Multi-dimensional interpretable deep learning-radiomics based on intra-tumoral and spatial habitat for preoperative prediction of thymic epithelial tumours risk categorisation
DOI:
https://doi.org/10.2340/1651-226X.2025.42982Keywords:
Radiomics, Deep learning, Machine learning, ThymomaAbstract
Background and purpose: This study aims to develop and compare combined models based on enhanced CT-based radiomics, multi-dimensional deep learning, clinical-conventional imaging and spatial habitat analysis to achieve accurate prediction of thymoma risk classification.
Materials and Methods: 205 consecutive patients with thymoma confirmed by surgical pathology were recruited from three medical centers. Venous phase enhanced CT images were used to delineate the tumor, and radiomics, 2D and 3D deep learning models based on the whole tumor were established and feature extraction was performed. The tumors were divided into different sub-regions by K-means clustering method and the corresponding features were obtained. The clinical-conventional imaging data of the patients were collected and evaluated, and the univariate and multivariate analysis were used for screening. The above types of features were fused with each other to construct a variety of combined models. Quantitative indicators such as area under the receiver operating characteristic (ROC) curve (AUC) were calculated to evaluate the performance of the model.
Results: The AUC of RDLCSM developed based on LightGBM classifier was 0.953 in the training cohort, 0.930 in the internal validation cohort, 0.924 and 0.903 in the two external validation cohorts, respectively. RDLCSM performs better than RDLM (AUC range: 0.831-0.890) and 2DLCSM (AUC range: 0.785-0.916) based on KNN. In addition, RDLCSM had the highest accuracy (0.818-0.882) and specificity (0.926-1.000).
Interpretation: The RDLCSM, which combines whole-tumor radiomics, 2D and 3D deep learning, clinical-visual radiology, and subregional omics, can be used as a non-invasive tool to predict thymoma risk classification.
Downloads
References
Araki T, Nishino M, Gao W, Dupuis J, Washko GR, Hunninghake GM, et al. Anterior mediastinal masses in the Framingham heart study: prevalence and CT image characteristics. Eur J Radiol Open. 2015;2:26–31.
https://doi.org/10.1016/j.ejro.2014.12.003 DOI: https://doi.org/10.1016/j.ejro.2014.12.003
Scorsetti M, Leo F, Trama A, D’Angelillo R, Serpico D, Macerelli M, et al. Thymoma and thymic carcinomas. Crit Rev Oncol Hematol. 2016;99:332–50.
https://doi.org/10.1016/j.critrevonc.2016.01.012 DOI: https://doi.org/10.1016/j.critrevonc.2016.01.012
Hishida T, Asamura H, Yoshida K, Tsuboi M, Yokoi K, Toyooka S, et al. Clinical features and prognostic impact of coexisting autoimmune disease other than myasthenia gravis in resected thymomas: analysis of a Japanese multi-institutional retrospective database. Eur J Cardiothorac Surg. 2021;59(3):641–9.
https://doi.org/10.1093/ejcts/ezaa362 DOI: https://doi.org/10.1093/ejcts/ezaa362
Aizemaiti R, Wang Y. Advances in staging system of thymoma. Zhongguo Fei Ai Za Zhi. 2011;14(2):170–2. Chinese.
https://doi.org/10.3779/j.issn.1009-3419.2011.02.13
Meurgey A, Girard N, Merveilleux du Vignaux C, Maury JM, Tronc F, Thivolet-Bejui F, et al. Assessment of the ITMIG statement on the WHO histological classification and of the eighth TNM staging of thymic epithelial tumors of a series of 188 thymic epithelial tumors. J Thorac Oncol. 2017;12(10):1571–81.
https://doi.org/10.1016/j.jtho.2017.06.072 DOI: https://doi.org/10.1016/j.jtho.2017.06.072
Suster S, Moran CA. Histologic classification of thymoma: the World Health Organization and beyond. Hematol Oncol Clin North Am. 2008;22(3):381–92.
https://doi.org/10.1016/j.hoc.2008.03.001 DOI: https://doi.org/10.1016/j.hoc.2008.03.001
Sorejs O, Pesek M, Finek J, Fiala O. Octreotide in the treatment of malignant thymoma – case report. Rep Pract Oncol Radiother. 2020;25(6):882–5.
https://doi.org/10.1016/j.rpor.2020.08.012 DOI: https://doi.org/10.1016/j.rpor.2020.08.012
Detterbeck FC, Stratton K, Giroux D, Asamura H, Crowley J, Falkson C, et al. The IASLC/ITMIG thymic epithelial tumors staging project: proposal for an evidence-based stage classification system for the forthcoming (8th) edition of the TNM classification of malignant tumors. J Thorac Oncol. 2014;9(9 Suppl. 2):S65–72.
https://doi.org/10.1097/JTO.0000000000000290 DOI: https://doi.org/10.1097/JTO.0000000000000290
Chen G, Marx A, Chen WH, Yong J, Puppe B, Stroebel P, et al. New WHO histologic classification predicts prognosis of thymic epithelial tumors: a clinicopathologic study of 200 thymoma cases from China. Cancer. 2002;95(2):420–9.
https://doi.org/10.1002/cncr.10665 DOI: https://doi.org/10.1002/cncr.10665
Ettinger DS, Riely GJ, Akerley W, Borghaei H, Chang AC, Cheney RT, et al. Thymomas and thymic carcinomas: clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2013;11(5):562–76.
https://doi.org/10.6004/jnccn.2013.0072 DOI: https://doi.org/10.6004/jnccn.2013.0072
Yanagawa M, Tomiyama N. Prediction of thymoma histology and stage by radiographic criteria. Thorac Surg Clin. 2011;21(1):1–12.
https://doi.org/10.1016/j.thorsurg.2010.08.008 DOI: https://doi.org/10.1016/j.thorsurg.2010.08.008
Kumar V, Gu Y, Basu S, Berglund A, Eschrich SA, Schabath MB, et al. Radiomics: the process and the challenges. Magn Reson Imaging. 2012;30(9):1234–48.
https://doi.org/10.1016/j.mri.2012.06.010 DOI: https://doi.org/10.1016/j.mri.2012.06.010
Truhn D, Schrading S, Haarburger C, Schneider H, Merhof D, Kuhl C. Radiomic versus convolutional neural networks analysis for classification of contrast-enhancing lesions at multiparametric breast MRI. Radiology. 2019;290(2):290–7.
https://doi.org/10.1148/radiol.2018181352 DOI: https://doi.org/10.1148/radiol.2018181352
Dai M, Liu Y, Hu Y, Li G, Zhang J, Xiao Z, et al. Combining multiparametric MRI features-based transfer learning and clinical parameters: application of machine learning for the differentiation of uterine sarcomas from atypical leiomyomas. Eur Radiol. 2022;32(11):7988–97.
https://doi.org/10.1007/s00330-022-08783-7 DOI: https://doi.org/10.1007/s00330-022-08783-7
Pease M, Arefan D, Barber J, Yuh E, Puccio A, Hochberger K, et al. Outcome prediction in patients with severe traumatic brain injury using deep learning from head CT scans. Radiology. 2022;304(2):385–94.
https://doi.org/10.1148/radiol.212181 DOI: https://doi.org/10.1148/radiol.212181
Cho HH, Kim H, Nam SY, Lee JE, Han BK, Ko EY, et al. Measurement of perfusion heterogeneity within tumor habitats on magnetic resonance imaging and its association with prognosis in breast cancer patients. Cancers (Basel). 2022;14(8):1858.
https://doi.org/10.3390/cancers14081858 DOI: https://doi.org/10.3390/cancers14081858
Feng XL, Wang SZ, Chen HH, Huang YX, Xin YK, Zhang T, et al. Optimizing the radiomics-machine-learning model based on non-contrast enhanced CT for the simplified risk categorization of thymic epithelial tumors: a large cohort retrospective study. Lung Cancer. 2022;166:150–60.
https://doi.org/10.1016/j.lungcan.2022.03.007 DOI: https://doi.org/10.1016/j.lungcan.2022.03.007
Liu W, Wang W, Zhang H, Guo M, Xu Y, Liu X. Development and validation of multi-omics thymoma risk classification model based on transfer learning. J Digit Imaging. 2023;36(5):2015–24.
https://doi.org/10.1007/s10278-023-00855-4 DOI: https://doi.org/10.1007/s10278-023-00855-4
Kocak B, Ates E, Durmaz ES, Ulusan MB, Kilickesmez O. Influence of segmentation margin on machine learning-based high-dimensional quantitative CT texture analysis: a reproducibility study on renal clear cell carcinomas. Eur Radiol. 2019;29(9):4765–75.
https://doi.org/10.1007/s00330-019-6003-8 DOI: https://doi.org/10.1007/s00330-019-6003-8
Ma X, Wei J, Gu D, Zhu Y, Feng B, Liang M, et al. Preoperative radiomics nomogram for microvascular invasion prediction in hepatocellular carcinoma using contrast-enhanced CT. Eur Radiol. 2019;29(7):3595–605.
https://doi.org/10.1007/s00330-018-5985-y DOI: https://doi.org/10.1007/s00330-018-5985-y
van Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational radiomics system to decode the radiographic phenotype. Cancer Res. 2017;77(21):e104–7.
https://doi.org/10.1158/0008-5472.CAN-17-0339 DOI: https://doi.org/10.1158/0008-5472.CAN-17-0339
Xu L, Yang P, Liang W, Liu W, Wang W, Luo C, et al. A radiomics approach based on support vector machine using MR images for preoperative lymph node status evaluation in intrahepatic cholangiocarcinoma. Theranostics. 2019;9(18):5374–85.
https://doi.org/10.7150/thno.34149 DOI: https://doi.org/10.7150/thno.34149
Wu J, Aguilera T, Shultz D, Gudur M, Rubin DL, Loo BW Jr, et al. Early-stage non-small cell lung cancer: quantitative imaging characteristics of (18)F fluorodeoxyglucose PET/CT allow prediction of distant metastasis. Radiology. 2016;281(1):270–8.
https://doi.org/10.1148/radiol.2016151829 DOI: https://doi.org/10.1148/radiol.2016151829
Wang W, Liang H, Zhang Z, Xu C, Wei D, Li W, et al. Comparing three-dimensional and two-dimensional deep-learning, radiomics, and fusion models for predicting occult lymph node metastasis in laryngeal squamous cell carcinoma based on CT imaging: a multicentre, retrospective, diagnostic study. EClinicalMedicine. 2024;67:102385.
https://doi.org/10.1016/j.eclinm.2023.102385 DOI: https://doi.org/10.1016/j.eclinm.2023.102385
Rodríguez-Pérez R, Bajorath J. Interpretation of compound activity predictions from complex machine learning models using local approximations and Shapley value. J Med Chem. 2020;63(16):8761–77.
https://doi.org/10.1021/acs.jmedchem.9b01101 DOI: https://doi.org/10.1021/acs.jmedchem.9b01101
Huang YY, Wu LL, Liu X, Liang SH, Ma GW. Nomogram predict relapse-free survival of patients with thymic epithelial tumors after surgery. BMC Cancer. 2021;21(1):847.
https://doi.org/10.1186/s12885-021-08585-y DOI: https://doi.org/10.1186/s12885-021-08585-y
Ströbel P, Bauer A, Puppe B, Kraushaar T, Krein A, Toyka K, et al. Tumor recurrence and survival in patients treated for thymomas and thymic squamous cell carcinomas: a retrospective analysis. J Clin Oncol. 2004;22(8):1501–9.
https://doi.org/10.1200/JCO.2004.10.113 DOI: https://doi.org/10.1200/JCO.2004.10.113
Padda SK, Terrone D, Tian L, Khuong A, Neal JW, Riess JW, et al. Computed tomography features associated with the eighth edition TNM stage classification for thymic epithelial tumors. J Thorac Imaging. 2018;33(3):176–83.
https://doi.org/10.1097/RTI.0000000000000310 DOI: https://doi.org/10.1097/RTI.0000000000000310
Li Y, Liu X, Qian Z, Sun Z, Xu K, Wang K, et al. Genotype prediction of ATRX mutation in lower-grade gliomas using an MRI radiomics signature. Eur Radiol. 2018;28(7):2960–8.
https://doi.org/10.1007/s00330-017-5267-0 DOI: https://doi.org/10.1007/s00330-017-5267-0
Tomiyama N, Johkoh T, Mihara N, Honda O, Kozuka T, Koyama M, et al. Using the World Health Organization classification of thymic epithelial neoplasms to describe CT findings. AJR Am J Roentgenol. 2002;179(4):881–6.
https://doi.org/10.2214/ajr.179.4.1790881 DOI: https://doi.org/10.2214/ajr.179.4.1790881
Liu W, Wang W, Guo R, Zhang H, Guo M. Deep learning for risk stratification of thymoma pathological subtypes based on preoperative CT images. BMC Cancer. 2024;24(1):651.
https://doi.org/10.1186/s12885-024-12394-4 DOI: https://doi.org/10.1186/s12885-024-12394-4
Lambin P, Leijenaar RTH, Deist TM, Peerlings J, de Jong EEC, van Timmeren J, et al. Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol. 2017;14(12):749–62.
https://doi.org/10.1038/nrclinonc.2017.141 DOI: https://doi.org/10.1038/nrclinonc.2017.141
Additional Files
Published
How to Cite
Issue
Section
Categories
License
Copyright (c) 2025 Yuhua Yang, Jia Cheng, Can Cui, Huijie Huang, Meiling Cheng, Jiayi Wang, Minjing Zuo
This work is licensed under a Creative Commons Attribution 4.0 International License.
