Abnormal Infant Movements Classification with Deep Learning on Pose-based Features

Edmond S. L. Ho*, Kevin McCay, Hubert P. H. Shum, Gerhard Fehringer, Claire Marcroft, Nicholas Embleton

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

58 Citations (Scopus)
125 Downloads (Pure)

Abstract

The pursuit of early diagnosis of cerebral palsy has been an active research area with some very promising results using tools such as the General Movements Assessment (GMA). In our previous work, we explored the feasibility of extracting pose-based features from video sequences to automatically classify infant body movement into two categories, normal and abnormal. The classification was based upon the GMA, which was carried out on the video data by an independent expert reviewer. In this paper we extend our previous work by extracting the normalised pose-based feature sets, Histograms of Joint Orientation 2D (HOJO2D) and Histograms of Joint Displacement 2D (HOJD2D), for use in new deep learning architectures. We explore the viability of using these pose-based feature sets for automated classification within a deep learning framework by carrying out extensive experiments on five new deep learning architectures. Experimental results show that the proposed fully connected neural network FCNet performed robustly across different feature sets. Furthermore, the proposed convolutional neural network architectures demonstrated excellent performance in handling features in higher dimensionality. We make the code, extracted features and associated GMA labels publicly available.

Original languageEnglish
Article number9034058
Pages (from-to)51582-51592
Number of pages11
JournalIEEE Access
Volume8
Early online date12 Mar 2020
DOIs
Publication statusPublished - 24 Mar 2020

Keywords

  • Deep learning
  • classification
  • feature extraction
  • infants
  • pose-based features

Fingerprint

Dive into the research topics of 'Abnormal Infant Movements Classification with Deep Learning on Pose-based Features'. Together they form a unique fingerprint.

Cite this