Automatic Spine Segmentation and Parameter Measurement for Radiological Analysis of Whole-Spine Lateral Radiographs Using Deep Learning and Computer Vision

被引:8
作者
Kim, Yong-Tae [1 ]
Jeong, Tae Seok [2 ]
Kim, Young Jae [1 ]
Kim, Woo Seok [3 ]
Kim, Kwang Gi [1 ]
Yee, Gi Taek [2 ]
机构
[1] Gachon Univ, Coll Med, Gil Med Ctr, Dept Biomed Engn, Incheon, South Korea
[2] Gachon Univ, Gil Med Ctr, Dept Neurosurg, Coll Med, Incheon, South Korea
[3] Gachon Univ, Coll Med, Gil Med Ctr, Dept Traumatol, Incheon, South Korea
关键词
Mask R-CNN; Spine; Computer vision; Computer-assisted diagnosis; SPINOPELVIC PARAMETERS; SAGITTAL BALANCE; DEFORMITY; INTRAOBSERVER; SCOLIOSIS; KEY; INTEROBSERVER; RELIABILITY; VALIDATION; ALIGNMENT;
D O I
10.1007/s10278-023-00830-z
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Radiographic examination is essential for diagnosing spinal disorders, and the measurement of spino-pelvic parameters provides important information for the diagnosis and treatment planning of spinal sagittal deformities. While manual measurement methods are the golden standard for measuring parameters, they can be time consuming, inefficient, and rater dependent. Previous studies that have used automatic measurement methods to alleviate the downsides of manual measurements showed low accuracy or could not be applied to general films. We propose a pipeline for automated measurement of spinal parameters by combining a Mask R-CNN model for spine segmentation with computer vision algorithms. This pipeline can be incorporated into clinical workflows to provide clinical utility in diagnosis and treatment planning. A total of 1807 lateral radiographs were used for the training (n = 1607) and validation (n = 200) of the spine segmentation model. An additional 200 radiographs, which were also used for validation, were examined by three surgeons to evaluate the performance of the pipeline. Parameters automatically measured by the algorithm in the test set were statistically compared to parameters measured manually by the three surgeons. The Mask R-CNN model achieved an average precision at 50% intersection over union (AP50) of 96.2% and a Dice score of 92.6% for the spine segmentation task in the test set. The mean absolute error values of the spino-pelvic parameters measurement results were within the range of 0.4 degrees (pelvic tilt) to 3.0 degrees (lumbar lordosis, pelvic incidence), and the standard error of estimate was within the range of 0.5 degrees (pelvic tilt) to 4.0 degrees (pelvic incidence). The intraclass correlation coefficient values ranged from 0.86 (sacral slope) to 0.99 (pelvic tilt, sagittal vertical axis).
引用
收藏
页码:1447 / 1459
页数:13
相关论文
共 36 条
  • [1] MEASUREMENT IN MEDICINE - THE ANALYSIS OF METHOD COMPARISON STUDIES
    ALTMAN, DG
    BLAND, JM
    [J]. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES D-THE STATISTICIAN, 1983, 32 (03) : 307 - 317
  • [2] MEASUREMENT OF SCOLIOSIS AND KYPHOSIS RADIOGRAPHS - INTRAOBSERVER AND INTEROBSERVER VARIATION
    CARMAN, DL
    BROWNE, RH
    BIRCH, JG
    [J]. JOURNAL OF BONE AND JOINT SURGERY-AMERICAN VOLUME, 1990, 72A (03) : 328 - 333
  • [3] Chen K, 2019, ARXIV, DOI [DOI 10.48550/ARXIV.1906.07155, 10.48550/arXiv.1906.07155]
  • [4] Vertebral centroid measurement of lumbar lordosis compared with the Cobb technique
    Chen, YL
    [J]. SPINE, 1999, 24 (17) : 1786 - 1790
  • [5] Automated Measurement of Lumbar Lordosis on Radiographs Using Machine Learning and Computer Vision
    Cho, Brian H.
    Kaji, Deepak
    Cheung, Zoe B.
    Ye, Ivan B.
    Tang, Ray
    Ahn, Amy
    Carrillo, Oscar
    Schwartz, John T.
    Valliani, Aly A.
    Oermann, Eric K.
    Arvind, Varun
    Ranti, Daniel
    Sun, Li
    Kim, Jun S.
    Cho, Samuel K.
    [J]. GLOBAL SPINE JOURNAL, 2020, 10 (05) : 611 - 618
  • [6] Fully automated radiological analysis of spinal disorders and deformities: a deep learning approach
    Galbusera, Fabio
    Niemeyer, Frank
    Wilke, Hans-Joachim
    Bassani, Tito
    Casaroli, Gloria
    Anania, Carla
    Costa, Francesco
    Brayda-Bruno, Marco
    Sconfienza, Luca Maria
    [J]. EUROPEAN SPINE JOURNAL, 2019, 28 (05) : 951 - 960
  • [7] The impact of positive sagittal balance in adult spinal deformity
    Glassman, SD
    Bridwell, K
    Dimar, JR
    Horton, W
    Berven, S
    Schwab, F
    [J]. SPINE, 2005, 30 (18) : 2024 - 2029
  • [8] Inter- and intraobserver reliability assessment of the Cobb angle: manual versus digital measurement tools
    Gstoettner, Michaela
    Sekyra, Katrin
    Walochnik, Nadja
    Winter, Peter
    Wachter, Roland
    Bach, Christian M.
    [J]. EUROPEAN SPINE JOURNAL, 2007, 16 (10) : 1587 - 1592
  • [9] Harrison D E, 2001, Spine (Phila Pa 1976), V26, pE227, DOI 10.1097/00007632-200106010-00002
  • [10] Cobb method or Harrison posterior tangent method
    Harrison, DE
    Harrison, DD
    Cailliet, R
    Troyanovich, SJ
    Janik, TJ
    Holland, B
    [J]. SPINE, 2000, 25 (16) : 2072 - 2078