Clear cell renal cell carcinoma: CT-based radiomics features for the prediction of Fuhrman grade

To evaluate the performance of a multi-parametric MRI (mp-MRI)-based radiomics signature for discriminating between clinically significant prostate cancer (csPCa) and insignificant PCa (ciPCa). Materials and methods: Two hundred and eighty patients with pathology-proven PCa were enrolled and were randomly divided into training and test cohorts. Eight hundred and nineteen radiomics features were extracted from mp-MRI for each patient. The minority group in the training cohort was balanced via the synthetic minority over-sampling technique (SMOTE) method. We used minimum-redundancy maximum-relevance (mRMR) selection and the LASSO algorithm for feature selection and radiomics signature building. The classification performance of the radiomics signature for csPCa and ciPCa was evaluated by receiver operating characteristic curve analysis in the training and test cohorts. Results: Nine features were selected for the radiomics signature building. Significant differences in the radiomics signature existed between the csPCa and ciPCa groups in both the training and test cohorts (p < 0.01 for both). The AUC, sensitivity and specificity of the radiomics signature were 0.872 (95% CI: 0.823−0.921), 0.883, and 0.753, respectively, in the training cohort, and 0.823 (95% CI: 0.669−0.976), 0.841, and 0.727, respectively, in the test cohort. Conclusion: Mp-MRI-based radiomics signature have the potential to noninvasively discriminate between csPCa and ciPCa.

show abstract

“…Features with non-zero coefficients were selected from the candidate features and were combined linearly to construct a radiomics signature [17].…”

Section: Discussionmentioning

confidence: 99%

Multi-parametric MRI-based radiomics signature for discriminating between clinically significant and insignificant prostate cancer: Cross-validation of a machine learning method

Min

Dong

et al. 2019

European Journal of Radiology

106

View full text Add to dashboard Cite

show abstract

“…Although currently histopathology is the gold standard, for tumor subtype and grade there is an intensive search for non-invasive imaging biomarkers which can provide prognostic information preoperatively and reduce the need for biopsy. Radiomics has generated significant interest with multiple studies finding a satisfactory diagnostic performance in grading and subtyping RCC on contrast-enhanced CT images using texture analysis (13)(14)(15)(16)(17)(18)(19)(20).…”

Section: Discussionmentioning

confidence: 99%

“…Several studies have used TA to subtype and grade RCC on contrast-enhanced CT images (13)(14)(15)(16)(17)(18)(19)(20). Many authors have also used machine learning algorithms in interpreting and validating the data in order to generate classifiers which could enhance the findings of individual metrics and save time in the process.…”

Section: Introductionmentioning

confidence: 99%

Role of MR texture analysis in histological subtyping and grading of renal cell carcinoma: a preliminary study

et al. 2019

View full text Add to dashboard Cite

The study evaluated the usefulness of magnetic resonance imaging (MRI) texture parameters in differentiating clear cell renal carcinoma (CC-RCC) from non-clear cell carcinoma (NC-RCC) and in the histological grading of CC-RCC. Material and Methods: After institutional ethical approval, this retrospective study analyzed 33 patients with 34 RCC masses (29 CC-RCC and five NC-RCC; 19 low grade and 10 high grade CC-RCC), who underwent MRI between January 2011 and December 2012 on a 1.5-T scanner (Avanto, Siemens, Erlangen, Germany). The MRI protocol included T2-weighted imaging (T2WI), diffusion-weighted imaging [DWI; at b 0, 500 and 1000 s/mm 2 with apparent diffusion coefficient (ADC) maps] and T1-weighted pre and postcontrast [corticomedullary (CM) and nephrographic (NG) phase] acquisition. MR texture analysis (MRTA) was performed using the TexRAD research software (Feedback Medical Ltd., Cambridge, UK) by a single reader who placed free-hand polygonal region of interest (ROI) on the slice showing the maximum viable tumor. Filtration histogrambased texture analysis was used to generate six first-order statistical parameters [mean intensity, standard deviation (SD), mean of positive pixels (MPP), entropy, skewness and kurtosis] at five spatial scaling factors (SSF) as well as on the unfiltered image. Mann-Whitney test was used to compare the texture parameters of CC-RCC vs. NC-RCC, and high grade vs. low grade CC-RCC. P-value < 0.05 was considered significant. A 3-step feature selection was used to obtain the best texture metrics for each MRI sequence and included the receiver operating characteristic (ROC) curve analysis and Pearson's correlation test. Results: The best performing texture parameters in differentiating CC-RCC from NC-RCC for each sequence included (area under the curve in parentheses): entropy at SSF 4 (0.807) on T2WI, SD at SSF 4 (0.814) on DWI b500, SD at SSF 6 (0.879) on DWI b1000, mean at SSF 0 (0.848) on ADC, skewness at SSF 2 (0.854) on T1WI and skewness at SSF 3 (0.908) on CM phase. In differentiating high from low grade CC-RCC, the best parameters were: entropy at SSF 6 (0.823) on DWI b1000, mean at SSF 3 (0.889) on CM phase and MPP at SSF 5 (0.870) on NG phase. Conclusion: Several MR texture parameters showed excellent diagnostic performance (AUC> 0.8) in differentiating CC-RCC from NC-RCC, and high grade from low grade CC-RCC. MRTA could serve as a useful non-invasive tool for this purpose.

show abstract

“…Finally, the least absolute shrinkage and selection operator (LASSO) feature selection algorithm was used to select the relevant features based on the optimal λ parameters, and the coefficients were calculated for each feature; then, radiomic features with non-zero coefficients were obtained. The LASSO algorithm can be used to reduce the dimensions of features and select the most meaningful features effectively [13,14]. Further, using the T test on the optimum features between chRCC and RO patients, a probability value (p value) is calculated.…”

Section: Feature Extraction and Selectionmentioning

confidence: 99%

Value of radiomics in differential diagnosis of chromophobe renal cell carcinoma and renal oncocytoma

Huang

Xia³

2019

Abdom Radiol

View full text Add to dashboard Cite

Purpose To explore the value of CT-enhanced quantitative features combined with machine learning for differential diagnosis of renal chromophobe cell carcinoma (chRCC) and renal oncocytoma (RO). Methods Sixty-one cases of renal tumors (chRCC = 44; RO = 17) that were pathologically confirmed at our hospital between 2008 and 2018 were retrospectively analyzed. All patients had undergone preoperative enhanced CT scans including the corticomedullary (CMP), nephrographic (NP), and excretory phases (EP) of contrast enhancement. Volumes of interest (VOIs), including lesions on the images, were manually delineated using the RadCloud platform. A LASSO regression algorithm was used to screen the image features extracted from all VOIs. Five machine learning classifications were trained to distinguish chRCC from RO by using a fivefold cross-validation strategy. The performance of the classifier was mainly evaluated by areas under the receiver operating characteristic (ROC) curve and accuracy. Results In total, 1029 features were extracted from CMP, NP, and EP. The LASSO regression algorithm was used to screen out the four, four, and six best features, respectively, and eight features were selected when CMP and NP were combined. All five classifiers had good diagnostic performance, with area under the curve (AUC) values greater than 0.850, and support vector machine (SVM) classifier showed a diagnostic accuracy of 0.945 (AUC 0.964 ± 0.054; sensitivity 0.999; specificity 0.800), showing the best performance. Conclusions Accurate preoperative differential diagnosis of chRCC and RO can be facilitated by a combination of CTenhanced quantitative features and machine learning.

show abstract

Clear cell renal cell carcinoma: CT-based radiomics features for the prediction of Fuhrman grade

Cited by 109 publications

References 33 publications

Multi-parametric MRI-based radiomics signature for discriminating between clinically significant and insignificant prostate cancer: Cross-validation of a machine learning method

Multi-parametric MRI-based radiomics signature for discriminating between clinically significant and insignificant prostate cancer: Cross-validation of a machine learning method

Role of MR texture analysis in histological subtyping and grading of renal cell carcinoma: a preliminary study

Value of radiomics in differential diagnosis of chromophobe renal cell carcinoma and renal oncocytoma

Contact Info

Product

Resources

About