Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model

Okita, Yasutaka; Hirano, Toru; Wang, Bowen; Nakashima, Yuta; Minoda, Saki; Nagahara, Hajime; Kumanogoh, Atsushi

doi:https://doi.org/10.1186/s13075-023-03172-x

Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model

Okita, Yasutaka; Hirano, Toru; Wang, Bowen et al.

Arthritis Research and Therapy, 2023, 25(1), 181

Number of Access:1,119（2025-08-14 03:52 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/11094/93229

Link to primary information

File	Format	Terms of use	Size	Views	Date.Available	Description	information
ArthritisResTher_25_1	pdf	None	1.38 MB	67	2023.12.05

Creative Commons : Attribution

Item Information

Output File Export EndNote Basic Export Mendeley

Title	Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model
Creator	researchmap permalink y_okita y_okita Scopus Author ID 57206222555 57206222555 NRID 1000020797039 1000020797039 CreatorName Okita, Yasutaka
	NRID 1000010437402 1000010437402 CreatorName Hirano, Toru
	ORCID 0000-0002-2911-5595 0000-0002-2911-5595 researchmap permalink wbw520 wbw520 CreatorName Wang, Bowen
	NRID 1000070633551 1000070633551 ORCID 0000-0001-8000-3567 0000-0001-8000-3567 researchmap permalink n-yuta n-yuta Scopus Author ID 56022946900 56022946900 CreatorName Nakashima, Yuta
	CreatorName Minoda, Saki
	NRID 1000080362648 1000080362648 researchmap permalink read0054705 read0054705 Scopus Author ID 7006675539 7006675539 ORCID 0000-0003-1579-8767 0000-0003-1579-8767 CreatorName Nagahara, Hajime
	NRID 1000010294125 1000010294125 researchmap permalink read0051725 read0051725 Scopus Author ID 6602726311 6602726311 ORCID 0000-0003-4749-7117 0000-0003-4749-7117 CreatorName Kumanogoh, Atsushi
Subject	Atlantoaxial subluxation
	Cervical spine
	Deep learning
	Machine learning
	Rheumatoid arthritis
Description	Okita Y., Hirano T., Wang B., et al. Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model. Arthritis Research and Therapy 25, 181 (2023); 10.1186/s13075-023-03172-x.
Abstract	Background: This work aims to develop a deep learning model, assessing atlantoaxial subluxation (AAS) in rheumatoid arthritis (RA), which can often be ambiguous in clinical practice. Methods: We collected 4691 X-ray images of the cervical spine of the 906 patients with RA. Among these images, 3480 were used for training the deep learning model, 803 were used for validating the model during the training process, and the remaining 408 were used for testing the performance of the trained model. The two-dimensional key points’ detection model of Deep High-Resolution Representation Learning for Human Pose Estimation was adopted as the base convolutional neural network model. The model inferred four coordinates to calculate the atlantodental interval (ADI) and space available for the spinal cord (SAC). Finally, these values were compared with those by clinicians to evaluate the performance of the model. Results: Among the 408 cervical images for testing the performance, the trained model correctly identified the four coordinates in 99.5% of the dataset. The values of ADI and SAC were positively correlated among the model and two clinicians. The sensitivity of AAS diagnosis with ADI or SAC by the model was 0.86 and 0.97 respectively. The specificity of that was 0.57 and 0.5 respectively. Conclusions: We present the development of a deep learning model for the evaluation of cervical lesions of patients with RA. The model was demonstrably shown to be useful for quantitative evaluation.
Publisher	BioMed Central Ltd
Source Title	Arthritis Research and Therapy
Volume (Issue)	25(1)
Page	181
Date of Issued	2023-09-25
Language	English
Handle URL	https://hdl.handle.net/11094/93229
PISSN	14786354
eISSN	14786362
Relation.isIdenticalTo	DOI (Publisher Version) https://doi.org/10.1186/s13075-023-03172-x
Access Rights	open access
CopyRight	This article is licensed under a Creative Commons Attribution 4.0 International License.
oaire:version	Version of Record
Category	学術雑誌論文 Journal Article

Resource Type	journal article
Local Type	Journal Article
DCMI Type	Text
DCTERMS.bibliographicCitation	Arthritis Research and Therapy. 2023, 25(1), p. 181
DC.title	Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model
DC.creator	Okita, Yasutaka
	Hirano, Toru
	Wang, Bowen
	Nakashima, Yuta
	Minoda, Saki
	Nagahara, Hajime
	Kumanogoh, Atsushi
DC.publisher	BioMed Central Ltd
DC.language' scheme='DCTERMS.RFC1766	English
DC.type' scheme='DCTERMS.DCMIType	Text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2023-09-25
DC.identifier	https://doi.org/10.1186/s13075-023-03172-x
DC.subject	Atlantoaxial subluxation
	Cervical spine
	Deep learning
	Machine learning
	Rheumatoid arthritis
DC.description	Okita Y., Hirano T., Wang B., et al. Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model. Arthritis Research and Therapy 25, 181 (2023); 10.1186/s13075-023-03172-x.
DCTERMS.abstract	Background: This work aims to develop a deep learning model, assessing atlantoaxial subluxation (AAS) in rheumatoid arthritis (RA), which can often be ambiguous in clinical practice. Methods: We collected 4691 X-ray images of the cervical spine of the 906 patients with RA. Among these images, 3480 were used for training the deep learning model, 803 were used for validating the model during the training process, and the remaining 408 were used for testing the performance of the trained model. The two-dimensional key points’ detection model of Deep High-Resolution Representation Learning for Human Pose Estimation was adopted as the base convolutional neural network model. The model inferred four coordinates to calculate the atlantodental interval (ADI) and space available for the spinal cord (SAC). Finally, these values were compared with those by clinicians to evaluate the performance of the model. Results: Among the 408 cervical images for testing the performance, the trained model correctly identified the four coordinates in 99.5% of the dataset. The values of ADI and SAC were positively correlated among the model and two clinicians. The sensitivity of AAS diagnosis with ADI or SAC by the model was 0.86 and 0.97 respectively. The specificity of that was 0.57 and 0.5 respectively. Conclusions: We present the development of a deep learning model for the evaluation of cervical lesions of patients with RA. The model was demonstrably shown to be useful for quantitative evaluation.
DC.rights	This article is licensed under a Creative Commons Attribution 4.0 International License.
DC.format	93229
citation_title	Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model
citation_author	Okita, Yasutaka
	Hirano, Toru
	Wang, Bowen
	Nakashima, Yuta
	Minoda, Saki
	Nagahara, Hajime
	Kumanogoh, Atsushi
citation_publisher	BioMed Central Ltd
citation_language	English
citation_date	2023-09-25
citation_journal_title	Arthritis Research and Therapy
citation_volume	25
citation_issue	1
citation_firstpage	181
citation_doi	https://doi.org/10.1186/s13075-023-03172-x
citation_public_url	https://hdl.handle.net/11094/93229

Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model

Link to primary information

Item Information

Show details