Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs

Abstract Human spinal balance assessment relies considerably on sagittal radiographic parameter measurement. Deep learning could be applied for automatic landmark detection and alignment analysis, with mild to moderate standard errors and favourable correlations with manual measurement. In this stud...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Yu-Cheng Yeh, Chi-Hung Weng, Yu-Jui Huang, Chen-Ju Fu, Tsung-Ting Tsai, Chao-Yuan Yeh
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/81f7b934fe4245148f554804ed105461
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:81f7b934fe4245148f554804ed105461
record_format dspace
spelling oai:doaj.org-article:81f7b934fe4245148f554804ed1054612021-12-02T14:15:53ZDeep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs10.1038/s41598-021-87141-x2045-2322https://doaj.org/article/81f7b934fe4245148f554804ed1054612021-04-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-87141-xhttps://doaj.org/toc/2045-2322Abstract Human spinal balance assessment relies considerably on sagittal radiographic parameter measurement. Deep learning could be applied for automatic landmark detection and alignment analysis, with mild to moderate standard errors and favourable correlations with manual measurement. In this study, based on 2210 annotated images of various spinal disease aetiologies, we developed deep learning models capable of automatically locating 45 anatomic landmarks and subsequently generating 18 radiographic parameters on a whole-spine lateral radiograph. In the assessment of model performance, the localisation accuracy and learning speed were the highest for landmarks in the cervical area, followed by those in the lumbosacral, thoracic, and femoral areas. All the predicted radiographic parameters were significantly correlated with ground truth values (all p < 0.001). The human and artificial intelligence comparison revealed that the deep learning model was capable of matching the reliability of doctors for 15/18 of the parameters. The proposed automatic alignment analysis system was able to localise spinal anatomic landmarks with high accuracy and to generate various radiographic parameters with favourable correlations with manual measurements.Yu-Cheng YehChi-Hung WengYu-Jui HuangChen-Ju FuTsung-Ting TsaiChao-Yuan YehNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-15 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Yu-Cheng Yeh
Chi-Hung Weng
Yu-Jui Huang
Chen-Ju Fu
Tsung-Ting Tsai
Chao-Yuan Yeh
Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs
description Abstract Human spinal balance assessment relies considerably on sagittal radiographic parameter measurement. Deep learning could be applied for automatic landmark detection and alignment analysis, with mild to moderate standard errors and favourable correlations with manual measurement. In this study, based on 2210 annotated images of various spinal disease aetiologies, we developed deep learning models capable of automatically locating 45 anatomic landmarks and subsequently generating 18 radiographic parameters on a whole-spine lateral radiograph. In the assessment of model performance, the localisation accuracy and learning speed were the highest for landmarks in the cervical area, followed by those in the lumbosacral, thoracic, and femoral areas. All the predicted radiographic parameters were significantly correlated with ground truth values (all p < 0.001). The human and artificial intelligence comparison revealed that the deep learning model was capable of matching the reliability of doctors for 15/18 of the parameters. The proposed automatic alignment analysis system was able to localise spinal anatomic landmarks with high accuracy and to generate various radiographic parameters with favourable correlations with manual measurements.
format article
author Yu-Cheng Yeh
Chi-Hung Weng
Yu-Jui Huang
Chen-Ju Fu
Tsung-Ting Tsai
Chao-Yuan Yeh
author_facet Yu-Cheng Yeh
Chi-Hung Weng
Yu-Jui Huang
Chen-Ju Fu
Tsung-Ting Tsai
Chao-Yuan Yeh
author_sort Yu-Cheng Yeh
title Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs
title_short Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs
title_full Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs
title_fullStr Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs
title_full_unstemmed Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs
title_sort deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/81f7b934fe4245148f554804ed105461
work_keys_str_mv AT yuchengyeh deeplearningapproachforautomaticlandmarkdetectionandalignmentanalysisinwholespinelateralradiographs
AT chihungweng deeplearningapproachforautomaticlandmarkdetectionandalignmentanalysisinwholespinelateralradiographs
AT yujuihuang deeplearningapproachforautomaticlandmarkdetectionandalignmentanalysisinwholespinelateralradiographs
AT chenjufu deeplearningapproachforautomaticlandmarkdetectionandalignmentanalysisinwholespinelateralradiographs
AT tsungtingtsai deeplearningapproachforautomaticlandmarkdetectionandalignmentanalysisinwholespinelateralradiographs
AT chaoyuanyeh deeplearningapproachforautomaticlandmarkdetectionandalignmentanalysisinwholespinelateralradiographs
_version_ 1718391736836816896