A Machine Learning Model to Automatic Assessment of Gross Motor Development in Children using Posenet
DOI:
https://doi.org/10.5433/1679-0375.2023.v44.48131Keywords:
automatic assessment, machine learning, motor development, TGMD-3Abstract
Gross motor skills such as sitting, jumping, and running are activities that involve the large muscles of the human body. The Test of Gross Motor Development, or TGMD, is widely used by researchers, pediatricians, physiotherapists, and educators from different countries to assess these skills in children aged 3 to 11 years. An important part of the test is that the movement, performed by the children, needs to be recorded and assessed by two or more professionals. The assessment process is laborious and takes time, and its automation is one of the main points to be developed. In recent years, methods have been proposed to automate the assessment according to the TGMD. The hypothesis investigated in this work is that it is possible to induce a machine learning model to identify whether the movement executed by the child is correct, considering only the first criterion of the TGMD-3 jumping skill. The skeleton of the children was extracted using PoseNet. A dataset of 350 images of Brazilian children between 3 and 11 years old performing the preparatory movement for the jump was used. The experimental results show an accuracy of 84%.
Metrics
References
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M., . . . Zheng, X. (2015). TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. Arxiv.
Bisi, M., Pacini Panebianco, G., Polman, R., & Stagni, R. (2017). Objective assessment of movement competence in children using wearable sensors: An instrumented version of the TGMD-2 locomotor subtest. Gait & Posture, 56, 42–48. DOI: https://doi.org/10.1016/j.gaitpost.2017.04.025
Clark, J. E. (2005). From the Beginning: A Developmental Perspective on Movement and Mobility. Quest, 57, 37–45. DOI: https://doi.org/10.1080/00336297.2005.10491841
Estimativa de pose humana em tempo real no navegador com TensorFlow.js. (2018). TensorFlow.
Gallahue, D., Ozmun, J., & Goodway, J. (2013). Compreendendo o desenvolvimento motor-: bebês, crianças, adolescentes e adultos. AMGH Editora.
Gonzalez, S., Alvarez, V., & Nelson, E. (2019). Do Gross and Fine Motor Skills Differentially Contribute to Language Outcomes? A Systematic Review. Frontiers In Psychology, 10, 1–16. DOI: https://doi.org/10.3389/fpsyg.2019.02670
Henderson, S., Sugden, D., & Barnett, A. (2007). Movement assessment battery for children. (2nd. ed.). APA Psyc. DOI: https://doi.org/10.1037/t55281-000
Jo, B., & Kim, S. (2022). Comparative analysis of openpose, posenet, and movenet models for pose estimation in mobile devices. Traitement du Signal, 39(14), 119–124. DOI: https://doi.org/10.18280/ts.390111
Kiphard, E., & Schilling, V. (1974). KöperKoordinationstest für kinder KTK. Weinheim: Beltz Test.
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, L. (2014). Microsoft COCO: Common Objects in Context. In D. Fleet, T. Pajdla, B. Schiele, & T. Tuytelaars (Eds.). Computer Vision – ECCV 2014 (pp. 740-755, Lecture Notes in Computer Science, Vol. 8693). Springer Cham. DOI: https://doi.org/10.1007/978-3-319-10602-1_48
Manoel, E., & Connolly, K. (1995). Variability and the development of skilled actions. International Journal Of Psychophysiology, 19, 129–147. DOI: https://doi.org/10.1016/0167-8760(94)00078-S
Marques, I., Santos, C., & Medina-Papst, J. (2017). Teste de desenvolvimento motor para educação física escolar.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, E. (2011). Scikit-learn: Machine Learning in Python. Journal Of Machine Learning Research, 12, 2825–2830.
Rosa Neto, F. (2002). Manual de avaliação motora. Artmed.
Spinosa, R. M. d. O. (2019). Demonstração digital de habilidades motoras aplicada à instrumentos de valiação do desenvolvimento motor [Doctoral dissertation, Universidade Estadual de Londrina]. Biblioteca Digital. http://www.bibliotecadigital.uel.br/document/?code=vtls000222135
Suzuki, S., Amemiya, Y., & Sato, M. (2019). Enhancement of gross-motor action recognition for children by cnn with openpose. In Institute of Electrical and Electronics Engineers, IECON 2019 [Conference]. 45th Annual Conference of The IEEE Industrial Electronics Society, Lisbon, Portugal. DOI: https://doi.org/10.1109/IECON.2019.8927828
Suzuki, S., Amemiya, Y., & Sato, M. (2020). Enhancement of child gross-motor action recognition by motional time-series images conversion. In Institute of Electrical and Electronics Engineers, IEEE/SICE International Symposium on System Integration (SII), [Symposium]. Honolulu, HI, USA. DOI: https://doi.org/10.1109/SII46433.2020.9025833
Suzuki, S., Amemiya, Y., & Sato, M. (2021). Skeletonbased visualization of poor body movements in a child’s gross-motor assessment using convolutional auto-encoder. In Institute of Electrical and Electronics Engineers, IEEE International Conference on Mechatronics (ICM) [Conference]. Kashiwa, Japan. DOI: https://doi.org/10.1109/ICM46511.2021.9385618
Ulrich, D. (2000). The test of gross motor development. PRO-ED.
Ulrich, D. (2017). Introduction to the Special Section: Evaluation of the Psychometric Properties of the TGMD-3. Journal Of Motor Learning and Development, 5, 1–4. DOI: https://doi.org/10.1123/jmld.2017-0020
Valentini, N., Zanella, L., & Webster, E. (2017). Test of Gross Motor Development—Third Edition: Establishing Content and Construct Validity for Brazilian Children. Journal of Motor Learning and Development, 5, 15–28. DOI: https://doi.org/10.1123/jmld.2016-0002
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Edson Luiz Pilati Filho, Rodrigo Martins de Oliveira Spinosa, Jacques Duílio Brancher
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The Copyright Declaration for articles published in this journal is the author’s right. Since manuscripts are published in an open access Journal, they are free to use, with their own attributions, in educational and non-commercial applications. The Journal has the right to make, in the original document, changes regarding linguistic norms, orthography, and grammar, with the purpose of ensuring the standard norms of the language and the credibility of the Journal. It will, however, respect the writing style of the authors. When necessary, conceptual changes, corrections, or suggestions will be forwarded to the authors. In such cases, the manuscript shall be subjected to a new evaluation after revision. Responsibility for the opinions expressed in the manuscripts lies entirely with the authors.
This journal is licensed with a license Creative Commons Attribution-NonCommercial 4.0 International.
