Computed Tomography-Based Radiomics Diagnostic Model for Fat-Poor Small Renal Tumor Subtypes
Abstract
:1. Introduction
2. Materials and Methods
2.1. Patients
2.2. Radiomics Feature Analysis Workflow
2.3. Radiomics Feature Extraction
2.4. Feature Selection and Dimensionality Reduction
2.5. Reproducibility and Ethical Considerations
3. Results
3.1. Demographic Characteristics of Patients
3.2. Multi-Phase CT Radiomic Features for Renal Cell Tumor Subtype Classification
3.3. Single-Phase CT Radiomic Features for Renal Cell Tumor Subtype Classification
3.4. Comparison of Multi-Phase and Single-Phase Models
4. Discussion
5. Conclusions
Code Availability
Supplementary Materials
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Moch, H.; Cubilla, A.L.; Humphrey, P.A.; Reuter, V.E.; Ulbright, T.M. The 2016 WHO classification of tumours of the urinary system and male genital organs—Part A: Renal, penile, and testicular tumours. Eur. Urol. 2016, 70, 93–105. [Google Scholar] [CrossRef] [PubMed]
- American Cancer Society. Key Statistics About Kidney Cancer. Available online: https://www.cancer.org/cancer/types/kidney-cancer/about/key-statistics.html (accessed on 9 February 2025).
- Capitanio, U.; Bensalah, K.; Bex, A.; Boorjian, S.A.; Bray, F.; Coleman, J.; Gore, J.L.; Sun, M.; Wood, C.; Russo, P. Epidemiology of Renal Cell Carcinoma. Eur. Urol. 2019, 75, 74–84. (In English) [Google Scholar] [CrossRef] [PubMed]
- Kang, S.K.; Huang, W.C.; Pandharipande, P.V.; Chandarana, H. Solid renal masses: What the numbers tell us. AJR Am. J. Roentgenol. 2014, 202, 1196–1206. (In English) [Google Scholar] [CrossRef]
- Prasad, S.R.; Dalrymple, N.C.; Surabhi, V.R. Cross-sectional imaging evaluation of renal masses. Radiol. Clin. N. Am. 2008, 46, 95–111. [Google Scholar] [CrossRef]
- Young, J.R.; Margolis, D.; Sauk, S.; Pantuck, A.J.; Sayre, J.; Raman, S.S. Clear cell renal cell carcinoma: Discrimination from other renal cell carcinoma subtypes and oncocytoma at multiphasic multidetector CT. Radiology 2013, 267, 444–453. [Google Scholar] [CrossRef]
- Bauman, T.M.; Potretzke, A.M.; Wright, A.J.; Knight, B.A.; Vetter, J.M.; Figenshau, R.S. Partial nephrectomy for presumed renal-cell carcinoma: Incidence, predictors, and perioperative outcomes of benign lesions. J. Endourol. 2017, 31, 412–417. [Google Scholar] [CrossRef]
- Tanaka, T.; Huang, Y.; Marukawa, Y.; Tsuboi, Y.; Masaoka, Y.; Kojima, K.; Iguchi, T.; Hiraki, T.; Gobara, H.; Yanai, H. Differentiation of small (≤4 cm) renal masses on multiphase contrast-enhanced CT by deep learning. Am. J. Roentgenol. 2020, 214, 605–612. [Google Scholar] [CrossRef]
- De Fauw, J.; Ledsam, J.R.; Romera-Paredes, B.; Nikolov, S.; Tomasev, N.; Blackwell, S.; Askham, H.; Glorot, X.; O’Donoghue, B.; Visentin, D. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat. Med. 2018, 24, 1342–1350. [Google Scholar] [CrossRef]
- Esteva, A.; Kuprel, B.; Novoa, R.A.; Ko, J.; Swetter, S.M.; Blau, H.M.; Thrun, S. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2017, 542, 115–118. [Google Scholar] [CrossRef]
- Mei, X.; Lee, H.-C.; Diao, K.-y.; Huang, M.; Lin, B.; Liu, C.; Xie, Z.; Ma, Y.; Robson, P.M.; Chung, M. Artificial intelligence–enabled rapid diagnosis of patients with COVID-19. Nat. Med. 2020, 26, 1224–1228. [Google Scholar] [CrossRef]
- Ardila, D.; Kiraly, A.P.; Bharadwaj, S.; Choi, B.; Reicher, J.J.; Peng, L.; Tse, D.; Etemadi, M.; Ye, W.; Corrado, G. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nat. Med. 2019, 25, 954–961. [Google Scholar] [CrossRef] [PubMed]
- Uhm, K.H.; Jung, S.W.; Choi, M.H.; Shin, H.K.; Yoo, J.I.; Oh, S.W.; Kim, J.Y.; Kim, H.G.; Lee, Y.J.; Youn, S.Y.; et al. Deep learning for end-to-end kidney cancer diagnosis on multi-phase abdominal computed tomography. NPJ Precis. Oncol. 2021, 5, 54. (In English) [Google Scholar] [CrossRef] [PubMed]
- Zhao, W.; Jiang, D.; Queralta, J.P.; Westerlund, T. MSS U-Net: 3D segmentation of kidneys and tumors from CT images with a multi-scale supervised U-Net. Inform. Med. Unlocked 2020, 19, 100357. [Google Scholar] [CrossRef]
- Kim, S.H.; Kim, C.S.; Kim, M.J.; Cho, J.Y.; Cho, S.H. Differentiation of clear cell renal cell carcinoma from other subtypes and fat-poor angiomyolipoma by use of quantitative enhancement measurement during three-phase MDCT. Am. J. Roentgenol. 2016, 206, W21–W28. [Google Scholar] [CrossRef]
- Sun, X.-Y.; Feng, Q.-X.; Xu, X.; Zhang, J.; Zhu, F.-P.; Yang, Y.-H.; Zhang, Y.-D. Radiologic-radiomic machine learning models for differentiation of benign and malignant solid renal masses: Comparison with expert-level radiologists. Am. J. Roentgenol. 2020, 214, W44–W54. [Google Scholar] [CrossRef]
- de Leon, A.D.; Pedrosa, I. Imaging and screening of kidney cancer. Radiol. Clin. 2017, 55, 1235–1250. [Google Scholar] [CrossRef]
- Bang, S.; Wang, H.-H.; Kim, H.; Choi, M.H.; Cha, J.; Choi, Y.; Hong, S.-H. Development and Validation of a Prediction Model for Differentiation of Benign and Malignant Fat-Poor Renal Tumors Using CT Radiomics. Appl. Sci. 2023, 13, 11345. [Google Scholar] [CrossRef]
- Oberai, A.; Varghese, B.; Cen, S.; Angelini, T.; Hwang, D.; Gill, I.; Aron, M.; Lau, C.; Duddalwar, V. Deep learning based classification of solid lipid-poor contrast enhancing renal masses using contrast enhanced CT. Br. J. Radiol. 2020, 93, 20200002. [Google Scholar] [CrossRef]
- Kaur, R.; Juneja, M.; Mandal, A.K. Computer-aided diagnosis of renal lesions in CT images: A comprehensive survey and future prospects. Comput. Electr. Eng. 2019, 77, 423–434. [Google Scholar] [CrossRef]
- Chen, T.; Guestrin, C. Xgboost: A scalable tree boosting system. In Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016; pp. 785–794. [Google Scholar]
- Akiba, T.; Sano, S.; Yanase, T.; Ohta, T.; Koyama, M. Optuna: A next-generation hyperparameter optimization framework. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, AK, USA, 4–8 August 2019; pp. 2623–2631. [Google Scholar]
- LemaÃŽtre, G.; Nogueira, F.; Aridas, C.K. Imbalanced-learn: A python toolbox to tackle the curse of imbalanced datasets in machine learning. J. Mach. Learn. Res. 2017, 18, 1–5. [Google Scholar]
- Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830. [Google Scholar]
- Klontzas, M.E.; Gatti, A.A.; Tejani, A.S.; Kahn, C.E., Jr. AI reporting guidelines: How to select the best one for your research. Radiol. Artif. Intell. 2023, 5, e230055. [Google Scholar] [CrossRef] [PubMed]
- Van Griethuysen, J.J.; Fedorov, A.; Parmar, C.; Hosny, A.; Aucoin, N.; Narayan, V.; Beets-Tan, R.G.; Fillion-Robin, J.-C.; Pieper, S.; Aerts, H.J. Computational radiomics system to decode the radiographic phenotype. Cancer Res. 2017, 77, e104–e107. [Google Scholar] [CrossRef]
- van Timmeren, J.E.; Cester, D.; Tanadini-Lang, S.; Alkadhi, H.; Baessler, B. Radiomics in medical imaging-“how-to” guide and critical reflection. Insights Imaging 2020, 11, 91. (In English) [Google Scholar] [CrossRef]
- Li, S.; Wang, K.; Hou, Z.; Yang, J.; Ren, W.; Gao, S.; Meng, F.; Wu, P.; Liu, B.; Liu, J.; et al. Use of Radiomics Combined With Machine Learning Method in the Recurrence Patterns After Intensity-Modulated Radiotherapy for Nasopharyngeal Carcinoma: A Preliminary Study. Front. Oncol. 2018, 8, 648. (In English) [Google Scholar] [CrossRef]
- Shi, Y.; Zou, Y.; Liu, J.; Wang, Y.; Chen, Y.; Sun, F.; Yang, Z.; Cui, G.; Zhu, X.; Cui, X. Ultrasound-based radiomics XGBoost model to assess the risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual application of SHAP. Front. Oncol. 2022, 12, 897596. [Google Scholar] [CrossRef]
- Mühlbauer, J.; Egen, L.; Kowalewski, K.-F.; Grilli, M.; Walach, M.T.; Westhoff, N.; Nuhn, P.; Laqua, F.C.; Baessler, B.; Kriegmair, M.C. Radiomics in renal cell carcinoma—A systematic review and meta-analysis. Cancers 2021, 13, 1348. [Google Scholar] [CrossRef]
- Bertsimas, D.; Wiberg, H. Machine learning in oncology: Methods, applications, and challenges. JCO Clin. Cancer Inform. 2020, 4, CCI-20. [Google Scholar] [CrossRef]
- Alhussaini, A.J.; Steele, J.D.; Jawli, A.; Nabi, G. Radiomics Machine Learning Analysis of Clear Cell Renal Cell Carcinoma for Tumour Grade Prediction Based on Intra-Tumoural Sub-Region Heterogeneity. Cancers 2024, 16, 1454. [Google Scholar] [CrossRef]
- Lambin, P.; Leijenaar, R.T.; Deist, T.M.; Peerlings, J.; De Jong, E.E.; Van Timmeren, J.; Sanduleanu, S.; Larue, R.T.; Even, A.J.; Jochems, A. Radiomics: The bridge between medical imaging and personalized medicine. Nat. Rev. Clin. Oncol. 2017, 14, 749–762. [Google Scholar] [CrossRef]
- Uhlig, A.; Uhlig, J.; Leha, A.; Biggemann, L.; Bachanek, S.; Stöckle, M.; Reichert, M.; Lotz, J.; Zeuschner, P.; Maßmann, A. Radiomics and machine learning for renal tumor subtype assessment using multiphase computed tomography in a multicenter setting. Eur. Radiol. 2024, 34, 6254–6263. [Google Scholar] [CrossRef]
- Jeon, H.G.; Seo, S.I.; Jeong, B.C.; Jeon, S.S.; Lee, H.M.; Choi, H.-Y.; Song, C.; Hong, J.H.; Kim, C.-S.; Ahn, H. Percutaneous kidney biopsy for a small renal mass: A critical appraisal of results. J. Urol. 2016, 195, 568–573. [Google Scholar] [CrossRef]
Training Dataset | Test Dataset | p Value | |
---|---|---|---|
Sex | 289 | 210 | 0.639 |
Male (%) | 162 (49.2%) | 123 (58.6%) | |
Female (%) | 127 (50.8%) | 87 (41.4%) | |
Age (years) (range) | 57.0 (22, 83) | 57.0 (27, 81) | 0.816 |
Cancer size (cm) (range) | 3.629 (0.8, 17.5) | 3.357 (0.7, 12.0) | 0.216 |
Renal cell tumor subtype | 0.881 | ||
oncocytoma (%) | 25 (8.6%) | 23 (11%) | |
angiomyolipoma (%) | 47 (16.3%) | 31 (14.7%) | |
chromophobe renal cell carcinoma (%) | 48 (16.6%) | 36 (17.1%) | |
papillary renal cell carcinoma (%) | 50 (17.3%) | 39 (18.6%) | |
clear cell renal cell carcinoma (%) | 119 (41.2%) | 81 (38.6%) |
(A) | ||||||||||
F1 | AU-PRC | ACC | AU-ROC | |||||||
Linear SVM | 0.423 ± 0.008 | 0.393 ± 0.016 | 0.423 ± 0.008 | 0.703 ± 0.009 | ||||||
Rbf SVM | 0.433 ± 0.017 | 0.408 ± 0.011 | 0.433 ± 0.017 | 0.722 ± 0.006 | ||||||
XGBoost | 0.453 ± 0.007 | 0.474 ± 0.006 | 0.453 ± 0.007 | 0.744 ± 0.004 | ||||||
Random Forest | 0.444 ± 0.009 | 0.443 ± 0.006 | 0.444 ± 0.009 | 0.742 ± 0.003 | ||||||
(B) | ||||||||||
AU-PRC | AU-ROC | |||||||||
Oncocytoma | AML | chRCC | pRCC | ccRCC | Oncocytoma | AML | chRCC | pRCC | ccRCC | |
vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | |
Linear SVM | 0.096 ± 0.01 | 0.285 ± 0.039 | 0.257 ± 0.044 | 0.246 ± 0.022 | 0.613 ± 0.019 | 0.511 ± 0.042 | 0.688 ± 0.03 | 0.605 ± 0.037 | 0.607 ± 0.023 | 0.697 ± 0.011 |
Rbf SVM | 0.101 ± 0.014 | 0.292 ± 0.017 | 0.242 ± 0.013 | 0.276 ± 0.02 | 0.570 ± 0.023 | 0.488 ± 0.042 | 0.707 ± 0.017 | 0.618 ± 0.01 | 0.648 ± 0.022 | 0.663 ± 0.015 |
XGBoost | 0.151 ± 0.018 | 0.31 ± 0.009 | 0.3 ± 0.015 | 0.318 ± 0.015 | 0.673 ± 0.08 | 0.606 ± 0.021 | 0.698 ± 0.009 | 0.666 ± 0.007 | 0.681 ± 0.014 | 0.738 ± 0.007 |
Random Forest | 0.142 ± 0021 | 0.366 ± 0.01 | 0.259 ± 0.022 | 0.298 ± 0.017 | 0.617 ± 0.011 | 0.616 ± 0.017 | 0.769 ± 0.009 | 0.607 ± 0.014 | 0.686 ± 0.016 | 0.695 ± 0.01 |
(C) | ||||||||||
F1 | AU-PRC | ACC | AU-ROC | |||||||
Linear SVM | 0.423 ± 0.011 | 0.396 ± 0.019 | 0.423 ± 0.011 | 0.716 ± 0.001 | ||||||
Rbf SVM | 0.424 ± 0.0 | 0.322 ± 0.001 | 0.424 ± 0.0 | 0.66 ± 0.004 | ||||||
XGBoost | 0.475 ± 0.011 | 0.495 ± 0.009 | 0.475 ± 0.011 | 0.751 ± 0.006 | ||||||
Random Forest | 0.433 ± 0.005 | 0.47 ± 0.011 | 0.433 ± 0.005 | 0.75 ± 0.003 | ||||||
(D) | ||||||||||
AU-PRC | AU-ROC | |||||||||
Oncocytoma | AML | chRCC | pRCC | ccRCC | Oncocytoma | AML | chRCC | pRCC | ccRCC | |
vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | vs. Rest | |
Linear SVM | 0.141 ± 0.015 | 0.278 ± 0.022 | 0.230 ± 0.032 | 0.288 ± 0.020 | 0.598 ± 0.04 | 0.631 ± 0.027 | 0.734 ± 0.013 | 0.574 ± 0.042 | 0.658 ± 0.017 | 0.673 ± 0.026 |
Rbf SVM | 0.096 ± 0.0 | 0.148 ± 0.0 | 0.164 ± 0.0 | 0.168 ± 0.0 | 0.424 ± 0.0 | 0.5 ± 0.0 | 0.5 ± 0.0 | 0.5 ± 0.0 | 0.5 ± 0.0 | 0.5 ± 0.0 |
XGBoost | 0.124 ± 0.012 | 0.410 ± 0.029 | 0.289 ± 0.017 | 0.294 ± 0.017 | 0.686 ± 0.012 | 0.578 ± 0.029 | 0.748 ± 0.006 | 0.646 ± 0.013 | 0.674 ± 0.02 | 0.750 ± 0.011 |
Random Forest | 0.183 ± 0.033 | 0.33 ± 0.015 | 0.258 ± 0.012 | 0.316 ± 0.017 | 0.677 ± 0.020 | 0.645 ± 0.026 | 0.743 ± 0.014 | 0.672 ± 0.016 | 0.672 ± 0.018 | 0.742 ± 0.013 |
(A) | |||||||||||||||||||||||
Arterial Phase | Delayed Phase | Non-Contrast Phase | Portal Phase | ||||||||||||||||||||
Training Dataset | Test Dataset | Training Dataset | Test Dataset | Training Dataset | Test Dataset | Training Dataset | Test Dataset | ||||||||||||||||
Renal cell tumor subtype | 243 | 75 | 257 | 83 | 280 | 139 | 253 | 188 | |||||||||||||||
oncocytoma (%) | 22 (9%) | 6 (8%) | 23 (9%) | 8 (10%) | 25 (9%) | 13 (10%) | 20 (8%) | 21 (11%) | |||||||||||||||
angiomyolipoma (%) | 40 (16%) | 10 (13%) | 42 (16%) | 10 (12%) | 44 (16%) | 20 (14%) | 42 (17%) | 27 (14%) | |||||||||||||||
chromophobe renal cell carcinoma (%) | 43 (18%) | 13 (17%) | 42 (16%) | 15 (18%) | 48 (17%) | 25 (18%) | 40 (16%) | 28 (15%) | |||||||||||||||
papillary renal cell carcinoma (%) | 36 (15%) | 14 (19%) | 43 (17%) | 13 (16%) | 48 (17%) | 24 (17%) | 46 (18%) | 37 (20%) | |||||||||||||||
clear cell renal cell carcinoma (%) | 102 (42%) | 32 (43%) | 107 (42%) | 37 (44%) | 115 (41%) | 57 (41%) | 105 (41%) | 75 (40%) | |||||||||||||||
(B) | |||||||||||||||||||||||
F1 score | AU-PRC | ACC | AU-ROC | ||||||||||||||||||||
CT Phase | Arterial | Delayed | Non-contrast | Portal | Arterial | Delayed | Non-contrast | Portal | Arterial | Delayed | Non-contrast | Portal | Arterial | Delayed | Non-contrast | Portal | |||||||
Linear SVM | 0.421 ± 0.036 | 0.378 ± 0.032 | 0.397 ± 0.02 | 0.371 ± 0.011 | 0.42 ± 0.026 | 0.368 ± 0.032 | 0.345 ± 0.018 | 0.341 ± 0.019 | 0.421 ± 0.036 | 0.378 ± 0.032 | 0.397 ± 0.02 | 0.371 ± 0.011 | 0.719 ± 0.022 | 0.676 ± 0.024 | 0.671 ± 0.019 | 0.659 ± 0.012 | |||||||
Rbf SVM | 0.461 ± 0.028 | 0.477 ± 0.009 | 0.425 ± 0.006 | 0.386 ± 0.004 | 0.48 ± 0.051 | 0.416 ± 0.031 | 0.303 ± 0.024 | 0.525 ± 0.05 | 0.461 ± 0.028 | 0.477 ± 0.09 | 0.425 ± 0.006 | 0.386 ± 0.04 | 0.756 ± 0.027 | 0.725 ± 0.021 | 0.67 ± 0.023 | 0.629 ± 0.018 | |||||||
XGBoost | 0.594 ± 0.037 | 0.479 ± 0.009 | 0.439 ± 0.015 | 0.456 ± 0.015 | 0.608 ± 0.022 | 0.442 ± 0.18 | 0.419 ± 0.014 | 0.463 ± 0.012 | 0.594 ± 0.037 | 0.479 ± 0.009 | 0.439 ± 0.015 | 0.456 ± 0.015 | 0.83 ± 0.012 | 0.728 ± 0.013 | 0.712 ± 0.006 | 0.739 ± 0.007 | |||||||
Random Forest | 0.478 ± 0.013 | 0.493 ± 0.015 | 0.461 ± 0.011 | 0.441 ± 0.014 | 0.511 ± 0.016 | 0.492 ± 0.31 | 0.471 ± 0.013 | 0.473 ± 0.015 | 0.478 ± 0.013 | 0.493 ± 0.015 | 0.461 ± 0.011 | 0.441 ± 0.014 | 0.778 ± 0.013 | 0.766 ± 0.016 | 0.744 ± 0.008 | 0.756 ± 0.008 | |||||||
(C) | |||||||||||||||||||||||
F1 score | AU-PRC | ACC | AU-ROC | ||||||||||||||||||||
CT Phase | Arterial | Delayed | Non-contrast | Portal | Arterial | Delayed | Non-contrast | Portal | Arterial | Delayed | Non-contrast | Portal | Arterial | Delayed | Non-contrast | Portal | |||||||
Linear SVM | 0.396 ± 0.028 | 0.385 ± 0.027 | 0.414 ± 0.022 | 0.297 ± 0.045 | 0.353 ± 0.048 | 0.321 ± 0.036 | 0.354 ± 0.031 | 0.267 ± 0.031 | 0.396 ± 0.028 | 0.385 ± 0.027 | 0.414 ± 0.022 | 0.297 ± 0.045 | 0.669 ± 0.036 | 0.627 ± 0.032 | 0.688 ± 0.022 | 0.581 ± 0.026 | |||||||
Rbf SVM | 0.437 ± 0.0 | 0.462 ± 0.0 | 0.413 ± 0.0 | 0.382 ± 0.0 | 0.337 ± 0.01 | 0.337 ± 0.016 | 0.317 ± 0.002 | 0.296 ± 0.007 | 0.437 ± 0.0 | 0.462 ± 0.0 | 0.413 ± 0.0 | 0.382 ± 0.0 | 0.677 ± 0.006 | 0.67 ± 0.008 | 0.665 ± 0.005 | 0.633 ± 0.007 | |||||||
XGBoost | 0.531 ± 0.02 | 0.466 ± 0.029 | 0.354 ± 0.024 | 0.447 ± 0.019 | 0.565 ± 0.014 | 0.507 ± 0.021 | 0.382 ± 0.022 | 0.429 ± 0.013 | 0.531 ± 0.002 | 0.466 ± 0.029 | 0.354 ± 0.024 | 0.447 ± 0.019 | 0793 ± 0.005 | 0.766 ± 0.001 | 0.686 ± 0.023 | 0.732 ± 0.008 | |||||||
Random Forest | 0.504 ± 0.022 | 0.47 ± 0.016 | 0.45 ± 0.006 | 0.427 ± 0.013 | 0.52 ± 0.026 | 0.529 ± 0.02 | 0.433 ± 0.009 | 0.449 ± 0.014 | 0.504 ± 0.022 | 0.497 ± 0.016 | 0.45 ± 0.006 | 0.427 ± 0.013 | 0.797 ± 0.018 | 0.73 ± 0.014 | 0.16 ± 0.007 | 0.735 ± 0.007 |
(A) | ||||||||||||||||||||
Arterial | Delayed | Non-Contrast | Portal | |||||||||||||||||
CT Phase | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC |
Linear SVM | 0.147 ± 0.068 | 0.290 ± 0.077 | 0.280 ± 0.028 | 0.298 ± 0.078 | 0.656 ± 0.065 | 0.151 ± 0.056 | 0.205 ± 0.041 | 0.310 ± 0.079 | 0.260 ± 0.059 | 0.564 ± 0.037 | 0.109 ± 0.016 | 0.214 ± 0.025 | 0.202 ± 0.051 | 0.206 ± 0.036 | 0.503 ± 0.030 | 0.141 ± 0.016 | 0.194 ± 0.027 | 0.217 ± 0.019 | 0.275 ± 0.028 | 0.550 ± 0.054 |
Rbf SVM | 0.133 ± 0.038 | 0.382 ± 0.135 | 0.285 ± 0.041 | 0.598 ± 0.119 | 0.638 ± 0.085 | 0.128 ± 0.059 | 0.241 ± 0.044 | 0.268 ± 0.1 | 0.421 ± 0.095 | 0.547 ± 0.07 | 0.084 ± 0.008 | 0.340 ± 0.105 | 0.170 ± 0.013 | 0.168 ± 0.038 | 0.463 ± 0.034 | 0.117 ± 0.014 | 0.190 ± 0.060 | 0.148 ± 0.008 | 0.200 ± 0.025 | 0.384 ± 0.048 |
XGBoost | 0.089 ± 0.017 | 0.605 ± 0.076 | 0.590 ± 0.063 | 0.393 ± 0.1 | 0.757 ± 0.033 | 0.105 ± 0.01 | 0.420 ± 0.028 | 0.351 ± 0.038 | 0.346 ± 0.049 | 0.586 ± 0.023 | 0.111 ± 0.015 | 0.401 ± 0.022 | 0.320 ± 0.033 | 0.196 ± 0.016 | 0.556 ± 0.024 | 0.189 ± 0.033 | 0.411 ± 0.016 | 0.273 ± 0.035 | 0.333 ± 0.31 | 0.682 ± 0.02 |
Random Forest | 0.163 ± 0.026 | 0.384 ± 0.035 | 0.265 ± 0.049 | 0.413 ± 0.052 | 0.711 ± 0.026 | 0.158 ± 0.023 | 0.356 ± 0.021 | 0.277 ± 0.038 | 0.344 ± 0.093 | 0.712 ± 0.046 | 0.149 ± 0.029 | 0.499 ± 0.023 | 0.239 ± 0.027 | 0.238 ± 0.019 | 0.637 ± 0.039 | 0.214 ± 0.032 | 0.479 ± 0.052 | 0.271 ± 0.027 | 0.295 ± 0.016 | 0.643 ± 0.02 |
(B) | ||||||||||||||||||||
Arterial | Delayed | Non-Contrast | Portal | |||||||||||||||||
CT Phase | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC |
Linear SVM | 0.625 ± 0.149 | 0.681 ± 0.073 | 0.572 ± 0.053 | 0.687 ± 0.056 | 0.663 ± 0.049 | 0.584 ± 0.094 | 0.631 ± 0.048 | 0.561 ± 0.084 | 0.645 ± 0.067 | 0.595 ± 0.042 | 0.529 ± 0.056 | 0.685 ± 0.049 | 0.465 ± 0.075 | 0.492 ± 0.049 | 0.595 ± 0.022 | 0.561 ± 0.040 | 0.639 ± 0.047 | 0.537 ± 0.053 | 0.584 ± 0.046 | 0.655 ± 0.044 |
Rbf SVM | 0.133 ± 0.038 | 0.382 ± 0.135 | 0.285 ± 0.041 | 0.598 ± 0.119 | 0.638 ± 0.084 | 0.444 ± 0.100 | 0.570 ± 0.079 | 0.547 ± 0.117 | 0.699 ± 0.113 | 0.635 ± 0.082 | 0.458 ± 0.064 | 0.580 ± 0.139 | 0.415 ± 0.065 | 0.452 ± 0.090 | 0.524 ± 0.047 | 0.504 ± 0.055 | 0.424 ± 0.133 | 0.459 ± 0.018 | 0.442 ± 0.018 | 0.500 ± 0.074 |
XGBoost | 0.507 ± 0.05 | 0.843 ± 0.028 | 0.775 ± 0.032 | 0.781 ± 0.04 | 0.824 ± 0.023 | 0.500 ± 0.042 | 0.715 ± 0.019 | 0.707 ± 0.029 | 0.616 ± 0.041 | 0.675 ± 0.011 | 0.513 ± 0.049 | 0.693 ± 0.02 | 0.668 ± 0.026 | 0.522 ± 0.014 | 0.645 ± 0.018 | 0.580 ± 0.033 | 0.785 ± 0.016 | 0.659 ± 0.024 | 0.667 ± 0.015 | 0.761 ± 0.013 |
Random Forest | 0.664 ± 0.052 | 0.755 ± 0.045 | 0.636 ± 0.048 | 0.769 ± 0.033 | 0.775 ± 0.026 | 0.644 ± 0.059 | 0.749 ± 0.025 | 0.659 ± 0.037 | 0.637 ± 0.113 | 0.785 ± 0.037 | 0.603 ± 0.056 | 0.780 ± 0.026 | 0.585 ± 0.019 | 0.577 ± 0.038 | 0.706 ± 0.025 | 0.669 ± 0.033 | 0.806 ± 0.017 | 0.651 ± 0.022 | 0.656 ± 0.023 | 0.731 ± 0.017 |
(C) | ||||||||||||||||||||
Arterial | Delayed | Non-Contrast | Portal | |||||||||||||||||
CT Phase | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC |
Linear SVM | 0.106 ± 0.06 | 0.217 ± 0.085 | 0.188 ± 0.075 | 0.207 ± 0.063 | 0.583 ± 0.059 | 0.111 ± 0.025 | 0.189 ± 0.082 | 0.196 ± 0.049 | 0.218 ± 0.054 | 0.469 ± 0.061 | 0.129 ± 0.030 | 0.331 ± 0.055 | 0.207 ± 0.04 | 0.199 ± 0.035 | 0.512 ± 0.042 | 0.111 ± 0.017 | 0.163 ± 0.029 | 0.17 ± 0.018 | 0.217 ± 0.025 | 0.408 ± 0.04 |
Rbf SVM | 0.185 ± 0.072 | 0.215 ± 0.046 | 0.165 ± 0.007 | 0.168 ± 0.015 | 0.420 ± 0.019 | 0.18 ± 0.049 | 0.212 ± 0.097 | 0.162 ± 0.024 | 0.181 ± 0.043 | 0.412 ± 0.031 | 0.091 ± 0.0 | 0.147 ± 0.0 | 0.175 ± 0.0 | 0.175 ± 0.0 | 0.413 ± 0.0 | 0.111 ± 0.009 | 0.248 ± 0.034 | 0.154 ± 0.009 | 0.205 ± 0.024 | 0.361 ± 0.02 |
XGBoost | 0.177 ± 0.056 | 0.584 ± 0.037 | 0.261 ± 0.023 | 0.393 ± 0.067 | 0.726 ± 0.026 | 0.184 ± 0.077 | 0.399 ± 0.048 | 0.239 ± 0.026 | 0.34 ± 0.036 | 0.714 ± 0.032 | 0.119 ± 0.025 | 0.443 ± 0.048 | 0.22 ± 0.023 | 0.222 ± 0.025 | 0.549 ± 0.029 | 0.147 ± 0.016 | 0.452 ± 0.038 | 0.318 ± 0.045 | 0.394 ± 0.027 | 0.552 ± 0.025 |
Random Forest | 0.249 ± 0.072 | 0.451 ± 0.028 | 0.326 ± 0.079 | 0.368 ± 0.076 | 0.671 ± 0.041 | 0.136 ± 0.02 | 0.533 ± 0.072 | 0.252 ± 0.052 | 0.388 ± 0.051 | 0.741 ± 0.032 | 0.141 ± 0.025 | 0.395 ± 0.021 | 0.212 ± 0.04 | 0.221 ± 0.032 | 0.627 ± 0.022 | 0.123 ± 0.021 | 0.428 ± 0.044 | 0.261 ± 0.027 | 0.318 ± 0.031 | 0.622 ± 0.034 |
(D) | ||||||||||||||||||||
Arterial | Delayed | Non-Contrast | Portal | |||||||||||||||||
CT Phase | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC | Onco- cytoma | AML | chRCC | pRCC | ccRCC |
Linear SVM | 0.509 ± 0.135 | 0.566 ± 0.135 | 0.478 ± 0.132 | 0.537 ± 0.139 | 0.580 ± 0.061 | 0.466 ± 0.094 | 0.514 ± 0.161 | 0.481 ± 0.079 | 0.475 ± 0.094 | 0.5 ± 0.066 | 0.598 ± 0.068 | 0.744 ± 0.036 | 0.53 ± 0.044 | 0.488 ± 0.062 | 0.605 ± 0.034 | 0.46 ± 0.058 | 0.495 ± 0.079 | 0.484 ± 0.043 | 0.493 ± 0.041 | 0.491 ± 0.055 |
Rbf SVM | 0.497 ± 0.085 | 0.56 ± 0.105 | 0.457 ± 0.045 | 0.441 ± 0.039 | 0.43 ± 0.049 | 0.492 ± 0.061 | 0.529 ± 0.115 | 0.429 ± 0.066 | 0.433 ± 0.082 | 0.387 ± 0.072 | 0.5 ± 0.0 | 0.5 ± 0. | 0.5 ± 0. | 0.5 ± 0. | 0.5 ± 0. | 0.483 ± 0.041 | 0.587 ± 0.042 | 0.412 ± 0.05 | 0.518 ± 0.031 | 0.409 ± 0.045 |
XGBoost | 0.489 ± 0.039 | 0.805 ± 0.028 | 0.671 ± 0.027 | 0.707 ± 0.029 | 0.751 ± 0.015 | 0.594 ± 0.054 | 0.762 ± 0.019 | 0.630 ± 0.016 | 0.766 ± 0.029 | 0.731 ± 0.023 | 0.503 ± 0.045 | 0.747 ± 0.021 | 0.545 ± 0.022 | 0.607 ± 0.041 | 0.632 ± 0.025 | 0.560 ± 0.028 | 0.796 ± 0.019 | 0.678 ± 0.031 | 0.726 ± 0.021 | 0.648 ± 0.031 |
Random Forest | 0.648 ± 0.074 | 0.78 ± 0.028 | 0.717 ± 0.077 | 0.717 ± 0.077 | 0.716 ± 0.024 | 0.541 ± 0.062 | 0.799 ± 0.033 | 0.609 ± 0.045 | 0.712 ± 0.061 | 0.777 ± 0.03 | 0.581 ± 0.037 | 0.757 ± 0.016 | 0.513 ± 0031 | 0.554 ± 0.048 | 0.723 ± 0.028 | 0.567 ± 0.049 | 0.804 ± 0.015 | 0.658 ± 0.036 | 0.669 ± 0.025 | 0.706 ± 0.019 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Bang, S.; Wang, H.; Bae, H.; Hong, S.-H.; Cha, J.; Choi, M.H. Computed Tomography-Based Radiomics Diagnostic Model for Fat-Poor Small Renal Tumor Subtypes. Diagnostics 2025, 15, 1365. https://doi.org/10.3390/diagnostics15111365
Bang S, Wang H, Bae H, Hong S-H, Cha J, Choi MH. Computed Tomography-Based Radiomics Diagnostic Model for Fat-Poor Small Renal Tumor Subtypes. Diagnostics. 2025; 15(11):1365. https://doi.org/10.3390/diagnostics15111365
Chicago/Turabian StyleBang, Seokhwan, Heehwan Wang, Hoyoung Bae, Sung-Hoo Hong, Jiook Cha, and Moon Hyung Choi. 2025. "Computed Tomography-Based Radiomics Diagnostic Model for Fat-Poor Small Renal Tumor Subtypes" Diagnostics 15, no. 11: 1365. https://doi.org/10.3390/diagnostics15111365
APA StyleBang, S., Wang, H., Bae, H., Hong, S.-H., Cha, J., & Choi, M. H. (2025). Computed Tomography-Based Radiomics Diagnostic Model for Fat-Poor Small Renal Tumor Subtypes. Diagnostics, 15(11), 1365. https://doi.org/10.3390/diagnostics15111365