Predictive models composed by radiomic features extracted from multi-detector computed tomography images for predicting low- and high- grade clear cell renal cell carcinoma: A STARD-compliant article

A contrast-enhanced computed tomography-based radiomics nomogram for preoperative differentiation between benign and malignant cystic renal lesions

Background

There is lack of discrimination as to traditional imaging diagnostic methods of cystic renal lesions (CRLs). This study aimed to evaluate the value of machine learning models based on clinical data and contrast-enhanced computed tomography (CECT) radiomics features in the differential diagnosis of benign and malignant CRL.

Methods

There were 192 patients with CRL (Bosniak class ≥ II) enrolled through histopathological examination, including 144 benign cystic renal lesions (BCRLs) and 48 malignant cystic renal lesions (MCRLs). Radiomics features were extracted from CECT images taken during the medullary phase. Using the light gradient boosting machine (LightGBM) algorithm, the clinical, radiomics and combined models were constructed. A comprehensive nomogram was developed by integrating the radiomics score (Rad-score) with independent clinical factors. Receiver operating characteristic (ROC) curves were plotted. The corresponding area under the curve (AUC) value was worked out to quantify the discrimination performance of the three models in training and validation cohorts. Calibration curves were worked out to assess the accuracy of the probability values predicted by the models. Decision curve analysis (DCA) was worked out to assess the performance of models at different thresholds.

Results

Maximum diameter and Bosniak class were independent risk factors of patients with MCRL in the clinical model. Twenty-one radiomics features were extracted to work out a Rad-score. The performance of the clinical model in the training cohort was AUC =0.948, 95% confidence interval (CI): 0.917-0.980, and the performance in the validation cohort was AUC =0.936, 95% CI: 0.859-1.000 (P<0.05). The performance of the radiomics model in the training cohort was AUC =0.990, 95% CI: 0.979-1.000, and the performance in the validation cohort was AUC =0.959, 95% CI: 0.903-1.000 (P<0.05). Compared with the above models, the combined radiomics nomogram had an AUC of 0.989 (95% CI: 0.977-1.000) in the training cohort and an AUC of 0.962 (95% CI: 0.905-1.000) in the validation cohort (P<0.05), showing the best diagnostic efficacy.

Conclusions

The radiomics nomogram integrating clinical independent risk factors and radiomics signature improved the diagnostic accuracy in differentiating between BCRL and MCRL, which can provide a reference for clinical decision-making and help clinicians develop individualized treatment strategies for patients.

CT-based radiomics model using stability selection for predicting the World Health Organization/International Society of Urological Pathology grade of clear cell renal cell carcinoma

OBJECTIVES

This study aimed to develop a model to predict World Health Organization/International Society of Urological Pathology (WHO/ISUP) low-grade or high-grade clear cell renal cell carcinoma (ccRCC) using 3D multiphase enhanced CT radiomics features (RFs).

METHODS

CT data of 138 low-grade and 60 high-grade ccRCC cases were included. RFs were extracted from four CT phases: non-contrast phase (NCP), corticomedullary phase, nephrographic phase, and excretory phase (EP). Models were developed using various combinations of RFs and subjected to cross-validation.

RESULTS

There were 107 RFs extracted from each phase of the CT images. The NCP-EP model had the best overall predictive value (AUC = 0.78), but did not significantly differ from that of the NCP model (AUC = 0.76). By considering the predictive ability of the model, the level of radiation exposure, and model simplicity, the overall best model was the Conventional image and clinical features (CICFs)-NCP model (AUC = 0.77; sensitivity 0.75, specificity 0.69, positive predictive value 0.85, negative predictive value 0.54, accuracy 0.73). The second-best model was the NCP model (AUC = 0.76).

CONCLUSIONS

Combining clinical features with unenhanced CT images of the kidneys seems to be optimal for prediction of WHO/ISUP grade of ccRCC. This noninvasive method may assist in guiding more accurate treatment decisions for ccRCC.

ADVANCES IN KNOWLEDGE

This study innovatively employed stability selection for RFs, enhancing model reliability. The CICFs-NCP model's simplicity and efficacy mark a significant advancement, offering a practical tool for clinical decision-making in ccRCC management.

Machine learning-based CT radiomics approach for predicting WHO/ISUP nuclear grade of clear cell renal cell carcinoma: an exploratory and comparative study

Purpose

To investigate the predictive performance of machine learning-based CT radiomics for differentiating between low- and high-nuclear grade of clear cell renal cell carcinomas (CCRCCs).

Methods

This retrospective study enrolled 406 patients with pathologically confirmed low- and high-nuclear grade of CCRCCs according to the WHO/ISUP grading system, which were divided into the training and testing cohorts. Radiomics features were extracted from nephrographic-phase CT images using PyRadiomics. A support vector machine (SVM) combined with three feature selection algorithms such as least absolute shrinkage and selection operator (LASSO), recursive feature elimination (RFE), and ReliefF was performed to determine the most suitable classification model, respectively. Clinicoradiological, radiomics, and combined models were constructed using the radiological and clinical characteristics with significant differences between the groups, selected radiomics features, and a combination of both, respectively. Model performance was evaluated by receiver operating characteristic (ROC) curve, calibration curve, and decision curve analyses.

Results

SVM-ReliefF algorithm outperformed SVM-LASSO and SVM-RFE in distinguishing low- from high-grade CCRCCs. The combined model showed better prediction performance than the clinicoradiological and radiomics models (p < 0.05, DeLong test), which achieved the highest efficacy, with an area under the ROC curve (AUC) value of 0.887 (95% confidence interval [CI] 0.798–0.952), 0.859 (95% CI 0.748–0.935), and 0.828 (95% CI 0.731–0.929) in the training, validation, and testing cohorts, respectively. The calibration and decision curves also indicated the favorable performance of the combined model.

Conclusion

A combined model incorporating the radiomics features and clinicoradiological characteristics can better predict the WHO/ISUP nuclear grade of CCRCC preoperatively, thus providing effective and noninvasive assessment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13244-021-01107-1.

CT-based radiomics for differentiating renal tumours: a systematic review

PURPOSE

Differentiating renal tumours into grades and tumour subtype from medical imaging is important for patient management; however, there is an element of subjectivity when performed qualitatively. Quantitative analysis such as radiomics may provide a more objective approach. The purpose of this article is to systematically review the literature on computed tomography (CT) radiomics for grading and differentiating renal tumour subtypes. An educational perspective will also be provided.

METHODS

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist was followed. PubMed, Scopus and Web of Science were searched for relevant articles. The quality of each study was assessed using the Radiomic Quality Score (RQS).

RESULTS

13 studies were found. The main outcomes were prediction of pathological grade and differentiating between renal tumour types, measured as area under the curve (AUC) for either the receiver operator curve or precision recall curve. Features extracted to predict pathological grade or tumour subtype included shape, intensity, texture and wavelet (a type of higher order feature). Four studies differentiated between low-grade and high-grade clear cell renal cell cancer (RCC) with good performance (AUC = 0.82-0.978). One other study differentiated low- and high-grade chromophobe with AUC = 0.84. Finally, eight studies used radiomics to differentiate between tumour types such as clear cell RCC, fat-poor angiomyolipoma, papillary RCC, chromophobe RCC and renal oncocytoma with high levels of performance (AUC 0.82-0.96).

CONCLUSION

Renal tumours can be pathologically classified using CT-based radiomics with good performance. The main radiomic feature used for tumour differentiation was texture. Fuhrman was the most common pathologic grading system used in the reviewed studies. Renal tumour grading studies should be extended beyond clear cell RCC and chromophobe RCC. Further research with larger prospective studies, performed in the clinical setting, across multiple institutions would help with clinical translation to the radiologist's workstation.

MDCT-Based Radiomics Features for the Differentiation of Serous Borderline Ovarian Tumors and Serous Malignant Ovarian Tumors

Objective

To investigate whether multidetector computed tomography (MDCT)-based radiomics features can discriminate between serous borderline ovarian tumors (SBOTs) and serous malignant ovarian tumors (SMOTs).

Patients and Methods

Eighty patients with SBOTs and 102 patients with SMOTs, confirmed by pathology (training set: n = 127; validation set: n = 55) from December 2017 to June 2020, were enrolled in this study. The interclass correlation coefficient (ICC) and least absolute shrinkage and selection operator (LASSO) regression were applied to select radiomics parameters derived from MDCT images on the arterial phase (AP), venous phase (VP), and equilibrium phase (EP). Receiver operating characteristic (ROC) analysis of each selected parameter was carried out. Heat maps were created to illustrate the distribution of the radiomics parameters. Three models incorporating selected radiomics parameters generated by support vector machine (SVM) classifiers in each phase were analyzed by ROC and compared using the DeLong test.

Results

The most predictive features selected by ICC and LASSO regression between SBOTs and SMOTs included 9 radiomics parameters on AP, VP, and EP each. Three models on AP, VP, and EP incorporating the selected features generated by SVM classifiers produced AUCs of 0.80 (accuracy, 0.75; sensitivity, 0.74; specificity, 0.75), 0.86 (accuracy, 0.78; sensitivity, 0.80; specificity, 0.75), and 0.73 (accuracy, 0.69; sensitivity, 0.71; specificity, 0.67), respectively. There were no significant differences in the AUCs among the three models (AP vs. VP, P = 0.199; AP vs. EP, P = 0.260; VP vs. EP, P = 0.793).

Conclusion

MDCT-based radiomics features could be used as biomarkers for the differentiation of SBOTs and SMOTs.

Preoperative Predicting the WHO/ISUP Nuclear Grade of Clear Cell Renal Cell Carcinoma by Computed Tomography-Based Radiomics Features

Nuclear grade is important for treatment selection and prognosis in patients with clear cell renal cell carcinoma (ccRCC). This study aimed to determine the ability of preoperative four-phase multiphasic multidetector computed tomography (MDCT)-based radiomics features to predict the WHO/ISUP nuclear grade. In all 102 patients with histologically confirmed ccRCC, the training set (n = 62) and validation set (n = 40) were randomly assigned. In both datasets, patients were categorized according to the WHO/ISUP grading system into low-grade ccRCC (grades 1 and 2) and high-grade ccRCC (grades 3 and 4). The feature selection process consisted of three steps, including least absolute shrinkage and selection operator (LASSO) regression analysis, and the radiomics scores were developed using 48 radiomics features (10 in the unenhanced phase, 17 in the corticomedullary (CM) phase, 14 in the nephrographic (NP) phase, and 7 in the excretory phase). The radiomics score (Rad-Score) derived from the CM phase achieved the best predictive ability, with a sensitivity, specificity, and an area under the curve (AUC) of 90.91%, 95.00%, and 0.97 in the training set. In the validation set, the Rad-Score derived from the NP phase achieved the best predictive ability, with a sensitivity, specificity, and an AUC of 72.73%, 85.30%, and 0.84. We constructed a complex model, adding the radiomics score for each of the phases to the clinicoradiological characteristics, and found significantly better performance in the discrimination of the nuclear grades of ccRCCs in all MDCT phases. The highest AUC of 0.99 (95% CI, 0.92-1.00, p < 0.0001) was demonstrated for the CM phase. Our results showed that the MDCT radiomics features may play a role as potential imaging biomarkers to preoperatively predict the WHO/ISUP grade of ccRCCs.

Shape and texture-based radiomics signature on CT effectively discriminates benign from malignant renal masses

OBJECTIVES

Using a radiomics framework to quantitatively analyze tumor shape and texture features in three dimensions, we tested its ability to objectively and robustly distinguish between benign and malignant renal masses. We assessed the relative contributions of shape and texture metrics separately and together in the prediction model.

MATERIALS AND METHODS

Computed tomography (CT) images of 735 patients with 539 malignant and 196 benign masses were segmented in this retrospective study. Thirty-three shape and 760 texture metrics were calculated per tumor. Tumor classification models using shape, texture, and both metrics were built using random forest and AdaBoost with tenfold cross-validation. Sensitivity analyses on five sub-cohorts with respect to the acquisition phase were conducted. Additional sensitivity analyses after multiple imputation were also conducted. Model performance was assessed using AUC.

RESULTS

Random forest classifier showed shape metrics featuring within the top 10% performing metrics regardless of phase, attaining the highest variable importance in the corticomedullary phase. Convex hull perimeter ratio is a consistently high-performing shape feature. Shape metrics alone achieved an AUC ranging 0.64-0.68 across multiple classifiers, compared with 0.67-0.75 and 0.68-0.75 achieved by texture-only and combined models, respectively.

CONCLUSION

Shape metrics alone attain high prediction performance and high variable importance in the combined model, while being independent of the acquisition phase (unlike texture). Shape analysis therefore should not be overlooked in its potential to distinguish benign from malignant tumors, and future radiomics platforms powered by machine learning should harness both shape and texture metrics.

KEY POINTS

• Current radiomics research is heavily weighted towards texture analysis, but quantitative shape metrics should not be ignored in their potential to distinguish benign from malignant renal tumors. • Shape metrics alone can attain high prediction performance and demonstrate high variable importance in the combined shape and texture radiomics model. • Any future radiomics platform powered by machine learning should harness both shape and texture metrics, especially since tumor shape (unlike texture) is independent of the acquisition phase and more robust from the imaging variations.

Prediction models for clear cell renal cell carcinoma ISUP/WHO grade: comparison between CT radiomics and conventional contrast-enhanced CT

OBJECTIVE

Comparing the prediction models for the ISUP/WHO grade of clear cell renal cell carcinoma (ccRCC) based on CT radiomics and conventional contrast-enhanced CT (CECT).

METHODS

The corticomedullary phase images of 119 cases of low-grade (I and II) and high-grade (III and IV) ccRCC based on 2016 ISUP/WHO pathological grading criteria were analyzed retrospectively. The patients were randomly divided into training and validation set by stratified sampling according to 7:3 ratio. Prediction models of ccRCC differentiation were constructed using CT radiomics and conventional CECT findings in the training setandwere validated using validation set. The discrimination, calibration, net reclassification index (NRI) and integrated discrimination improvement index (IDI) of the two prediction models were further compared. The decision curve was used to analyze the net benefit of patients under different probability thresholds of the two models.

RESULTS

In the training set, the C-statistics of radiomics prediction model was statistically higher than that of CECT (p < 0.05), with NRI of 9.52% and IDI of 21.6%, both with statistical significance (p < 0.01).In the validation set, the C-statistics of radiomics prediction model was also higher but did not show statistical significance (p = 0.07). The NRI and IDI was 14.29 and 33.7%, respectively, both statistically significant (p < 0.01). Validation set decision curve analysis showed the net benefit improvement of CT radiomics prediction model in the range of 3-81% over CECT.

CONCLUSION

The prediction model using CT radiomics in corticomedullary phase is more effective for ccRCC ISUP/WHO grade than conventional CECT.

ADVANCES IN KNOWLEDGE

As a non-invasive analysis method, radiomics can predict the ISUP/WHO grade of ccRCC more effectively than traditional enhanced CT.

Predicting the ISUP grade of clear cell renal cell carcinoma with multiparametric MR and multiphase CT radiomics

OBJECTIVE

To investigate externally validated magnetic resonance (MR)-based and computed tomography (CT)-based machine learning (ML) models for grading clear cell renal cell carcinoma (ccRCC).

MATERIALS AND METHODS

Patients with pathologically proven ccRCC in 2009-2018 were retrospectively included for model development and internal validation; patients from another independent institution and The Cancer Imaging Archive dataset were included for external validation. Features were extracted from T1-weighted, T2-weighted, corticomedullary-phase (CMP), and nephrographic-phase (NP) MR as well as precontrast-phase (PCP), CMP, and NP CT. CatBoost was used for ML-model investigation. The reproducibility of texture features was assessed using intraclass correlation coefficient (ICC). Accuracy (ACC) was used for ML-model performance evaluation.

RESULTS

Twenty external and 440 internal cases were included. Among 368 and 276 texture features from MR and CT, 322 and 250 features with good to excellent reproducibility (ICC ≥ 0.75) were included for ML-model development. The best MR- and CT-based ML models satisfactorily distinguished high- from low-grade ccRCCs in internal (MR-ACC = 73% and CT-ACC = 79%) and external (MR-ACC = 74% and CT-ACC = 69%) validation. Compared to single-sequence or single-phase images, the classifiers based on all-sequence MR (71% to 73% in internal and 64% to 74% in external validation) and all-phase CT (77% to 79% in internal and 61% to 69% in external validation) images had significant increases in ACC.

CONCLUSIONS

MR- and CT-based ML models are valuable noninvasive techniques for discriminating high- from low-grade ccRCCs, and multiparameter MR- and multiphase CT-based classifiers are potentially superior to those based on single-sequence or single-phase imaging.

KEY POINTS

• Both the MR- and CT-based machine learning models are reliable predictors for differentiating high- from low-grade ccRCCs. • ML models based on multiparameter MR sequences and multiphase CT images potentially outperform those based on single-sequence or single-phase images in ccRCC grading.