Diffusion-weighted MRI radiomics of spine bone tumors: feature stability and machine learning-based classification performance.
Machine learning
Radiomics
Reproducibility
Spine
Texture analysis
Tumor
Journal
La Radiologia medica
ISSN: 1826-6983
Titre abrégé: Radiol Med
Pays: Italy
ID NLM: 0177625
Informations de publication
Date de publication:
May 2022
May 2022
Historique:
received:
27
02
2021
accepted:
11
02
2022
pubmed:
24
3
2022
medline:
18
5
2022
entrez:
23
3
2022
Statut:
ppublish
Résumé
To evaluate stability and machine learning-based classification performance of radiomic features of spine bone tumors using diffusion- and T2-weighted magnetic resonance imaging (MRI). This retrospective study included 101 patients with histology-proven spine bone tumor (22 benign; 38 primary malignant; 41 metastatic). All tumor volumes were manually segmented on morphologic T2-weighted sequences. The same region of interest (ROI) was used to perform radiomic analysis on ADC map. A total of 1702 radiomic features was considered. Feature stability was assessed through small geometrical transformations of the ROIs mimicking multiple manual delineations. Intraclass correlation coefficient (ICC) quantified feature stability. Feature selection consisted of stability-based (ICC > 0.75) and significance-based selections (ranking features by decreasing Mann-Whitney p-value). Class balancing was performed to oversample the minority (i.e., benign) class. Selected features were used to train and test a support vector machine (SVM) to discriminate benign from malignant spine tumors using tenfold cross-validation. A total of 76.4% radiomic features were stable. The quality metrics for the SVM were evaluated as a function of the number of selected features. The radiomic model with the best performance and the lowest number of features for classifying tumor types included 8 features. The metrics were 78% sensitivity, 68% specificity, 76% accuracy and AUC 0.78. SVM classifiers based on radiomic features extracted from T2- and diffusion-weighted imaging with ADC map are promising for classification of spine bone tumors. Radiomic features of spine bone tumors show good reproducibility rates.
Identifiants
pubmed: 35320464
doi: 10.1007/s11547-022-01468-7
pii: 10.1007/s11547-022-01468-7
pmc: PMC9098537
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
518-525Informations de copyright
© 2022. The Author(s).
Références
Albano D, Messina C, Gitto S et al (2019) Differential diagnosis of spine tumors: my favorite mistake. Semin Musculoskelet Radiol 23:026–035. https://doi.org/10.1055/s-0038-1675551
doi: 10.1055/s-0038-1675551
Cannavò L, Albano D, Messina C et al (2019) Accuracy of CT and MRI to assess resection margins in primary malignant bone tumours having histology as the reference standard. Clin Radiol 74:736.e13-736.e21. https://doi.org/10.1016/j.crad.2019.05.022
doi: 10.1016/j.crad.2019.05.022
Jung H-S, Jee W-H, McCauley TR et al (2003) Discrimination of metastatic from acute osteoporotic compression spinal fractures with MR imaging 1. Radiographics 23:179–187. https://doi.org/10.1148/rg.231025043
doi: 10.1148/rg.231025043
pubmed: 12533652
Suh CH, Yun SJ, Jin W et al (2018) ADC as a useful diagnostic tool for differentiating benign and malignant vertebral bone marrow lesions and compression fractures: a systematic review and meta-analysis. Eur Radiol 28:2890–2902. https://doi.org/10.1007/s00330-018-5330-5
doi: 10.1007/s00330-018-5330-5
pubmed: 29450718
Pozzi G, Albano D, Messina C et al (2018) Solid bone tumors of the spine: Diagnostic performance of apparent diffusion coefficient measured using diffusion-weighted MRI using histology as a reference standard. J Magn Reson Imaging 47:1034–1042. https://doi.org/10.1002/jmri.25826
doi: 10.1002/jmri.25826
pubmed: 28755383
Gillies RJ, Kinahan PE, Hricak H (2016) Radiomics: images are more than pictures, they are data. Radiology 278:563–577. https://doi.org/10.1148/radiol.2015151169
doi: 10.1148/radiol.2015151169
pubmed: 26579733
Gitto S, Cuocolo R, Albano D et al (2021) CT and MRI radiomics of bone and soft-tissue sarcomas: a systematic review of reproducibility and validation strategies. Insights Imaging 12:68. https://doi.org/10.1186/s13244-021-01008-3
doi: 10.1186/s13244-021-01008-3
pubmed: 34076740
pmcid: 8172744
Lubner MG, Smith AD, Sandrasegaran K et al (2017) CT texture analysis: definitions, applications, biologic correlates, and challenges. Radiographics 37:1483–1503. https://doi.org/10.1148/rg.2017170056
doi: 10.1148/rg.2017170056
pubmed: 28898189
Gitto S, Cuocolo R, Emili I et al (2021) Effects of interobserver variability on 2D and 3D CT- and MRI-based texture feature reproducibility of cartilaginous bone tumors. J Digit Imaging. https://doi.org/10.1007/s10278-021-00498-3
doi: 10.1007/s10278-021-00498-3
pubmed: 34405298
pmcid: 8455795
Gitto S, Cuocolo R, Albano D et al (2020) MRI radiomics-based machine-learning classification of bone chondrosarcoma. Eur J Radiol 128:109043. https://doi.org/10.1016/j.ejrad.2020.109043
doi: 10.1016/j.ejrad.2020.109043
pubmed: 32438261
Gitto S, Cuocolo R, Annovazzi A et al (2021) CT radiomics-based machine learning classification of atypical cartilaginous tumours and appendicular chondrosarcomas. EBioMedicine 68:103407. https://doi.org/10.1016/j.ebiom.2021.103407
doi: 10.1016/j.ebiom.2021.103407
pubmed: 34051442
pmcid: 8170113
Yushkevich PA, Piven J, Hazlett HC et al (2006) User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 31:1116–1128. https://doi.org/10.1016/j.neuroimage.2006.01.015
doi: 10.1016/j.neuroimage.2006.01.015
pubmed: 16545965
Tustison NJ, Cook PA, Gee JC (2010) N4ITK: improved N3 bias correction. IEEE Trans Med Imaging 29:1310–1320. https://doi.org/10.1109/TMI.2010.2046908.N4ITK
doi: 10.1109/TMI.2010.2046908.N4ITK
pubmed: 20378467
pmcid: 3071855
Leijenaar RT, Bogowicz M, Jochems A et al (2018) Development and validation of a radiomic signature to predict HPV (p16) status from standard CT imaging: a multicenter study. Br J Radiol 91:20170498. https://doi.org/10.1259/bjr.20170498
doi: 10.1259/bjr.20170498
pubmed: 29451412
pmcid: 6223271
van Griethuysen JJM, Fedorov A, Parmar C et al (2017) Computational radiomics system to decode the radiographic phenotype. Cancer Res 77:e104–e107. https://doi.org/10.1158/0008-5472.CAN-17-0339
doi: 10.1158/0008-5472.CAN-17-0339
pubmed: 29092951
pmcid: 5672828
Aerts HJWL, Velazquez ER, Leijenaar RTH et al (2014) Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat Commun 5:4006
doi: 10.1038/ncomms5006
Bologna M, Corino VDA, Montin E et al (2018) Assessment of stability and discrimination capacity of radiomic features on apparent diffusion coefficient images. J Digit Imaging 31:879–894. https://doi.org/10.1007/s10278-018-0092-9
doi: 10.1007/s10278-018-0092-9
pubmed: 29725965
pmcid: 6261192
Mcgraw KO (1996) Forming inferences about some intraclass correlation coefficients. Psychol Methods 1:30–46. https://doi.org/10.1037/1082-989X.1.1.30
doi: 10.1037/1082-989X.1.1.30
Koo TK, Li MY (2016) A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med 15:155–163. https://doi.org/10.1016/j.jcm.2016.02.012
doi: 10.1016/j.jcm.2016.02.012
pubmed: 27330520
pmcid: 4913118
Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) SMOTE: synthetic minority over-sampling technique. J Artif Intell Res 16:321–357. https://doi.org/10.1613/jair.953
doi: 10.1613/jair.953
Filograna L, Lenkowicz J, Cellini F et al (2019) Identification of the most significant magnetic resonance imaging (MRI) radiomic features in oncological patients with vertebral bone marrow metastatic disease: a feasibility study. Radiol Med 124:50–57. https://doi.org/10.1007/s11547-018-0935-y
doi: 10.1007/s11547-018-0935-y
pubmed: 30191445
Chianca V, Cuocolo R, Gitto S et al (2021) Radiomic machine learning classifiers in spine bone tumors: a multi-software. Multi-Scanner Study Eur J Radiol 137:109586. https://doi.org/10.1016/j.ejrad.2021.109586
doi: 10.1016/j.ejrad.2021.109586
pubmed: 33610852
Lang N, Zhang Y, Zhang E et al (2019) Differentiation of spinal metastases originated from lung and other cancers using radiomics and deep learning based on DCE-MRI. Magn Reson Imaging 64:4–12. https://doi.org/10.1016/j.mri.2019.02.013
doi: 10.1016/j.mri.2019.02.013
pubmed: 30826448
pmcid: 6713616
Galavis PE, Hollensen C, Jallow N et al (2010) Variability of textural features in FDG PET images due to different acquisition modes and reconstruction parameters. Acta Oncol (Madr) 49:1012–1016. https://doi.org/10.3109/0284186X.2010.498437
doi: 10.3109/0284186X.2010.498437
Gitto S, Cuocolo R, van Langevelde K et al (2022) MRI radiomics-based machine learning classification of a typical cartilaginous tumour and grade II chondrosarcoma of long bones. EBioMedicine 75:103757. https://doi.org/10.1016/j.ebiom.2021.103757
doi: 10.1016/j.ebiom.2021.103757
pubmed: 34933178