Cargando…

SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation

The Cobb angle (CA) serves as the principal method for assessing spinal deformity, but manual measurements of the CA are time-consuming and susceptible to inter- and intra-observer variability. While learning-based methods, such as SpineHRNet+, have demonstrated potential in automating CA measuremen...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhao, Moxin, Meng, Nan, Cheung, Jason Pui Yin, Yu, Chenxi, Lu, Pengyu, Zhang, Teng
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10669780/
https://www.ncbi.nlm.nih.gov/pubmed/38002457
http://dx.doi.org/10.3390/bioengineering10111333
_version_ 1785149252529291264
author Zhao, Moxin
Meng, Nan
Cheung, Jason Pui Yin
Yu, Chenxi
Lu, Pengyu
Zhang, Teng
author_facet Zhao, Moxin
Meng, Nan
Cheung, Jason Pui Yin
Yu, Chenxi
Lu, Pengyu
Zhang, Teng
author_sort Zhao, Moxin
collection PubMed
description The Cobb angle (CA) serves as the principal method for assessing spinal deformity, but manual measurements of the CA are time-consuming and susceptible to inter- and intra-observer variability. While learning-based methods, such as SpineHRNet+, have demonstrated potential in automating CA measurement, their accuracy can be influenced by the severity of spinal deformity, image quality, relative position of rib and vertebrae, etc. Our aim is to create a reliable learning-based approach that provides consistent and highly accurate measurements of the CA from posteroanterior (PA) X-rays, surpassing the state-of-the-art method. To accomplish this, we introduce SpineHRformer, which identifies anatomical landmarks, including the vertices of endplates from the 7th cervical vertebra (C7) to the 5th lumbar vertebra (L5) and the end vertebrae with different output heads, enabling the calculation of CAs. Within our SpineHRformer, a backbone HRNet first extracts multi-scale features from the input X-ray, while transformer blocks extract local and global features from the HRNet outputs. Subsequently, an output head to generate heatmaps of the endplate landmarks or end vertebra landmarks facilitates the computation of CAs. We used a dataset of 1934 PA X-rays with diverse degrees of spinal deformity and image quality, following an 8:2 ratio to train and test the model. The experimental results indicate that SpineHRformer outperforms SpineHRNet+ in landmark detection (Mean Euclidean Distance: 2.47 pixels vs. 2.74 pixels), CA prediction (Pearson correlation coefficient: 0.86 vs. 0.83), and severity grading (sensitivity: normal-mild; 0.93 vs. 0.74, moderate; 0.74 vs. 0.77, severe; 0.74 vs. 0.7). Our approach demonstrates greater robustness and accuracy compared to SpineHRNet+, offering substantial potential for improving the efficiency and reliability of CA measurements in clinical settings.
format Online
Article
Text
id pubmed-10669780
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-106697802023-11-20 SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation Zhao, Moxin Meng, Nan Cheung, Jason Pui Yin Yu, Chenxi Lu, Pengyu Zhang, Teng Bioengineering (Basel) Article The Cobb angle (CA) serves as the principal method for assessing spinal deformity, but manual measurements of the CA are time-consuming and susceptible to inter- and intra-observer variability. While learning-based methods, such as SpineHRNet+, have demonstrated potential in automating CA measurement, their accuracy can be influenced by the severity of spinal deformity, image quality, relative position of rib and vertebrae, etc. Our aim is to create a reliable learning-based approach that provides consistent and highly accurate measurements of the CA from posteroanterior (PA) X-rays, surpassing the state-of-the-art method. To accomplish this, we introduce SpineHRformer, which identifies anatomical landmarks, including the vertices of endplates from the 7th cervical vertebra (C7) to the 5th lumbar vertebra (L5) and the end vertebrae with different output heads, enabling the calculation of CAs. Within our SpineHRformer, a backbone HRNet first extracts multi-scale features from the input X-ray, while transformer blocks extract local and global features from the HRNet outputs. Subsequently, an output head to generate heatmaps of the endplate landmarks or end vertebra landmarks facilitates the computation of CAs. We used a dataset of 1934 PA X-rays with diverse degrees of spinal deformity and image quality, following an 8:2 ratio to train and test the model. The experimental results indicate that SpineHRformer outperforms SpineHRNet+ in landmark detection (Mean Euclidean Distance: 2.47 pixels vs. 2.74 pixels), CA prediction (Pearson correlation coefficient: 0.86 vs. 0.83), and severity grading (sensitivity: normal-mild; 0.93 vs. 0.74, moderate; 0.74 vs. 0.77, severe; 0.74 vs. 0.7). Our approach demonstrates greater robustness and accuracy compared to SpineHRNet+, offering substantial potential for improving the efficiency and reliability of CA measurements in clinical settings. MDPI 2023-11-20 /pmc/articles/PMC10669780/ /pubmed/38002457 http://dx.doi.org/10.3390/bioengineering10111333 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Zhao, Moxin
Meng, Nan
Cheung, Jason Pui Yin
Yu, Chenxi
Lu, Pengyu
Zhang, Teng
SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation
title SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation
title_full SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation
title_fullStr SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation
title_full_unstemmed SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation
title_short SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation
title_sort spinehrformer: a transformer-based deep learning model for automatic spine deformity assessment with prospective validation
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10669780/
https://www.ncbi.nlm.nih.gov/pubmed/38002457
http://dx.doi.org/10.3390/bioengineering10111333
work_keys_str_mv AT zhaomoxin spinehrformeratransformerbaseddeeplearningmodelforautomaticspinedeformityassessmentwithprospectivevalidation
AT mengnan spinehrformeratransformerbaseddeeplearningmodelforautomaticspinedeformityassessmentwithprospectivevalidation
AT cheungjasonpuiyin spinehrformeratransformerbaseddeeplearningmodelforautomaticspinedeformityassessmentwithprospectivevalidation
AT yuchenxi spinehrformeratransformerbaseddeeplearningmodelforautomaticspinedeformityassessmentwithprospectivevalidation
AT lupengyu spinehrformeratransformerbaseddeeplearningmodelforautomaticspinedeformityassessmentwithprospectivevalidation
AT zhangteng spinehrformeratransformerbaseddeeplearningmodelforautomaticspinedeformityassessmentwithprospectivevalidation