We present a series of new and improved features, and a feature fusion pipeline for this classification task. We also introduce the RVI-38 dataset, a series of videos captured as part of routine clinical care. By utilising this challenging dataset we establish the robustness of several motion features for classification, subsequently informing the design of our proposed feature fusion framework based upon the General Movements Assessment (GMA).
The features, classification frameowrk and the dataset are presented in the article (see the Citation section). Since we are still prearing the final version of the article, the code will be available soon. Stay tuned!
For accessing the dataset (skeletal pose sequences with annotations), please contact Edmond S. L. Ho ([email protected]).
Please cite these papers in your publications if it helps your research:
@ARTICLE{McCay:TNSRE2022,
author={McCay, Kevin D. and Hu, Pengpeng and Shum, Hubert P. H. and Woo, Wai Lok and Marcroft, Claire and Embleton, Nicholas D. and Munteanu, Adrian and Ho, Edmond S. L.},
journal={IEEE Transactions on Neural Systems and Rehabilitation Engineering},
title={A Pose-Based Feature Fusion and Classification Framework for the Early Prediction of Cerebral Palsy in Infants},
year={2022},
volume={30},
number={},
pages={8-19},
doi={10.1109/TNSRE.2021.3138185}}
}
@INPROCEEDINGS{McCay:EMBC2019,
author={McCay, Kevin D. and Ho, Edmond S. L. and Marcroft, Claire and Embleton, Nicholas D.},
booktitle={2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)},
title={Establishing Pose Based Features Using Histograms for the Detection of Abnormal Infant Movements},
year={2019},
pages={5469-5472},
doi={10.1109/EMBC.2019.8857680}
}
The program is developed by Kevin McCay ([email protected]) and Edmond S. L. Ho ([email protected]). Currently, it is being maintained by Edmond S. L. Ho.