Segmentation of anatomical structures in chest radiographs using supervised methods: a comparative study on a public database
- PMID: 15919232
- DOI: 10.1016/j.media.2005.02.002
Segmentation of anatomical structures in chest radiographs using supervised methods: a comparative study on a public database
Abstract
The task of segmenting the lung fields, the heart, and the clavicles in standard posterior-anterior chest radiographs is considered. Three supervised segmentation methods are compared: active shape models, active appearance models and a multi-resolution pixel classification method that employs a multi-scale filter bank of Gaussian derivatives and a k-nearest-neighbors classifier. The methods have been tested on a publicly available database of 247 chest radiographs, in which all objects have been manually segmented by two human observers. A parameter optimization for active shape models is presented, and it is shown that this optimization improves performance significantly. It is demonstrated that the standard active appearance model scheme performs poorly, but large improvements can be obtained by including areas outside the objects into the model. For lung field segmentation, all methods perform well, with pixel classification giving the best results: a paired t-test showed no significant performance difference between pixel classification and an independent human observer. For heart segmentation, all methods perform comparably, but significantly worse than a human observer. Clavicle segmentation is a hard problem for all methods; best results are obtained with active shape models, but human performance is substantially better. In addition, several hybrid systems are investigated. For heart segmentation, where the separate systems perform comparably, significantly better performance can be obtained by combining the results with majority voting. As an application, the cardio-thoracic ratio is computed automatically from the segmentation results. Bland and Altman plots indicate that all methods perform well when compared to the gold standard, with confidence intervals from pixel classification and active appearance modeling very close to those of a human observer. All results, including the manual segmentations, have been made publicly available to facilitate future comparative studies.
Similar articles
-
Cavity contour segmentation in chest radiographs using supervised learning and dynamic programming.Med Phys. 2014 Jul;41(7):071912. doi: 10.1118/1.4881096. Med Phys. 2014. PMID: 24989390
-
Clavicle segmentation in chest radiographs.Med Image Anal. 2012 Dec;16(8):1490-502. doi: 10.1016/j.media.2012.06.009. Epub 2012 Jul 31. Med Image Anal. 2012. PMID: 22998970
-
Minimal shape and intensity cost path segmentation.IEEE Trans Med Imaging. 2007 Aug;26(8):1115-29. doi: 10.1109/TMI.2007.896924. IEEE Trans Med Imaging. 2007. PMID: 17695131
-
Segmentation of the posterior ribs in chest radiographs using iterated contextual pixel classification.IEEE Trans Med Imaging. 2006 May;25(5):602-11. doi: 10.1109/TMI.2006.872747. IEEE Trans Med Imaging. 2006. PMID: 16689264
-
Automatic segmentation of lung fields in chest radiographs.Med Phys. 2000 Oct;27(10):2445-55. doi: 10.1118/1.1312192. Med Phys. 2000. PMID: 11099215
Cited by
-
Towards Foundation Models Learned from Anatomy in Medical Imaging via Self-supervision.Domain Adapt Represent Transf (2023). 2024;14293:94-104. doi: 10.1007/978-3-031-45857-6_10. Epub 2023 Oct 14. Domain Adapt Represent Transf (2023). 2024. PMID: 38752223 Free PMC article.
-
Self-supervised learning for medical image analysis: Discriminative, restorative, or adversarial?Med Image Anal. 2024 May;94:103086. doi: 10.1016/j.media.2024.103086. Epub 2024 Jan 28. Med Image Anal. 2024. PMID: 38537414
-
Segmentation-based cardiomegaly detection based on semi-supervised estimation of cardiothoracic ratio.Sci Rep. 2024 Mar 8;14(1):5695. doi: 10.1038/s41598-024-56079-1. Sci Rep. 2024. PMID: 38459104 Free PMC article.
-
[Non-local attention and multi-task learning based lung segmentation in chest X-ray].Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2023 Oct 25;40(5):912-919. doi: 10.7507/1001-5515.202211079. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2023. PMID: 37879920 Free PMC article. Chinese.
-
Evaluating artificial intelligence for comparative radiography.Int J Legal Med. 2024 Jan;138(1):307-327. doi: 10.1007/s00414-023-03080-4. Epub 2023 Oct 6. Int J Legal Med. 2024. PMID: 37801115
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Other Literature Sources
Research Materials