An Automated and Accurate Spine Curve Analysis System
We present a new Adaptive Error Correction Net (AEC-Net) to formulate the estimation of Cobb anges from spinal X-rays as a high-precision regression task. Our AEC-Net introduces two novel innovations. (1) The AEC-Net contains two networks calculating landmarks and Cobb angles separately, which robus...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2019-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8819955/ |
id |
doaj-4e34115269844e3eb1d0c6a57446ba5e |
---|---|
record_format |
Article |
spelling |
doaj-4e34115269844e3eb1d0c6a57446ba5e2021-03-29T23:18:39ZengIEEEIEEE Access2169-35362019-01-01712459612460510.1109/ACCESS.2019.29384028819955An Automated and Accurate Spine Curve Analysis SystemBo Chen0Qiuhao Xu1Liansheng Wang2https://orcid.org/0000-0002-2096-454XStephanie Leung3Jonathan Chung4Shuo Li5https://orcid.org/0000-0002-5184-3230Department of Medical Imaging, Digital Imaging Group of London, Western University, London, CanadaDepartment of Computer Science, Xiamen University, Xiamen, ChinaDepartment of Computer Science, Xiamen University, Xiamen, ChinaDepartment of Medical Imaging, Western University, London, CanadaDepartment of Medical Imaging, Western University, London, CanadaDepartment of Medical Imaging, Digital Imaging Group of London, Western University, London, CanadaWe present a new Adaptive Error Correction Net (AEC-Net) to formulate the estimation of Cobb anges from spinal X-rays as a high-precision regression task. Our AEC-Net introduces two novel innovations. (1) The AEC-Net contains two networks calculating landmarks and Cobb angles separately, which robustly solve the disadvantage of ambiguity in X-rays since these networks focus on more features. It effectively handles the nonlinear relationship between input images and quantitative outputs, while explicitly capturing the intrinsic features of input images. (2) Based on the two estimated angles, the AEC-Net proposed a new loss function to calculate the final Cobb angles. The optimization of the loss function is based on a high-precision calculation method. The deep learning structure is used to complete this optimization, which achieves higher accuracy and efficiency. We validate our method with the spinal X-rays dataset of 581 subjects with signs of scoliosis at varying extents. The proposed method achieves high accuracy and robustness on the Cobb angle estimations. Comparing to the exsiting conventional methods suffering from tremendous variability and low reliability caused by high ambiguity and variability around boundaries of the vertebrae, the AEC-Net obtain Cobb angles accurately and robustly, which indicates its great potential in clinical use. The highly accurate Cobb angles produced by our framework can be used by clinicians for comprehensive scoliosis assessment, and possibly be further extended to other clinical applications.https://ieeexplore.ieee.org/document/8819955/AEC-NetCobb angle estimationdeep learningdirect estimationhigh-precision calculation |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Bo Chen Qiuhao Xu Liansheng Wang Stephanie Leung Jonathan Chung Shuo Li |
spellingShingle |
Bo Chen Qiuhao Xu Liansheng Wang Stephanie Leung Jonathan Chung Shuo Li An Automated and Accurate Spine Curve Analysis System IEEE Access AEC-Net Cobb angle estimation deep learning direct estimation high-precision calculation |
author_facet |
Bo Chen Qiuhao Xu Liansheng Wang Stephanie Leung Jonathan Chung Shuo Li |
author_sort |
Bo Chen |
title |
An Automated and Accurate Spine Curve Analysis System |
title_short |
An Automated and Accurate Spine Curve Analysis System |
title_full |
An Automated and Accurate Spine Curve Analysis System |
title_fullStr |
An Automated and Accurate Spine Curve Analysis System |
title_full_unstemmed |
An Automated and Accurate Spine Curve Analysis System |
title_sort |
automated and accurate spine curve analysis system |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2019-01-01 |
description |
We present a new Adaptive Error Correction Net (AEC-Net) to formulate the estimation of Cobb anges from spinal X-rays as a high-precision regression task. Our AEC-Net introduces two novel innovations. (1) The AEC-Net contains two networks calculating landmarks and Cobb angles separately, which robustly solve the disadvantage of ambiguity in X-rays since these networks focus on more features. It effectively handles the nonlinear relationship between input images and quantitative outputs, while explicitly capturing the intrinsic features of input images. (2) Based on the two estimated angles, the AEC-Net proposed a new loss function to calculate the final Cobb angles. The optimization of the loss function is based on a high-precision calculation method. The deep learning structure is used to complete this optimization, which achieves higher accuracy and efficiency. We validate our method with the spinal X-rays dataset of 581 subjects with signs of scoliosis at varying extents. The proposed method achieves high accuracy and robustness on the Cobb angle estimations. Comparing to the exsiting conventional methods suffering from tremendous variability and low reliability caused by high ambiguity and variability around boundaries of the vertebrae, the AEC-Net obtain Cobb angles accurately and robustly, which indicates its great potential in clinical use. The highly accurate Cobb angles produced by our framework can be used by clinicians for comprehensive scoliosis assessment, and possibly be further extended to other clinical applications. |
topic |
AEC-Net Cobb angle estimation deep learning direct estimation high-precision calculation |
url |
https://ieeexplore.ieee.org/document/8819955/ |
work_keys_str_mv |
AT bochen anautomatedandaccuratespinecurveanalysissystem AT qiuhaoxu anautomatedandaccuratespinecurveanalysissystem AT lianshengwang anautomatedandaccuratespinecurveanalysissystem AT stephanieleung anautomatedandaccuratespinecurveanalysissystem AT jonathanchung anautomatedandaccuratespinecurveanalysissystem AT shuoli anautomatedandaccuratespinecurveanalysissystem AT bochen automatedandaccuratespinecurveanalysissystem AT qiuhaoxu automatedandaccuratespinecurveanalysissystem AT lianshengwang automatedandaccuratespinecurveanalysissystem AT stephanieleung automatedandaccuratespinecurveanalysissystem AT jonathanchung automatedandaccuratespinecurveanalysissystem AT shuoli automatedandaccuratespinecurveanalysissystem |
_version_ |
1724189780710785024 |