TY  - JOUR
T1  - Integration of radiomics with MRI clear cell likelihood score for classification of clear cell renal cell carcinoma
TT  - Şeffaf hücreli renal hücreli karsinomun sınıflandırılması için manyetik rezonans görüntüleme temelli şeffaf hücreli olabilirlik skoru ile radyomik analizin entegrasyonu
AU  - Karabulut, Ahmet Kasım
AU  - Koska, İlker Özgür
AU  - Turgut, Ali Çağlar
AU  - Sarsik Kumbaraci, Banu
AU  - Kızılay, Fuat
AU  - Güler, Ezgi
PY  - 2026
DA  - June
Y2  - 2026
DO  - 10.19161/etd.1882022
JF  - Ege Tıp Dergisi
JO  - EJM
PB  - Ege University
WT  - DergiPark
SN  - 1016-9113
SP  - 313
EP  - 320
VL  - 65
IS  - 2
LA  - en
AB  - Aim: To compare the diagnostic performance of the MRI-based clear cell likelihood score (ccLS), radiomics-based machine learning models, and their combination for differentiating clear cell renal cell carcinoma (ccRCC) from other renal tumor subtypes.Materials and Methods: This single-center retrospective study included patients with solid renal masses who underwent multiparametric MRI and had histopathologic confirmation. Lesions were evaluated using ccLS by two independent readers. Radiomic features were extracted from T1-weighted and T2-weighted images following standardized preprocessing. Multiple machine learning pipelines combining different feature selection methods and classifiers were evaluated using stratified 10-fold cross-validation with four repetitions. Using multivariable logistic regression, the radiomics score together with clinical factors and semantic imaging features were evaluated, and a nomogram was constructed based on the selected variables.Results: For differentiating ccRCC from other renal tumor subtypes, the ccLS model achieved an area under the receiver operating characteristic curve (AUC) of 0.832 (0.759–0.899). The best-performing radiomics-based machine learning model achieved a mean AUC of 0.89 ± 0.10. The combined model demonstrated higher diagnostic performance, with a mean AUC of 0.96 ± 0.01.Conclusions: Combining radiomics-based machine learning with established semantic MRI assessments was associated with improved differentiation of ccRCC from other renal tumor subtypes compared with individual approaches. Further validation in larger, multicenter cohorts is warranted before broader clinical application.
KW  - renal cell carcinoma
KW  - magnetic resonance imaging
KW  - clear cell
KW  - radiomics
KW  - machine learning
N2  - Amaç: Manyetik rezonans görüntüleme (MRG) tabanlı şeffaf hücreli olabilirlik skoru, radyomik tabanlı makine öğrenmesi modelleri ve bu yaklaşımların kombinasyonunun, şeffaf hücreli renal hücreli karsinomu diğer renal tümör alt tiplerinden ayırt etmedeki tanısal performanslarını karşılaştırmaktır.Gereç ve Yöntem: Tek merkezli retrospektif çalışmaya, multiparametrik MRG incelemesi olan ve histopatolojik doğrulaması mevcut solid böbrek kitlesi olan hastalar dahil edildi. Lezyonlar iki bağımsız okuyucu tarafından şeffaf hücreli olabilirlik skoru kullanılarak değerlendirildi. Radyomik öznitelikler, standart ön işleme adımlarını takiben T1-ağırlıklı ve T2-ağırlıklı görüntülerden çıkarıldı.Farklı öznitelik seçimi yöntemleri ve sınıflandırıcıları içeren modeller, dört tekrar içeren katmanlı 10-katlı çapraz doğrulama ile değerlendirildi. Çok değişkenli lojistik regresyon kullanılarak, radyomik skoru yanı sıra klinik faktörler ile semantik görüntüleme özellikleri değerlendirilerek, seçilen değişkenlerle nomogram görselleştirmesi yapıldı.Bulgular: Şeffaf hücreli olabilirlik skor modelinin eğri altında kalan alan değeri 0,832 (0,759–0,899) olarak bulundu. En iyi performans gösteren radyomik tabanlı makine öğrenmesi modelinde bu değer ortalama 0,89 ± 0,10 idi. Kombine model ise 0,96 ± 0,01 değeri ile en yüksek ayırt edici performansı göstermiştir.Sonuç: Radyomik tabanlı makine öğrenmesinin semantik MRG ölçümleriyle birleştirilmesi, şeffaf hücreli renal hücreli karsinomun diğer alt tiplerden ayrımında tekil yaklaşımlara kıyasla daha yüksek tanısal performans göstermiştir. Bu bulguların klinik uygulamaya geçebilmesi için daha geniş ve çok merkezli çalışmalarla doğrulanması gerekmektedir.
CR  - May AM, Guduru A, Fernelius J, Raza SJ, et al. Current Trends in Partial Nephrectomy After Guideline Release: Health Disparity for Small Renal Mass. Kidney Cancer 2019;3:183–88.
CR  - Remzi M, Özsoy M, Klingler H-C, et al. Are Small Renal Tumors Harmless? Analysis of Histopathological Features According to Tumors 4 Cm or Less in Diameter. J Urol 2006;176(3):896-99.
CR  - Finelli A, Cheung DC, Al-Matar A, et al. Small Renal Mass Surveillance: Histology-specific Growth Rates in a Biopsy-characterized Cohort. Eur Urol. 2020;78:460–67.
CR  - Pedrosa I. Invited Commentary: MRI Clear Cell Likelihood Score for Indeterminate Solid Renal Masses: Is There a Path for Broad Clinical Adoption? RadioGraphics 2023;43(7):e230042.
CR  - Zhong J, Hu Y, Xing Y, et al. Is there enough evidence supporting the clinical adoption of clear cell likelihood score (ccLS)? An updated systematic review and meta-analysis. Insights Imaging 2024;15(1):242.
CR  - Shetty AS, Fraum TJ, Ballard DH, et al. Renal Mass Imaging with MRI Clear Cell Likelihood Score: A User’s Guide. RadioGraphics 2023;43(7):e220209.
CR  - Suarez-Ibarrola R, Hein S, Reis G, Gratzke C, Miernik A. Current and future applications of machine and deep learning in urology: a review of the literature on urolithiasis, renal cell carcinoma, and bladder and prostate cancer. World J Urol 2020;38:2329–47.
CR  - Matsumoto S, Arita Y, Yoshida S, et al. Utility of radiomics features of diffusion-weighted magnetic resonance imaging for differentiation of fat-poor angiomyolipoma from clear cell renal cell carcinoma: model development and external validation. Abdom Radiol (NY) 2022;47:2178–86.
CR  - Massa’a RN, Stoeckl EM, Lubner MG, et al. Differentiation of benign from malignant solid renal lesions with MRI-based radiomics and machine learning. Abdom Radiol (NY) 2022;47:2896–2904.
CR  - Fedorov A, Beichel R, Kalpathy-Cramer J, et al. 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn Reson Imaging 2012;30:1323–41.
CR  - Van Griethuysen JJM, Fedorov A, Parmar C, et al. Computational radiomics system to decode the radiographic phenotype. Cancer Res 2017;77:e104–e7.
CR  - Ferro M, Crocetto F, Barone B, et al. Artificial intelligence and radiomics in evaluation of kidney lesions: a comprehensive literature review. Ther Adv Urol 2023;15:17562872231164803.
CR  - Razik A, Goyal A, Sharma R, et al. MR texture analysis in differentiating renal cell carcinoma from lipid-poor angiomyolipoma and oncocytoma. Br J Radiol 2020;93(1114):20200569.
CR  - Cakir IM, Eryuruk U, Gurun E, et al. Improving the diagnostic accuracy of small renal masses: Integration of radiomics and clear cell likelihood scores in multiparametric MRI. Eur J Radiol 2025;189:112174.
CR  - Hastie T, Tibshirani R, Friedman J. Support Vector Machines and Flexible Discriminants. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd ed. New York, NY: Springer; 2009:417-58.
UR  - https://doi.org/10.19161/etd.1882022
L1  - https://dergipark.org.tr/en/download/article-file/5685351
ER  -