Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study - PubMed (original) (raw)
Comparative Study
Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study
Oskar Maier et al. PLoS One. 2015.
Erratum in
- Correction: Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study.
Maier O, Schröder C, Forkert ND, Martinetz T, Handels H. Maier O, et al. PLoS One. 2016 Feb 12;11(2):e0149828. doi: 10.1371/journal.pone.0149828. eCollection 2016. PLoS One. 2016. PMID: 26871444 Free PMC article. No abstract available.
Abstract
Motivation: Ischemic stroke, triggered by an obstruction in the cerebral blood supply, leads to infarction of the affected brain tissue. An accurate and reproducible automatic segmentation is of high interest, since the lesion volume is an important end-point for clinical trials. However, various factors, such as the high variance in lesion shape, location and appearance, render it a difficult task.
Methods: In this article, nine classification methods (e.g. Generalized Linear Models, Random Decision Forests and Convolutional Neural Networks) are evaluated and compared with each other using 37 multiparametric MRI datasets of ischemic stroke patients in the sub-acute phase in terms of their accuracy and reliability for ischemic stroke lesion segmentation. Within this context, a multi-spectral classification approach is compared against mono-spectral classification performance using only FLAIR MRI datasets and two sets of expert segmentations are used for inter-observer agreement evaluation.
Results and conclusion: The results of this study reveal that high-level machine learning methods lead to significantly better segmentation results compared to the rather simple classification methods, pointing towards a difficult non-linear problem. The overall best segmentation results were achieved by a Random Decision Forest and a Convolutional Neural Networks classification approach, even outperforming all previously published results. However, none of the methods tested in this work are capable of achieving results in the range of the human observer agreement and the automatic ischemic stroke lesion segmentation remains a complicated problem that needs to be explored in more detail to improve the segmentation results.
Conflict of interest statement
Competing Interests: The authors have declared that no competing interests exist.
Figures
Fig 1. Processing pipeline.
Fig 2. Results for case 21.
Slice 21 with besteffort scenario trained on GTG. (a) ground-truth (GTG). (b) 100NN. (c) 10NN. (d) 5NN. (e) AdaBoost. (f) ET. (g) GNB. (h) GLM. (i) GB. (j) RDF.
Fig 3. Results for case 04.
Slice 30 with besteffort scenario trained on GTG. Note the presence of other white matter hyperintensities. (a) ground-truth (GTG). (b) 100NN. (c) 5NN. (d) AdaBoost. (e) ET. (f) GNB. (g) GLM. (h) GB. (i) RDF. (j) tunedET.
Fig 4. ROC curves for both evaluation scenarios computed over the GTG ground truth.
(a) besteffort scenario. (b) flair scenario.
Fig 5. Cases failed by at least one classifier.
(a) 09/24 ground-truth (GTG). (b) 11/29 ground-truth (GTG). (c) 39/36 ground-truth (GTG). (d) 41/24 ground-truth (GTG).
Fig 6. Worst two cases.
See text for description. (a) 37/44 ground-truth (GTG). (b) 37/32 tunedET. (c) 44/24 ground-truth (GTG). (d) 44/24 CNN.
Fig 7. Best overall case 36 and the worst (GNB, DM = 0.61) as well as best (ET, DM = 0.86) result obtained over all methods.
(a) 36/27 ground-truth (GTG). (b) 36/27 GNB. (c) 36/27 ET.
Fig 8. Case with low agreement between methods in flair scenario.
(a) 18/28 ground-truth (GTG). (b) 18/28 AdaBoost. (c) 18/28 CNN. (d) 18/28 ET.
Similar articles
- Extra tree forests for sub-acute ischemic stroke lesion segmentation in MR sequences.
Maier O, Wilms M, von der Gablentz J, Krämer UM, Münte TF, Handels H. Maier O, et al. J Neurosci Methods. 2015 Jan 30;240:89-100. doi: 10.1016/j.jneumeth.2014.11.011. Epub 2014 Nov 21. J Neurosci Methods. 2015. PMID: 25448384 - Machine learning identifies stroke features between species.
Castaneda-Vega S, Katiyar P, Russo F, Patzwaldt K, Schnabel L, Mathes S, Hempel JM, Kohlhofer U, Gonzalez-Menendez I, Quintanilla-Martinez L, Ziemann U, la Fougere C, Ernemann U, Pichler BJ, Disselhorst JA, Poli S. Castaneda-Vega S, et al. Theranostics. 2021 Jan 1;11(6):3017-3034. doi: 10.7150/thno.51887. eCollection 2021. Theranostics. 2021. PMID: 33456586 Free PMC article. - FLAIR lesion segmentation: application in patients with brain tumors and acute ischemic stroke.
Artzi M, Aizenstein O, Jonas-Kimchi T, Myers V, Hallevi H, Ben Bashat D. Artzi M, et al. Eur J Radiol. 2013 Sep;82(9):1512-8. doi: 10.1016/j.ejrad.2013.05.029. Epub 2013 Jun 21. Eur J Radiol. 2013. PMID: 23796882 - Lesion segmentation from multimodal MRI using random forest following ischemic stroke.
Mitra J, Bourgeat P, Fripp J, Ghose S, Rose S, Salvado O, Connelly A, Campbell B, Palmer S, Sharma G, Christensen S, Carey L. Mitra J, et al. Neuroimage. 2014 Sep;98:324-35. doi: 10.1016/j.neuroimage.2014.04.056. Epub 2014 May 2. Neuroimage. 2014. PMID: 24793830 - An East Coast Perspective on Artificial Intelligence and Machine Learning: Part 2: Ischemic Stroke Imaging and Triage.
Gupta R, Krishnam SP, Schaefer PW, Lev MH, Gonzalez RG. Gupta R, et al. Neuroimaging Clin N Am. 2020 Nov;30(4):467-478. doi: 10.1016/j.nic.2020.08.002. Neuroimaging Clin N Am. 2020. PMID: 33038997 Review.
Cited by
- Taxonomy of Acute Stroke: Imaging, Processing, and Treatment.
Nowinski WL. Nowinski WL. Diagnostics (Basel). 2024 May 19;14(10):1057. doi: 10.3390/diagnostics14101057. Diagnostics (Basel). 2024. PMID: 38786355 Free PMC article. Review. - Stroke-GFCN: ischemic stroke lesion prediction with a fully convolutional graph network.
Iporre-Rivas A, Saur D, Rohr K, Scheuermann G, Gillmann C. Iporre-Rivas A, et al. J Med Imaging (Bellingham). 2023 Jul;10(4):044502. doi: 10.1117/1.JMI.10.4.044502. Epub 2023 Jul 17. J Med Imaging (Bellingham). 2023. PMID: 37465592 Free PMC article. - Automatic Kidney Segmentation Method Based on an Enhanced Generative Adversarial Network.
Shan T, Ying Y, Song G. Shan T, et al. Diagnostics (Basel). 2023 Apr 6;13(7):1358. doi: 10.3390/diagnostics13071358. Diagnostics (Basel). 2023. PMID: 37046576 Free PMC article. - Differential Impact of Brain Network Efficiency on Poststroke Motor and Attentional Deficits.
Evangelista GG, Egger P, Brügger J, Beanato E, Koch PJ, Ceroni M, Fleury L, Cadic-Melchior A, Meyer NH, Rodríguez DL, Girard G, Léger B, Turlan JL, Mühl A, Vuadens P, Adolphsen J, Jagella CE, Constantin C, Alvarez V, San Millán D, Bonvin C, Morishita T, Wessel MJ, Van De Ville D, Hummel FC. Evangelista GG, et al. Stroke. 2023 Apr;54(4):955-963. doi: 10.1161/STROKEAHA.122.040001. Epub 2023 Feb 27. Stroke. 2023. PMID: 36846963 Free PMC article. - Hospital-acquired infections after acute ischaemic stroke and its association with healthcare-related costs and functional outcome.
Grieten J, Chevalier P, Lesenne A, Ernon L, Vandermeulen E, Panis E, Mesotten D. Grieten J, et al. Acta Neurol Belg. 2022 Oct;122(5):1281-1287. doi: 10.1007/s13760-022-01977-2. Epub 2022 Jun 30. Acta Neurol Belg. 2022. PMID: 35773572
References
- Rekik I, Allassonniere S, Carpenter TK, Wardlaw JM. Medical image analysis methods in MR/CT-imaged acute-subacute ischemic stroke lesion: Segmentation, prediction and insights into dynamic evolution simulation models. A critical appraisal. NeuroImage: Clinical. 2012;1(1):164–178. 10.1016/j.nicl.2012.10.003 - DOI - PMC - PubMed
- Forbes F, Doyle S, Garcia-Lorenzo D, Barillot C, Dojat M. Adaptive weighted fusion of multiple MR sequences for brain lesion segmentation In: Biomedical Imaging: From Nano to Macro, 2010 IEEE International Symposium on; 2010. p. 69–72.
Publication types
MeSH terms
Grants and funding
OM holds a scholarship of the Graduate School for Computing in Medicine and Life Sciences, Universität zu Lübeck, Lübeck, Germany. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
LinkOut - more resources
Full Text Sources
Other Literature Sources
Medical