 3686-93
- Cahill SV, Sharkey MS, Carter CW. Clinical assessment of generalized ligamentous laxity using a single test: is thumb-to-forearm apposition enough? J Pediatr Orthop B. 2021 May 1;30(3):296-300. doi: 10.1097/BPB.0000000000000732.
- Cao Z, Hidalgo G, Simon T, Wei SE, Sheikh Y. OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields. IEEE Trans Pattern Anal Mach Intell. 2021 Jan;43(1):172-186. doi: 10.1109/TPAMI.2019.2929257. Epub 2020 Dec 4.
- Critical Care Services Ontario, Ehlers-Danlos Syndrome Expert Panel Report, 2016. https://www.health.gov.on.ca/en/common/ministry/publications/reports/eds/Default.aspx.
- Fang H-S, Xie S, Tai Y-W, Lu C. RMPE: Regional Multi-person Pose Estimation. 2016 Nov 30; Available from: http://arxiv.org/abs/1612.00137
- He K, Gkioxari G, Dollár P, Girshick R. Mask R-CNN. In: Proceedings of the IEEE international conference on computer vision. 2017. p. 2961-9.
- Kim I-H, Jung I-H. A Study on Korea Sign Language Motion Recognition Using OpenPose Based on Deep Learning. 디지털콘텐츠학회논문지 (Journal of Digital Contents Society). 2021;22(4):681-7.
- Levy HP. Hypermobile Ehlers-Danlos Syndrome. 2004 Oct 22 [updated 2018 Jun 21]. In: Adam MP, Mirzaa GM, Pagon RA, Wallace SE, Bean LJH, Gripp KW, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2022. Available from http://www.ncbi.nlm.nih.gov/books/NBK1279/
- Lin T-Y, Maire M, Belongie S, Bourdev L, Girshick R, Hays J, et al. Microsoft COCO: Common Objects in Context. 2014 May 1; Available from: http://arxiv.org/abs/1405.0312
- Lu M, Zhao Q, Poston KL, Sullivan EV, Pfefferbaum A, Shahid M, Katz M, Kouhsari LM, Schulman K, Milstein A, Niebles JC, Henderson VW, Fei-Fei L, Pohl KM, Adeli E. Quantifying Parkinson's disease motor severity under uncertainty using MDS-UPDRS videos. Med Image Anal. 2021 Oct;73:102179. doi: 10.1016/j.media.2021.102179. Epub 2021 Jul 21.
- Lugaresi C, Tang J, Nash H, McClanahan C, Uboweja E, Hays M, et al. MediaPipe: A Framework for Building Perception Pipelines. 2019 Jun 14; Available from: https://arxiv.org/abs/1906.08172
- Mehdizadeh S, Nabavi H, Sabo A, Arora T, Iaboni A, Taati B. Concurrent validity of human pose tracking in video for measuring gait parameters in older adults: a preliminary analysis with multiple trackers, viewing angles, and walking directions. J Neuroeng Rehabil. 2021 Sep 15;18(1):139. doi: 10.1186/s12984-021-00933-0.
- Ota M, Tateuchi H, Hashiguchi T, Kato T, Ogino Y, Yamagata M, Ichihashi N. Verification of reliability and validity of motion analysis systems during bilateral squat using human pose tracking algorithm. Gait Posture. 2020 Jul;80:62-67. doi: 10.1016/j.gaitpost.2020.05.027. Epub 2020 May 25.
- Sabo A, Mehdizadeh S, Ng KD, Iaboni A, Taati B. Assessment of Parkinsonian gait in older adults with dementia via human pose tracking in video data. J Neuroeng Rehabil. 2020 Jul 14;17(1):97. doi: 10.1186/s12984-020-00728-9.
- Shi B, Brentari D, Shakhnarovich G, Livescu K. Fingerspelling Detection in American Sign Language. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021. p. 4166-75
- Slembrouck M, Luong H, Gerlo J, Schütte K, Van Cauwelaert D, De Clercq D, et al. Multiview 3d Markerless Human Pose Estimation from Openpose Skeletons. In: International Conference on Advanced Concepts for Intelligent Vision Systems. Springer; 2020. p. 166-78.
- Wang H, Xie Z, Lu L, Li L, Xu X. A computer-vision method to estimate joint angles and L5/S1 moments during lifting tasks through a single camera. J Biomech. 2021 Dec 2;129:110860. doi: 10.1016/j.jbiomech.2021.110860. Epub 2021 Nov 8.
- Williams S, Zhao Z, Hafeez A, Wong DC, Relton SD, Fang H, Alty JE. The discerning eye of computer vision: Can it measure Parkinson's finger tap bradykinesia? J Neurol Sci. 2020 Sep 15;416:117003. doi: 10.1016/j.jns.2020.117003. Epub 2020 Jun 30.
- Yahya M, Shah JA, Warsi A, Kadir K, Khan S, Izani M. Real time elbow angle estimation using single RGB camera. 2018 Aug 21; Available from: https://arxiv.org/abs/1808.07017
- Zhang F, Bazarevsky V, Vakunov A, Tkachenka A, Sung G, Chang C-L, et al. MediaPipe Hands: On-device Real-time Hand Tracking. 2020 Jun 17; Available from: http://arxiv.org/abs/2006.10214
- 22-5073