Towards Explainable Abnormal Infant Movements Identification: A Body-part Based Prediction and Visualisation Framework

Kevin McCay; Edmond S. L. Ho; Dimitrios Sakkos; Wai Lok Woo; Claire Marcroft; Patricia Dulson; Nicholas D. Embleton

doi:10.1109/BHI50953.2021.9508603

Towards Explainable Abnormal Infant Movements Identification: A Body-part Based Prediction and Visualisation Framework

Kevin McCay, Edmond S. L. Ho^*, Dimitrios Sakkos, Wai Lok Woo, Claire Marcroft, Patricia Dulson, Nicholas D. Embleton

^*Corresponding author for this work

Computer and Information Sciences Department

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

8 Citations (Scopus)

45 Downloads (Pure)

Abstract

Providing early diagnosis of cerebral palsy (CP) is key to enhancing the developmental outcomes for those affected. Diagnostic tools such as the General Movements Assessment (GMA), have produced promising results in early diagnosis, however these manual methods can be laborious.

In this paper, we propose a new framework for the automated classification of infant body movements, based upon the GMA, which unlike previous methods, also incorporates a visualization framework to aid with interpretability. Our proposed framework segments extracted features to detect the presence of Fidgety Movements (FMs) associated with the GMA spatiotemporally. These features are then used to identify the body-parts with the greatest contribution towards a classification decision and highlight the related body-part segment providing visual feedback to the user.

We quantitatively compare the proposed framework's classification performance with several other methods from the literature and qualitatively evaluate the visualization's veracity. Our experimental results show that the proposed method performs more robustly than comparable techniques in this setting whilst simultaneously providing relevant visual interpretability.

Original language	English
Title of host publication	2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI)
Place of Publication	Piscataway
Publisher	IEEE
Number of pages	4
ISBN (Electronic)	9781665447706
ISBN (Print)	9781665403580
DOIs	https://doi.org/10.1109/BHI50953.2021.9508603
Publication status	Published - 27 Jul 2021
Event	IEEE International Conference on Biomedical and Health Informatics (BHI) : Reshaping healthcare through advanced AI-enabled health informatics for a better quality of life - Virtual Duration: 27 Jul 2021 → 30 Jul 2021 https://www.bhi-bsn-2021.org/?page_id=2336

Publication series

Name	IEEE EMBS International Conference on Biomedical and Health Informatics
Publisher	IEEE
ISSN (Print)	2641-3590
ISSN (Electronic)	2641-3604

Conference

Conference	IEEE International Conference on Biomedical and Health Informatics (BHI)
Abbreviated title	IEEE BHI 2021
Period	27/07/21 → 30/07/21
Internet address	https://www.bhi-bsn-2021.org/?page_id=2336

Keywords

cerebral palsy
general movements assessment
machine learning
explainable AI
visualization

Access to Document

10.1109/BHI50953.2021.9508603

IEEE_BHI_2021 - accepted versionAccepted author manuscript, 2.82 MB

8 Citations
1 Article

A Pose-based Feature Fusion and Classification Framework for the Early Prediction of Cerebral Palsy in Infants
McCay, K. D., Hu, P., Shum, H., Woo, W. L., Marcroft, C., Embleton, N. D., Munteanu, A. & Ho, E. S. L., 28 Jan 2022, In: IEEE Transactions on Neural Systems and Rehabilitation Engineering. 30, p. 8-19 12 p.
Research output: Contribution to journal › Article › peer-review

Open Access
File
31 Citations (Scopus)

35 Downloads (Pure)

1 Finished

Autonomous Monitoring for Patients and Older People using Smart Environments with Sensor Fusion
Ho, E. (PI)
The Royal Society
19/08/19 → 18/08/21
Project: Research

Cite this

McCay, K., Ho, E. S. L., Sakkos, D., Woo, W. L., Marcroft, C., Dulson, P., & Embleton, N. D. (2021). Towards Explainable Abnormal Infant Movements Identification: A Body-part Based Prediction and Visualisation Framework. In 2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI) Article 9508603 (IEEE EMBS International Conference on Biomedical and Health Informatics). IEEE. https://doi.org/10.1109/BHI50953.2021.9508603

@inproceedings{63215dda816c453b911d539888a3bb2e,

title = "Towards Explainable Abnormal Infant Movements Identification: A Body-part Based Prediction and Visualisation Framework",

abstract = "Providing early diagnosis of cerebral palsy (CP) is key to enhancing the developmental outcomes for those affected. Diagnostic tools such as the General Movements Assessment (GMA), have produced promising results in early diagnosis, however these manual methods can be laborious. In this paper, we propose a new framework for the automated classification of infant body movements, based upon the GMA, which unlike previous methods, also incorporates a visualization framework to aid with interpretability. Our proposed framework segments extracted features to detect the presence of Fidgety Movements (FMs) associated with the GMA spatiotemporally. These features are then used to identify the body-parts with the greatest contribution towards a classification decision and highlight the related body-part segment providing visual feedback to the user.We quantitatively compare the proposed framework's classification performance with several other methods from the literature and qualitatively evaluate the visualization's veracity. Our experimental results show that the proposed method performs more robustly than comparable techniques in this setting whilst simultaneously providing relevant visual interpretability.",

keywords = "cerebral palsy, general movements assessment, machine learning, explainable AI, visualization",

author = "Kevin McCay and Ho, {Edmond S. L.} and Dimitrios Sakkos and Woo, {Wai Lok} and Claire Marcroft and Patricia Dulson and Embleton, {Nicholas D.}",

note = "This project is supported in part by the Royal Society (Ref: IES/R1/191147).; IEEE International Conference on Biomedical and Health Informatics (BHI) : Reshaping healthcare through advanced AI-enabled health informatics for a better quality of life, IEEE BHI 2021 ; Conference date: 27-07-2021 Through 30-07-2021",

year = "2021",

month = jul,

day = "27",

doi = "10.1109/BHI50953.2021.9508603",

language = "English",

isbn = "9781665403580",

series = "IEEE EMBS International Conference on Biomedical and Health Informatics",

publisher = "IEEE",

booktitle = "2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI)",

address = "United States",

url = "https://www.bhi-bsn-2021.org/?page_id=2336",

}

McCay, K, Ho, ESL, Sakkos, D, Woo, WL, Marcroft, C, Dulson, P & Embleton, ND 2021, Towards Explainable Abnormal Infant Movements Identification: A Body-part Based Prediction and Visualisation Framework. in 2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI)., 9508603, IEEE EMBS International Conference on Biomedical and Health Informatics, IEEE, Piscataway, IEEE International Conference on Biomedical and Health Informatics (BHI) , 27/07/21. https://doi.org/10.1109/BHI50953.2021.9508603

Towards Explainable Abnormal Infant Movements Identification: A Body-part Based Prediction and Visualisation Framework. / McCay, Kevin; Ho, Edmond S. L.; Sakkos, Dimitrios et al.
2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI). Piscataway: IEEE, 2021. 9508603 (IEEE EMBS International Conference on Biomedical and Health Informatics).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Towards Explainable Abnormal Infant Movements Identification

T2 - IEEE International Conference on Biomedical and Health Informatics (BHI)

AU - McCay, Kevin

AU - Ho, Edmond S. L.

AU - Sakkos, Dimitrios

AU - Woo, Wai Lok

AU - Marcroft, Claire

AU - Dulson, Patricia

AU - Embleton, Nicholas D.

N1 - This project is supported in part by the Royal Society (Ref: IES/R1/191147).

PY - 2021/7/27

Y1 - 2021/7/27

N2 - Providing early diagnosis of cerebral palsy (CP) is key to enhancing the developmental outcomes for those affected. Diagnostic tools such as the General Movements Assessment (GMA), have produced promising results in early diagnosis, however these manual methods can be laborious. In this paper, we propose a new framework for the automated classification of infant body movements, based upon the GMA, which unlike previous methods, also incorporates a visualization framework to aid with interpretability. Our proposed framework segments extracted features to detect the presence of Fidgety Movements (FMs) associated with the GMA spatiotemporally. These features are then used to identify the body-parts with the greatest contribution towards a classification decision and highlight the related body-part segment providing visual feedback to the user.We quantitatively compare the proposed framework's classification performance with several other methods from the literature and qualitatively evaluate the visualization's veracity. Our experimental results show that the proposed method performs more robustly than comparable techniques in this setting whilst simultaneously providing relevant visual interpretability.

AB - Providing early diagnosis of cerebral palsy (CP) is key to enhancing the developmental outcomes for those affected. Diagnostic tools such as the General Movements Assessment (GMA), have produced promising results in early diagnosis, however these manual methods can be laborious. In this paper, we propose a new framework for the automated classification of infant body movements, based upon the GMA, which unlike previous methods, also incorporates a visualization framework to aid with interpretability. Our proposed framework segments extracted features to detect the presence of Fidgety Movements (FMs) associated with the GMA spatiotemporally. These features are then used to identify the body-parts with the greatest contribution towards a classification decision and highlight the related body-part segment providing visual feedback to the user.We quantitatively compare the proposed framework's classification performance with several other methods from the literature and qualitatively evaluate the visualization's veracity. Our experimental results show that the proposed method performs more robustly than comparable techniques in this setting whilst simultaneously providing relevant visual interpretability.

KW - cerebral palsy

KW - general movements assessment

KW - machine learning

KW - explainable AI

KW - visualization

U2 - 10.1109/BHI50953.2021.9508603

DO - 10.1109/BHI50953.2021.9508603

M3 - Conference contribution

SN - 9781665403580

T3 - IEEE EMBS International Conference on Biomedical and Health Informatics

BT - 2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI)

PB - IEEE

CY - Piscataway

Y2 - 27 July 2021 through 30 July 2021

ER -

McCay K, Ho ESL, Sakkos D, Woo WL, Marcroft C, Dulson P et al. Towards Explainable Abnormal Infant Movements Identification: A Body-part Based Prediction and Visualisation Framework. In 2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI). Piscataway: IEEE. 2021. 9508603. (IEEE EMBS International Conference on Biomedical and Health Informatics). doi: 10.1109/BHI50953.2021.9508603

Towards Explainable Abnormal Infant Movements Identification: A Body-part Based Prediction and Visualisation Framework

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Research output

A Pose-based Feature Fusion and Classification Framework for the Early Prediction of Cerebral Palsy in Infants

Projects

Autonomous Monitoring for Patients and Older People using Smart Environments with Sensor Fusion

Cite this