Background: The shoulder joint complex is recognised for the wide range of movements it can perform and is extremely important for the performance of activities of daily living. Correct quantification of the range of motion is therefore essential in the assessment of patients with movement disorders. In this context, the universal goniometer stands out as the most widely used and economical tool, together with other technologies that have been validated in comparison with this device. Objective: This paper presents the validation of MediaPipe as a low-cost alternative tool to the universal goniometer for range of motion measurement, using a webcam as a sensor. A case study is also presented in which this tool has been implemented for the measurement and evolution of a patient’s range of motion in three rehabilitation sessions. Methods: To validate the agreement and repeatability of the MediaPipe tool, the Bland–Altman methodology was used by performing an exhaustive analysis of 360 shoulder abduction measurements for each arm in 40 volunteers, comparing them with the universal goniometer. Results: The results of the agreement analysis revealed that the mean differences between the MediaPipe and the universal goniometer for shoulder abduction were close to 0°, with limits of agreement ranging from −7.19° to 7.19° for the left arm and −6.39° to 6.39° for the right arm, indicating that the MediaPipe is valid in place of the universal goniometer. In addition, repeatability analysis showed mean differences of 0.51° for the left arm and 0.28° for the right arm, with repeatability limits ranging from −5.2° to 6.23°, suggesting reliable measurements under constant conditions. In relation to the case study, the results obtained with MediaPipe are consistent with the clinical evolution expected by the physiotherapist in shoulder injuries, considering the postoperative time and rehabilitation sessions. Conclusions: According to the results obtained, MediaPipe is a valid, reliable and low-cost tool to be used instead of the universal goniometer and can be implemented in the clinic for range of motion measurement in rehabilitation patients.
GillTKShanahanEMTuckerGR, et al.Shoulder range of movement in the general population: age and gender stratified normative data using a community-based cohort. BMC Muscoskelet Disord2020; 21(1): 676.
2.
ImagamaSAndoKKobayashiK, et al.Shoulder pain has most impact on poor quality of life among various types of musculoskeletal pain in middle-aged and elderly people: Yakumo study. Mod Rheumatol2019; 30: 1–5.
3.
YanikELAlvarezCClevelandRJ, et al.Occupational tasks associated with shoulder pain and upper extremity disability: a cross-sectional study in the Johnston County Osteoarthritis Project. BMC Muscoskelet Disord2024; 25(1): 374.
4.
García ParraPAnaya RojasMJiménez BravoB, et al.Correlación entre la exploración física y los hallazgos intraoperatorios de patología de hombro tratada mediante artroscopia. Análisis estadístico de 150 pacientes. Rev Esp Cir Ortop Traumatol2016; 60(5): 306–314.
5.
Martinez-CalderonJStruyfFMeeusM, et al.The association between pain beliefs and pain intensity and/or disability in people with shoulder pain: a systematic review. Musculoskelet Sci Pract2018; 37: 29–57.
6.
PribicevicM. The epidemiology of shoulder pain: a narrative review of the literature2012:147–185. DOI:10.5772/52931.
7.
VirtaLJorangerPBroxJI, et al.Costs of shoulder pain and resource use in primary health care: a cost-of-illness study in Sweden. BMC Muscoskelet Disord2012; 13: 17.
8.
WintersJCSobelJSGroenierKH, et al.The long-term course of shoulder complaints: a prospective study in general practice. Rheumatology1999; 38: 160–163.
9.
TaboadelaCHG. Una herramienta para la evaluación de las incapacidades laborales. Asociart SA Art, 2007.
10.
Pinto-CarralAFernandez-VillaTde la TorreAJM. Patient-reported mobility: a systematic review. Arch Phys Med Rehabil2016; 97(7): 1182–1194.
11.
ArmstrongADMacDermidJCChinchalkarS, et al.Reliability of range-of-motion measurement in the elbow and forearm. J Shoulder Elb Surg1998; 7(6): 573–580.
12.
BrutonAEllisBGoddardJ. Comparison of visual estimation and goniometry for assessment of metacarpophalangeal joint angle. Physiotherapy1999; 85(4): 201–208.
13.
ClarCCumminsEMcIntyreL, et al.Clinical and cost-effectiveness of autologous chondrocyte implantation for cartilage defects in knee joints: systematic review and economic evaluation. Health Technol Assess2005; 9: 1–82.
14.
MilaneseSGordonSBuettnerP, et al.Reliability and concurrent validity of knee angle measurement: smart phone app versus universal goniometer used by experienced and novice clinicians. Man Ther2014; 19(6): 569–574.
15.
MuralidaranSWilson-SmithARBasP, et al.Validation of a novel digital goniometer as a range of motion assessment tool for the lower extremity. J Orthop Res Ther2020; 5: 1158.
16.
LopezNPerezETelloE, et al.Statistical validation for clinical measures: repeatability and agreement of Kinect™-based software. BioMed Res Int2018; 2018: 6710595.
17.
IturrietaJPérezETettamantiF, et al.Validation and development of a posture quantification and classification technique using MediaPipe pose®. In: LopezNMTelloE (eds). Advances in Bioengineering and Clinical Engineering. SABI 2022. IFMBE Proceedings. Springer, 2024, vol 105.
18.
LatrecheAKelaiaiaRChemoriA, et al.Reliability and validity analysis of MediaPipe-based measurement system for some human rehabilitation motions. Measurement2023; 214: 112826.
19.
LugaresiCTangJNashH, et al.Mediapipe: a framework for building perception pipelines. arXiv preprint arXiv:1906.081722019. DOI:10.48550/arXiv.1906.08172.
20.
BazarevskyVGrishchenkoIRaveendranK, et al.Blazepose: On-device real-time body pose tracking. arXiv preprint arXiv:2006.102042020. DOI:10.48550/arXiv.2006.10204.
KiatkulanusornSLuangponNSrijuntoW, et al.Analysis of the concurrent validity and reliability of five common clinical goniometric devices. Sci Rep2023; 13(1): 20931.
