Automatic Movement Recognition for Evaluating the Gross Motor Development of Infants

Objective: The objective of this study was to early-detect gross motor abnormalities through video detection in Taiwanese infants aged 2–6 months. Background: The current diagnosis of infant developmental delays primarily relies on clinical examinations. However, during clinical visits, infants may...

Full description

Saved in:
Bibliographic Details
Main Authors: Yin-Zhang Yang, Jia-An Tsai, Ya-Lan Yu, Mary Hsin-Ju Ko, Hung-Yi Chiou, Tun-Wen Pai, Hui-Ju Chen
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Children
Subjects:
Online Access:https://www.mdpi.com/2227-9067/12/3/310
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850203808886423552
author Yin-Zhang Yang
Jia-An Tsai
Ya-Lan Yu
Mary Hsin-Ju Ko
Hung-Yi Chiou
Tun-Wen Pai
Hui-Ju Chen
author_facet Yin-Zhang Yang
Jia-An Tsai
Ya-Lan Yu
Mary Hsin-Ju Ko
Hung-Yi Chiou
Tun-Wen Pai
Hui-Ju Chen
author_sort Yin-Zhang Yang
collection DOAJ
description Objective: The objective of this study was to early-detect gross motor abnormalities through video detection in Taiwanese infants aged 2–6 months. Background: The current diagnosis of infant developmental delays primarily relies on clinical examinations. However, during clinical visits, infants may show atypical behaviors due to unfamiliar environments, which might not truly reflect their true developmental status. Methods: This study utilized videos of infants recorded in their home environments. Two pediatric neurologists manually annotated these clips to identify whether an infant possessed the characteristics of gross motor delays through an assessment of his/her gross motor movements. Using transfer learning techniques, four pose recognition models, including ViTPose, HRNet, DARK, and UDP, were applied to the infant gross motor dataset. Four machine learning classification models, including random forest, support vector machine, logistic regression, and XGBoost, were used to predict the developmental status of infants. Results: The experimental results of pose estimation and tracking indicate that the ViTPose model provided the best performance for pose recognition. A total of 227 features related to kinematics, motions, and postures were extracted and calculated. A one-way ANOVA analysis revealed 106 significant features that were retained for constructing prediction models. The results show that a random forest model achieved the best performance with an average F1-score of 0.94, a weighted average AUC of 0.98, and an average accuracy of 94%.
format Article
id doaj-art-4abd48ec297343cf911db73da94426db
institution OA Journals
issn 2227-9067
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Children
spelling doaj-art-4abd48ec297343cf911db73da94426db2025-08-20T02:11:25ZengMDPI AGChildren2227-90672025-02-0112331010.3390/children12030310Automatic Movement Recognition for Evaluating the Gross Motor Development of InfantsYin-Zhang Yang0Jia-An Tsai1Ya-Lan Yu2Mary Hsin-Ju Ko3Hung-Yi Chiou4Tun-Wen Pai5Hui-Ju Chen6Department of Computer Science and Information Engineering, National Taipei University of Technology, Taipei 10608, TaiwanDepartment of Computer Science and Information Engineering, National Taipei University of Technology, Taipei 10608, TaiwanDivision of Pediatric Neurology, Department of Pediatrics, MacKay Children Hospital, Taipei 104217, TaiwanDepartment of Pediatric Neurology, Hsinchu Municipal MacKay Children Hospital, Hsinchu 300195, TaiwanInstitute of Population Health Science, National Health Research Institute, Miaoli County 350401, TaiwanDepartment of Computer Science and Information Engineering, National Taipei University of Technology, Taipei 10608, TaiwanDivision of Pediatric Neurology, Department of Pediatrics, MacKay Children Hospital, Taipei 104217, TaiwanObjective: The objective of this study was to early-detect gross motor abnormalities through video detection in Taiwanese infants aged 2–6 months. Background: The current diagnosis of infant developmental delays primarily relies on clinical examinations. However, during clinical visits, infants may show atypical behaviors due to unfamiliar environments, which might not truly reflect their true developmental status. Methods: This study utilized videos of infants recorded in their home environments. Two pediatric neurologists manually annotated these clips to identify whether an infant possessed the characteristics of gross motor delays through an assessment of his/her gross motor movements. Using transfer learning techniques, four pose recognition models, including ViTPose, HRNet, DARK, and UDP, were applied to the infant gross motor dataset. Four machine learning classification models, including random forest, support vector machine, logistic regression, and XGBoost, were used to predict the developmental status of infants. Results: The experimental results of pose estimation and tracking indicate that the ViTPose model provided the best performance for pose recognition. A total of 227 features related to kinematics, motions, and postures were extracted and calculated. A one-way ANOVA analysis revealed 106 significant features that were retained for constructing prediction models. The results show that a random forest model achieved the best performance with an average F1-score of 0.94, a weighted average AUC of 0.98, and an average accuracy of 94%.https://www.mdpi.com/2227-9067/12/3/310developmental delaypose estimationtransfer learningmachine learning
spellingShingle Yin-Zhang Yang
Jia-An Tsai
Ya-Lan Yu
Mary Hsin-Ju Ko
Hung-Yi Chiou
Tun-Wen Pai
Hui-Ju Chen
Automatic Movement Recognition for Evaluating the Gross Motor Development of Infants
Children
developmental delay
pose estimation
transfer learning
machine learning
title Automatic Movement Recognition for Evaluating the Gross Motor Development of Infants
title_full Automatic Movement Recognition for Evaluating the Gross Motor Development of Infants
title_fullStr Automatic Movement Recognition for Evaluating the Gross Motor Development of Infants
title_full_unstemmed Automatic Movement Recognition for Evaluating the Gross Motor Development of Infants
title_short Automatic Movement Recognition for Evaluating the Gross Motor Development of Infants
title_sort automatic movement recognition for evaluating the gross motor development of infants
topic developmental delay
pose estimation
transfer learning
machine learning
url https://www.mdpi.com/2227-9067/12/3/310
work_keys_str_mv AT yinzhangyang automaticmovementrecognitionforevaluatingthegrossmotordevelopmentofinfants
AT jiaantsai automaticmovementrecognitionforevaluatingthegrossmotordevelopmentofinfants
AT yalanyu automaticmovementrecognitionforevaluatingthegrossmotordevelopmentofinfants
AT maryhsinjuko automaticmovementrecognitionforevaluatingthegrossmotordevelopmentofinfants
AT hungyichiou automaticmovementrecognitionforevaluatingthegrossmotordevelopmentofinfants
AT tunwenpai automaticmovementrecognitionforevaluatingthegrossmotordevelopmentofinfants
AT huijuchen automaticmovementrecognitionforevaluatingthegrossmotordevelopmentofinfants