Clear Cell Renal Cell Carcinoma: Machine Learning-Based Quantitative Computed Tomography Texture Analysis for Prediction of Fuhrman Nuclear Grade

Incremental value of automatically segmented perirenal adipose tissue for pathological grading of clear cell renal cell carcinoma: a multicenter cohort study

OBJECTIVES

Accurate preoperative prediction of the pathological grade of clear cell renal cell carcinoma (ccRCC) is crucial for optimal treatment planning and patient outcomes. This study aims to develop and validate a deep-learning (DL) algorithm to automatically segment renal tumours, kidneys, and perirenal adipose tissue (PRAT) from computed tomography (CT) images and extract radiomics features to predict the pathological grade of ccRCC.

METHODS

In this cross-ethnic retrospective study, a total of 614 patients were divided into a training set (383 patients from the local hospital), an internal validation set (88 patients from the local hospital), and an external validation set (143 patients from the public dataset). A two-dimensional TransUNet-based DL model combined with the train-while-annotation method was trained for automatic volumetric segmentation of renal tumours, kidneys, and visceral adipose tissue (VAT) on images from two groups of datasets. PRAT was extracted using a dilation algorithm by calculating voxels of VAT surrounding the kidneys. Radiomics features were subsequently extracted from three regions of interest of CT images, adopting multiple filtering strategies. The least absolute shrinkage and selection operator (LASSO) regression was used for feature selection, and the support vector machine (SVM) for developing the pathological grading model. Ensemble learning was used for imbalanced data classification. Performance evaluation included the Dice coefficient for segmentation and metrics such as accuracy and area under curve (AUC) for classification. The WHO/International Society of Urological Pathology (ISUP) grading models were finally interpreted and visualized using the SHapley Additive exPlanations (SHAP) method.

RESULTS

For automatic segmentation, the mean Dice coefficient achieved 0.836 for renal tumours and 0.967 for VAT on the internal validation dataset. For WHO/ISUP grading, a model built with features of PRAT achieved a moderate AUC of 0.711 (95% CI, 0.604-0.802) in the internal validation set, coupled with a sensitivity of 0.400 and a specificity of 0.781. While model built with combination features of the renal tumour, kidney, and PRAT showed an AUC of 0.814 (95% CI, 0.717-0.889) in the internal validation set, with a sensitivity of 0.800 and a specificity of 0.753, significantly higher than the model built with features solely from tumour lesion (0.760; 95% CI, 0.657-0.845), with a sensitivity of 0.533 and a specificity of 0.767.

CONCLUSION

Automated segmentation of kidneys and visceral adipose tissue (VAT) through TransUNet combined with a conventional image morphology processing algorithm offers a standardized approach to extract PRAT with high reproducibility. The radiomics features of PRAT and tumour lesions, along with machine learning, accurately predict the pathological grade of ccRCC and reveal the incremental significance of PRAT in this prediction.

Kidney medicine meets computer vision: a bibliometric analysis

BACKGROUND AND OBJECTIVE

Rapid advances in computer vision (CV) have the potential to facilitate the examination, diagnosis, and treatment of diseases of the kidney. The bibliometric study aims to explore the research landscape and evolving research focus of the application of CV in kidney medicine research.

METHODS

The Web of Science Core Collection was utilized to identify publications related to the research or applications of CV technology in the field of kidney medicine from January 1, 1900, to December 31, 2022. We analyzed emerging research trends, highly influential publications and journals, prolific researchers, countries/regions, research institutions, co-authorship networks, and co-occurrence networks. Bibliographic information was analyzed and visualized using Python, Matplotlib, Seaborn, HistCite, and Vosviewer.

RESULTS

There was an increasing trend in the number of publications on CV-based kidney medicine research. These publications mainly focused on medical image processing, surgical procedures, medical image analysis/diagnosis, as well as the application and innovation of CV technology in medical imaging. The United States is currently the leading country in terms of the quantities of published articles and international collaborations, followed by China. Deep learning-based segmentation and machine learning-based texture analysis are the most commonly used techniques in this field. Regarding research hotspot trends, CV algorithms are shifting toward artificial intelligence, and research objects are expanding to encompass a wider range of kidney-related objects, with data dimensions used in research transitioning from 2D to 3D while simultaneously incorporating more diverse data modalities.

CONCLUSION

The present study provides a scientometric overview of the current progress in the research and application of CV technology in kidney medicine research. Through the use of bibliometric analysis and network visualization, we elucidate emerging trends, key sources, leading institutions, and popular topics. Our findings and analysis are expected to provide valuable insights for future research on the use of CV in kidney medicine research.

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.

Multiphase comparative study for WHO/ISUP nuclear grading diagnostic model based on enhanced CT images of clear cell renal cell carcinoma

To compare and analyze the diagnostic value of different enhancement stages in distinguishing low and high nuclear grade clear cell renal cell carcinoma (ccRCC) based on enhanced computed tomography (CT) images by building machine learning classifiers. A total of 51 patients (Dateset1, including 41 low-grade and 10 high-grade) and 27 patients (Independent Dateset2, including 16 low-grade and 11 high-grade) with pathologically proven ccRCC were enrolled in this retrospective study. Radiomic features were extracted from the corticomedullary phase (CMP), nephrographic phase (NP), and excretory phase (EP) CT images, and selected using the recursive feature elimination cross-validation (RFECV) algorithm, the group differences were assessed using T-test and Mann-Whitney U test for continuous variables. The support vector machine (SVM), random forest (RF), XGBoost (XGB), VGG11, ResNet18, and GoogLeNet classifiers are established to distinguish low-grade and high-grade ccRCC. The classifiers based on CT images of NP (Dateset1, RF: AUC = 0.82 ± 0.05, ResNet18: AUC = 0.81 ± 0.02; Dateset2, XGB: AUC = 0.95 ± 0.02, ResNet18: AUC = 0.87 ± 0.07) obtained the best performance and robustness in distinguishing low-grade and high-grade ccRCC, while the EP-based classifier performance in poorer results. The CT images of enhanced phase NP had the best performance in diagnosing low and high nuclear grade ccRCC. Firstorder_Kurtosis and firstorder_90Percentile feature play a vital role in the classification task.

Study of radiomics based on dual-energy CT for nuclear grading and T-staging in renal clear cell carcinoma

INTRODUCTION

Clear cell renal cell carcinoma (ccRCC) is the most lethal subtype of renal cell carcinoma with a high invasive potential. Radiomics has attracted much attention in predicting the preoperative T-staging and nuclear grade of ccRCC.

OBJECTIVE

The objective was to evaluate the efficacy of dual-energy computed tomography (DECT) radiomics in predicting ccRCC grade and T-stage while optimizing the models.

METHODS

200 ccRCC patients underwent preoperative DECT scanning and were randomized into training and validation cohorts. Radiomics models based on 70 KeV, 100 KeV, 150 KeV, iodine-based material decomposition images (IMDI), virtual noncontrasted images (VNC), mixed energy images (MEI) and MEI + IMDI were established for grading and T-staging. Receiver operating characteristic analysis and decision curve analysis (DCA) were performed. The area under the curve (AUC) values were compared using Delong test.

RESULTS

For grading, the AUC values of these models ranged from 0.64 to 0.97 during training and from 0.54 to 0.72 during validation. In the validation cohort, the performance of MEI + IMDI model was optimal, with an AUC of 0.72, sensitivity of 0.71, and specificity of 0.70. The AUC value for the 70 KeV model was higher than those for the 100 KeV, 150 KeV, and MEI models. For T-staging, these models achieved AUC values of 0.83 to 1.00 in training and 0.59 to 0.82 in validation. The validation cohort demonstrated AUCs of 0.82 and 0.70, sensitivities of 0.71 and 0.71, and specificities of 0.80 and 0.60 for the MEI + IMDI and IMDI models, respectively. In terms of grading and T-staging, the MEI + IMDI model had the highest AUC in validation, with IMDI coming in second. There were statistically significant differences between the MEI + IMDI model and the 70 KeV, 100 KeV, 150 KeV, MEI, and VNC models in terms of grading (P < .05) and staging (P ≤ .001). DCA showed that both MEI + IDMI and IDMI models outperformed other models in predicting grade and stage of ccRCC.

CONCLUSIONS

DECT radiomics models were helpful in grading and T-staging of ccRCC. The combined model of MEI + IMDI achieved favorable results.

Machine learning-based CT texture analysis in the differentiation of testicular masses

Purpose

To evaluate the ability of texture features for distinguishing between benign and malignant testicular masses, and furthermore, for identifying primary testicular lymphoma in malignant tumors and identifying seminoma in testicular germ cell tumors, respectively.

Methods

We retrospectively collected 77 patients with an abdominal and pelvic enhanced computed tomography (CT) examination and a histopathologically confirmed testicular mass from a single center. The ROI of each mass was split into two parts by the largest cross-sectional slice and deemed to be two samples. After all processing steps, three-dimensional texture features were extracted from unenhanced and contrast-enhanced CT images. Excellent reproducibility of texture features was defined as intra-class correlation coefficient ≥0.8 (ICC ≥0.8). All the groups were balanced via the synthetic minority over-sampling technique (SMOTE) method. Dimension reduction was based on pearson correlation coefficient (PCC). Before model building, minimum-redundancy maximum-relevance (mRMR) selection and recursive feature elimination (RFE) were used for further feature selection. At last, three ML classifiers with the highest cross validation with 5-fold were selected: autoencoder (AE), support vector machine(SVM), linear discriminant analysis (LAD). Logistics regression (LR) and LR-LASSO were also constructed to compare with the ML classifiers.

Results

985 texture features with ICC ≥0.8 were extracted for further feature selection process. With the highest AUC of 0.946 (P <0.01), logistics regression was proved to be the best model for the identification of benign or malignant testicular masses. Besides, LR also had the best performance in identifying primary testicular lymphoma in malignant testicular tumors and in identifying seminoma in testicular germ cell tumors, with the AUC of 0.982 (P <0.01) and 0.928 (P <0.01), respectively.

Conclusion

Until now, this is the first study that applied CT texture analysis (CTTA) to assess the heterogeneity of testicular tumors. LR model based on CTTA might be a promising non-invasive tool for the diagnosis and differentiation of testicular masses. The accurate diagnosis of testicular masses would assist urologists in correct preoperative and perioperative decision making.

A computed tomography urography-based machine learning model for predicting preoperative pathological grade of upper urinary tract urothelial carcinoma

OBJECTIVES

Development and validation of a computed tomography urography (CTU)-based machine learning (ML) model for prediction of preoperative pathology grade of upper urinary tract urothelial carcinoma (UTUC).

METHODS

A total of 140 patients with UTUC who underwent CTU examination from January 2017 to August 2023 were retrospectively enrolled. Tumor lesions on the unenhanced, medullary, and excretory periods of CTU were used to extract Features, respectively. Feature selection was screened by the Pearson and Spearman correlation analysis, least absolute shrinkage and selection operator algorithm, random forest (RF), support vector machine (SVM), and eXtreme Gradient Boosting (XGBoost). The logistic regression (LR) was used to screen for independent influencing factors of clinical baseline characteristics. Machine learning models based on different feature datasets were constructed and validated using algorithms such as LR, RF, SVM, and XGBoost. By computing the selected features, a radiomics score was generated, and a diverse feature dataset was constructed. Based on the training set, 16 ML models were created, and their performance was evaluated using the validation set for metrics including sensitivity, specificity, accuracy, area under the receiver operating characteristic curve (AUC), and others.

RESULTS

The training set consisted of 98 patients (mean age: 64.5 ± 10.5 years; 30 males), whereas the validation set consisted of 42 patients (mean age: 65.3 ± 9.78 years; 17 males). Hydronephrosis was the best independent influence factor (p < 0.05). The RF model had the best performance in predicting high-grade UTUC, with AUC of 0.914 (95% Confidence Interval [95%CI] 0.852-0.977) and 0.903 (95%CI 0.809-0.997) in the training set and validation set, and accuracy of 0.878 and 0.857, respectively.

CONCLUSIONS

An ML model based on the RF algorithm exhibits excellent predictive performance, offering a non-invasive approach for predicting preoperative high-grade UTUC.

Five machine learning-based radiomics models for preoperative prediction of histological grade in hepatocellular carcinoma

PURPOSE

To compare the efficacy of radiomics models via five machine learning algorithms in predicting the histological grade of hepatocellular carcinoma (HCC) before surgery and to develop the most stable model to classify high-risk HCC patients.

METHODS

Contrast-enhanced computed tomography (CECT) images of 175 HCC patients before surgery were analysed, and radiomics features were extracted from CECT images (including arterial and portal phases). Five machine learning models, including Bayes, random forest (RF), k-nearest neighbors (KNN), logistic regression (LR), and support vector machine (SVM), were applied to establish the model. The stability of the five models was weighed by the relative standard deviation (RSD), and the lowest RSD value was chosen as the most stable model to predict the histological grade of HCC. The area under the curve (AUC) and Delong tests were devoted to assessing the predictive efficacy of the models.

RESULTS

High-grade HCC accounted for 28.57% (50/175) of the 175 patients. The RSD value of AUC via the RF machine learning model was the lowest (2.3%), followed by Bayes (3.2%), KNN (6.4%), SVM (8.7%) and LR (31.3%). In addition, the RF model (AUC = 0.995) was better than the other four models in the training set (p < 0.05), as well as obtained good predictive performance in the test set (AUC = 0.837).

CONCLUSION

Among the five machine learning models, the RF-based radiomics model was the most stable and performed excellently in identifying high histological grade of HCC.

Machine Learning-Based Magnetic Resonance Radiomics Analysis for Predicting Low- and High-Grade Clear Cell Renal Cell Carcinoma

PURPOSE

To explore whether high- and low-grade clear cell renal cell carcinomas (ccRCC) can be distinguished using radiomics features extracted from magnetic resonance imaging.

METHODS

In this retrospective study, 154 patients with pathologically proven clear ccRCC underwent contrast-enhanced 3 T magnetic resonance imaging and were assigned to the development (n = 122) and test (n = 32) cohorts in a temporal-split setup. A total of 834 radiomics features were extracted from whole-tumor volumes using 3 sequences: T2-weighted imaging (T2WI), diffusion-weighted imaging, and contrast-enhanced T1-weighted imaging. A random forest regressor was used to extract important radiomics features that were subsequently used for model development using the random forest algorithm. Tumor size, apparent diffusion coefficient value, and percentage of tumor-to-renal parenchymal signal intensity drop in the tumors were recorded by 2 radiologists for quantitative analysis. The area under the receiver operating characteristic curve (AUC) was generated to predict ccRCC grade.

RESULTS

In the development cohort, the T2WI-based radiomics model demonstrated the highest performance (AUC, 0.82). The T2WI-based radiomics and radiologic feature hybrid model showed AUCs of 0.79 and 0.83, respectively. In the test cohort, the T2WI-based radiomics model achieved an AUC of 0.82. The range of AUCs of the hybrid model of T2WI-based radiomics and radiologic features was 0.73 to 0.80.

CONCLUSION

Magnetic resonance imaging-based classifier models using radiomics features and machine learning showed satisfactory diagnostic performance in distinguishing between high- and low-grade ccRCC, thereby serving as a helpful noninvasive tool for predicting ccRCC grade.

Differentiation of Small Clear Renal Cell Carcinoma and Oncocytoma through Magnetic Resonance Imaging-Based Radiomics Analysis: Toward the End of Percutaneous Biopsy

PURPOSE

The aim of this study was to ascertain whether radiomics data can assist in differentiating small (<4 cm) clear cell renal cell carcinomas (ccRCCs) from small oncocytomas using T2-weighted magnetic resonance imaging (MRI).

MATERIAL AND METHODS

This retrospective study incorporated 48 tumors, 28 of which were ccRCCs and 20 were oncocytomas. All tumors were less than 4 cm in size and had undergone pre-biopsy or pre-surgery MRI. Following image pre-processing, 102 radiomics features were evaluated. A univariate analysis was performed using the Wilcoxon rank-sum test with Bonferroni correction. We compared multiple radiomics pipelines of normalization, feature selection, and machine learning (ML) algorithms, including random forest (RF), logistic regression (LR), AdaBoost, K-nearest neighbor, and support vector machine, using a supervised ML approach.

RESULTS

No statistically significant features were identified via the univariate analysis with Bonferroni correction. The most effective algorithm was identified using a pipeline incorporating standard normalization, RF-based feature selection, and LR, which achieved an area under the curve (AUC) of 83%, accuracy of 73%, sensitivity of 79%, and specificity of 65%. Subsequently, the most significant features were identified from this algorithm, and two groups of uncorrelated features were established based on Pearson correlation scores. Using these features, an algorithm was established after a pipeline of standard normalization and LR, achieving an AUC of 90%, an accuracy of 77%, sensitivity of 83%, and specificity of 69% for distinguishing ccRCCs from oncocytomas.

CONCLUSIONS

Radiomics analysis based on T2-weighted MRI can aid in distinguishing small ccRCCs from small oncocytomas. However, it is not superior to standard multiparameter renal MRI and does not yet allow us to dispense with percutaneous biopsy.

A CT-Based Clinical, Radiological and Radiomic Machine Learning Model for Predicting Malignancy of Solid Renal Tumors (UroCCR-75)

BACKGROUND

Differentiating benign from malignant renal tumors is important for patient management, and it may be improved by quantitative CT features analysis including radiomic.

PURPOSE

This study aimed to compare performances of machine learning models using bio-clinical, conventional radiologic and 3D-radiomic features for the differentiation of benign and malignant solid renal tumors using pre-operative multiphasic contrast-enhanced CT examinations.

MATERIALS AND METHODS

A unicentric retrospective analysis of prospectively acquired data from a national kidney cancer database was conducted between January 2016 and December 2020. Histologic findings were obtained by robotic-assisted partial nephrectomy. Lesion images were semi-automatically segmented, allowing for a 3D-radiomic features extraction in the nephrographic phase. Conventional radiologic parameters such as shape, content and enhancement were combined in the analysis. Biological and clinical features were obtained from the national database. Eight machine learning (ML) models were trained and validated using a ten-fold cross-validation. Predictive performances were evaluated comparing sensitivity, specificity, accuracy and AUC.

RESULTS

A total of 122 patients with 132 renal lesions, including 111 renal cell carcinomas (RCCs) (111/132, 84%) and 21 benign tumors (21/132, 16%), were evaluated (58 +/- 14 years, men 74%). Unilaterality (100/111, 90% vs. 13/21, 62%; p = 0.02), necrosis (81/111, 73% vs. 8/21, 38%; p = 0.02), lower values of tumor/cortex ratio at portal time (0.61 vs. 0.74, p = 0.01) and higher variation of tumor/cortex ratio between arterial and portal times (0.22 vs. 0.05, p = 0.008) were associated with malignancy. A total of 35 radiomics features were selected, and "intensity mean value" was associated with RCCs in multivariate analysis (OR = 0.99). After ten-fold cross-validation, a C5.0Tree model was retained for its predictive performances, yielding a sensitivity of 95%, specificity of 42%, accuracy of 87% and AUC of 0.74.

CONCLUSION

Our machine learning-based model combining clinical, radiologic and radiomics features from multiphasic contrast-enhanced CT scans may help differentiate benign from malignant solid renal tumors.

Role of AI and Radiomic Markers in Early Diagnosis of Renal Cancer and Clinical Outcome Prediction: A Brief Review

Globally, renal cancer (RC) is the 10th most common cancer among men and women. The new era of artificial intelligence (AI) and radiomics have allowed the development of AI-based computer-aided diagnostic/prediction (AI-based CAD/CAP) systems, which have shown promise for the diagnosis of RC (i.e., subtyping, grading, and staging) and prediction of clinical outcomes at an early stage. This will absolutely help reduce diagnosis time, enhance diagnostic abilities, reduce invasiveness, and provide guidance for appropriate management procedures to avoid the burden of unresponsive treatment plans. This survey mainly has three primary aims. The first aim is to highlight the most recent technical diagnostic studies developed in the last decade, with their findings and limitations, that have taken the advantages of AI and radiomic markers derived from either computed tomography (CT) or magnetic resonance (MR) images to develop AI-based CAD systems for accurate diagnosis of renal tumors at an early stage. The second aim is to highlight the few studies that have utilized AI and radiomic markers, with their findings and limitations, to predict patients' clinical outcome/treatment response, including possible recurrence after treatment, overall survival, and progression-free survival in patients with renal tumors. The promising findings of the aforementioned studies motivated us to highlight the optimal AI-based radiomic makers that are correlated with the diagnosis of renal tumors and prediction/assessment of patients' clinical outcomes. Finally, we conclude with a discussion and possible future avenues for improving diagnostic and treatment prediction performance.

Noninvasive Prediction of Ki-67 Expression in Hepatocellular Carcinoma Using Machine Learning-Based Ultrasomics: A Multicenter Study

OBJECTIVES

To investigate the ability of ultrasomics to predict Ki-67 expression in hepatocellular carcinoma (HCC).

METHODS

A total of 244 patients from three hospitals were retrospectively recruited (training dataset, n = 168; test dataset, n = 43; and validation dataset, n = 33). Lesion segmentation of the ultrasound images was performed manually by two radiologists. In total, 1409 ultrasomics features were extracted. Feature selection was conducted using the intra-class correlation coefficient, variance threshold, mutual information, and recursive feature elimination plus eXtreme Gradient Boosting. The support vector machine was combined with the learning curve and grid search parameter tuning to construct the clinical, ultrasomics, and combined models. The predictive performance of the models was assessed using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity and accuracy.

RESULTS

The ultrasomics model performed well on the training, test, and validation datasets. The AUC (95% confidence interval [CI]) for these datasets were 0.955 (0.912-0.981), 0.861 (0.721-0.947), and 0.665 (0.480-0.819), respectively. The combination of ultrasomics and clinical features significantly improved model performance on all three datasets. The AUC (95% CI), sensitivity, specificity, and accuracy were 0.986 (0.955-0.998), 0.973, 0.840, and 0.869 on the training dataset; 0.871 (0.734-0.954), 0.750, 0.829, and 0.814 on the test dataset; and 0.742 (0.560-0.878), 0.714, 0.808, and 0.788 on the validation dataset, respectively.

CONCLUSIONS

Ultrasomics was proved to be a potential noninvasive method to predict Ki-67 expression in HCC.

Radiologist's Disease: Imaging for Renal Cancer

There is a clear benefit of imaging-based differentiation of small indeterminate masses to its subtypes of clear cell renal cell carcinoma (RCC), chromophobe RCC, papillary RCC, fat poor angiomyolipoma and oncocytoma because it helps determine the next step options for the patients. The work thus far in radiology has explored different parameters in computed tomography, MRI, and contrast-enhanced ultrasound with the discovery of many reliable imaging features that suggest certain tissue subtypes. Likert score-based risk stratification systems can help determine management, and new techniques such as perfusion, radiogenomics, single-photon emission tomography, and artificial intelligence can add to the imaging-based evaluation of indeterminate renal masses.

Computed Tomography Urography: State of the Art and Beyond

Computed Tomography Urography (CTU) is a multiphase CT examination optimized for imaging kidneys, ureters, and bladder, complemented by post-contrast excretory phase imaging. Different protocols are available for contrast administration and image acquisition and timing, with different strengths and limits, mainly related to kidney enhancement, ureters distension and opacification, and radiation exposure. The availability of new reconstruction algorithms, such as iterative and deep-learning-based reconstruction has dramatically improved the image quality and reducing radiation exposure at the same time. Dual-Energy Computed Tomography also has an important role in this type of examination, with the possibility of renal stone characterization, the availability of synthetic unenhanced phases to reduce radiation dose, and the availability of iodine maps for a better interpretation of renal masses. We also describe the new artificial intelligence applications for CTU, focusing on radiomics to predict tumor grading and patients' outcome for a personalized therapeutic approach. In this narrative review, we provide a comprehensive overview of CTU from the traditional to the newest acquisition techniques and reconstruction algorithms, and the possibility of advanced imaging interpretation to provide an up-to-date guide for radiologists who want to better comprehend this technique.

Deep-learning-based ensemble method for fully automated detection of renal masses on magnetic resonance images

Purpose

Accurate detection of small renal masses (SRM) is a fundamental step for automated classification of benign and malignant or indolent and aggressive renal tumors. Magnetic resonance image (MRI) may outperform computed tomography (CT) for SRM subtype differentiation due to improved tissue characterization, but is less explored compared to CT. The objective of this study is to autonomously detect SRM on contrast-enhanced magnetic resonance images (CE-MRI).

Approach

In this paper, we described a novel, fully automated methodology for accurate detection and localization of SRM on CE-MRI. We first determine the kidney boundaries using a U-Net convolutional neural network. We then search for SRM within the localized kidney regions using a mixture-of-experts ensemble model based on the U-Net architecture. Our dataset contained CE-MRI scans of 118 patients with different solid kidney tumor subtypes including renal cell carcinomas, oncocytomas, and fat-poor renal angiomyolipoma. We evaluated the proposed model on the entire CE-MRI dataset using 5-fold cross validation.

Results

The developed algorithm reported a Dice similarity coefficient of 91.20 ± 5.41 % (mean ± standard deviation) for kidney segmentation from 118 volumes consisting of 25,025 slices. Our proposed ensemble model for SRM detection yielded a recall and precision of 86.2% and 83.3% on the entire CE-MRI dataset, respectively.

Conclusions

We described a deep-learning-based method for fully automated SRM detection using CE-MR images, which has not been studied previously. The results are clinically important as SRM localization is a pre-step for fully automated diagnosis of SRM subtypes.

Radiogenomics in Renal Cancer Management-Current Evidence and Future Prospects

Renal cancer management is challenging from diagnosis to treatment and follow-up. In cases of small renal masses and cystic lesions the differential diagnosis of benign or malignant tissues has potential pitfalls when imaging or even renal biopsy is applied. The recent artificial intelligence, imaging techniques, and genomics advancements have the ability to help clinicians set the stratification risk, treatment selection, follow-up strategy, and prognosis of the disease. The combination of radiomics features and genomics data has achieved good results but is currently limited by the retrospective design and the small number of patients included in clinical trials. The road ahead for radiogenomics is open to new, well-designed prospective studies, with large cohorts of patients required to validate previously obtained results and enter clinical practice.

A preliminary radiomics model for predicting perirenal fat invasion on renal cell carcinoma with contrast-enhanced CT images

OBJECTIVE

The aim is to develop a radiomics model based on contrast-enhanced CT scans for preoperative prediction of perirenal fat invasion (PFI) in patients with renal cell carcinoma (RCC).

METHODS

The CT data of 131 patients with pathology-confirmed PFI status (64 positives) were retrospectively collected and randomly assigned to the training and test datasets. The kidneys and the masses were annotated by semi-automatic segmentation. Eight types of regions of interest (ROI) were chosen for the training of the radiomics models. The areas under the curves (AUCs) from the receiver operating characteristic (ROC) curve analysis were used to analyze the diagnostic performance. Eight types of models with different ROIs have been developed. The models with the highest AUC in the test dataset were used for construction of the corresponding final model, and comparison with radiologists' diagnosis.

RESULTS

The AUCs of the models for each ROI was 0.783-0.926, and there was no statistically significant difference between them (P > 0.05). Model 4 was using the ROI of the outer half-ring which extended along the edge of the mass at the outer edge of the kidney into the perirenal fat space with a thickness of 3 mm. It yielded the highest AUC (0.926) and its diagnostic accuracy was higher than the radiologists' diagnosis.

CONCLUSION

We have developed and validated a radiomics model for prediction of PFI on RCC with contrast-enhanced CT scans. The model proved to be more accurate than the radiologists' diagnosis.

Whole-Lesion CT Texture Analysis as a Quantitative Biomarker for the Identification of Homogeneous Renal Tumors

Renal tumors are very common in the urinary system, and the preoperative differential diagnosis of homogeneous renal tumors remains a challenge. This study aimed to evaluate the feasibility of the whole-lesion CT texture analysis for the identification of homogeneous renal tumors including clear cell renal cell carcinoma (ccRCC), chromophobe RCC (chRCC), and renal oncocytoma (RO). This retrospective study was approved by our local IRB. Contrast-enhanced CT examination was performed in 128 patients and histopathologically confirmed ccRCC, chRCC, and RO. The one-way ANOVA test with Bonferroni corrections was used to compare the differences, and the receiver operating characteristic (ROC) curve analysis was applied to determine the diagnostic efficiency. The whole-lesion CT histogram analysis was used to demonstrate significant differences between ccRCC and chRCC in both arterial and venous phases, and the entropy demonstrated excellent performance in discriminating these two types of tumors (AUCs = 0.95, 0.91). The inhomogeneity of ccRCC was significantly higher than that of RO both in arterial and venous phases. The entropy of chRCC was significantly lower than that of RO, and the kurtosis and entropy yielded high sensitivity (91%) and moderate specificity (74%) in the arterial phase. The whole-lesion CT histogram analysis could be useful for the differential diagnosis of homogeneous ccRCC, chRCC, and RO.

An exploratory study of CT radiomics using differential network feature selection for WHO/ISUP grading and progression-free survival prediction of clear cell renal cell carcinoma

Objectives

To explore the feasibility of predicting the World Health Organization/International Society of Urological Pathology (WHO/ISUP) grade and progression-free survival (PFS) of clear cell renal cell cancer (ccRCC) using the radiomics features (RFs) based on the differential network feature selection (FS) method using the maximum-entropy probability model (MEPM).

Methods

175 ccRCC patients were divided into a training set (125) and a test set (50). The non-contrast phase (NCP), cortico-medullary phase, nephrographic phase, excretory phase phases, and all-phase WHO/ISUP grade prediction models were constructed based on a new differential network FS method using the MEPM. The diagnostic performance of the best phase model was compared with the other state-of-the-art machine learning models and the clinical models. The RFs of the best phase model were used for survival analysis and visualized using risk scores and nomograms. The performance of the above models was tested in both cross-validated and independent validation and checked by the Hosmer-Lemeshow test.

Results

The NCP RFs model was the best phase model, with an AUC of 0.89 in the test set, and performed superior to other machine learning models and the clinical models (all p &lt;0.05). Kaplan-Meier survival analysis, univariate and multivariate cox regression results, and risk score analyses showed the NCP RFs could predict PFS well (almost all p &lt; 0.05). The nomogram model incorporated the best two RFs and showed good discrimination, a C-index of 0.71 and 0.69 in the training and test set, and good calibration.

Conclusion

The NCP CT-based RFs selected by differential network FS could predict the WHO/ISUP grade and PFS of RCC.

Radiomics analysis of contrast-enhanced CT scans can distinguish between clear cell and non-clear cell renal cell carcinoma in different imaging protocols

Introduction

This study aimed to construct a radiomics-based machine learning (ML) model for differentiation between non-clear cell and clear cell renal cell carcinomas (ccRCC) that is robust against institutional imaging protocols and scanners.

Materials and methods

Preoperative unenhanced (UN), corticomedullary (CM), and excretory (EX) phase CT scans from 209 patients diagnosed with RCCs were retrospectively collected. After the three-dimensional segmentation, 107 radiomics features (RFs) were extracted from the tumor volumes in each contrast phase. For the ML analysis, the cases were randomly split into training and test sets with a 3:1 ratio. Highly correlated RFs were filtered out based on Pearson’s correlation coefficient (r > 0.95). Intraclass correlation coefficient analysis was used to select RFs with excellent reproducibility (ICC ≥ 0.90). The most predictive RFs were selected by the least absolute shrinkage and selection operator (LASSO). A support vector machine algorithm-based binary classifier (SVC) was constructed to predict tumor types and its performance was evaluated based-on receiver operating characteristic curve (ROC) analysis. The “Kidney Tumor Segmentation 2019” (KiTS19) publicly available dataset was used during external validation of the model. The performance of the SVC was also compared with an expert radiologist’s.

Results

The training set consisted of 121 ccRCCs and 38 non-ccRCCs, while the independent internal test set contained 40 ccRCCs and 13 non-ccRCCs. For external validation, 50 ccRCCs and 23 non-ccRCCs were identified from the KiTS19 dataset with the available UN, CM, and EX phase CTs. After filtering out the highly correlated and poorly reproducible features, the LASSO algorithm selected 10 CM phase RFs that were then used for model construction. During external validation, the SVC achieved an area under the ROC curve (AUC) value, accuracy, sensitivity, and specificity of 0.83, 0.78, 0.80, and 0.74, respectively. UN and/or EX phase RFs did not further increase the model’s performance. Meanwhile, in the same comparison, the expert radiologist achieved similar performance with an AUC of 0.77, an accuracy of 0.79, a sensitivity of 0.84, and a specificity of 0.69.

Conclusion

Radiomics analysis of CM phase CT scans combined with ML can achieve comparable performance with an expert radiologist in differentiating ccRCCs from non-ccRCCs.

A triple-classification for differentiating renal oncocytoma from renal cell carcinoma subtypes and CK7 expression evaluation: a radiomics analysis

Background

To investigate the value of computed tomography (CT)-based radiomics model analysis in differentiating renal oncocytoma (RO) from renal cell carcinoma subtypes (chromophobe renal cell carcinoma, clear cell carcinoma) and predicting the expression of Cytokeratin 7 (CK7).

Methods

In this retrospective study, radiomics was applied for patients with RO, chRCC and ccRCC who underwent surgery between January 2013 and December 2019 comprised the training cohort, and the testing cohort was collected between January and October 2020. The corticomedullary (CMP) and nephrographic phases (NP) were manually segmented, and radiomics texture parameters were extracted. Support vector machine was generated from CMP and NP after feature selection. Shapley additive explanations were applied to interpret the radiomics features. A radiomics signature was built using the selected features from the two phases, and the radiomics nomogram was constructed by incorporating the radiomics features and clinical factors. Receiver operating characteristic curve was calculated to evaluate the above models in the two sets. Furthermore, Rad-score was used for correlation analysis with CK7.

Results

A total of 123 patients with RO, chRCC and ccRCC were analyzed in the training cohort and 57 patients in the testing cohort. Subsequently, 396 radiomics features were selected from each phase. The radiomics features combining two phases yielded the highest area under the curve values of 0.941 and 0.935 in the training and testing sets, respectively. The Pearson’s correlation coefficient was statistically significant between Rad-score and CK7.

Conclusion

We proposed a non-invasive and individualized CT-based radiomics nomogram to differentiation among RO, chRCC and ccRCC preoperatively and predict the immunohistochemical protein expression for accurate clinical diagnosis and treatment decision.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12894-022-01099-0.

Ultrasomics prediction for cytokeratin 19 expression in hepatocellular carcinoma: A multicenter study

Objective

The purpose of this study was to investigate the preoperative prediction of Cytokeratin (CK) 19 expression in patients with hepatocellular carcinoma (HCC) by machine learning-based ultrasomics.

Methods

We retrospectively analyzed 214 patients with pathologically confirmed HCC who received CK19 immunohistochemical staining. Through random stratified sampling (ratio, 8:2), patients from institutions I and II were divided into training dataset (n = 143) and test dataset (n = 36), and patients from institution III served as external validation dataset (n = 35). All gray-scale ultrasound images were preprocessed, and then the regions of interest were then manually segmented by two sonographers. A total of 1409 ultrasomics features were extracted from the original and derived images. Next, the intraclass correlation coefficient, variance threshold, mutual information, and embedded method were applied to feature dimension reduction. Finally, the clinical model, ultrasonics model, and combined model were constructed by eXtreme Gradient Boosting algorithm. Model performance was assessed by area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy.

Results

A total of 12 ultrasomics signatures were used to construct the ultrasomics models. In addition, 21 clinical features were used to construct the clinical model, including gender, age, Child-Pugh classification, hepatitis B surface antigen/hepatitis C virus antibody (positive/negative), cirrhosis (yes/no), splenomegaly (yes/no), tumor location, tumor maximum diameter, tumor number, alpha-fetoprotein, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, glutamyl-transpeptidase, albumin, total bilirubin, conjugated bilirubin, creatinine, prothrombin time, fibrinogen, and international normalized ratio. The AUC of the ultrasomics model was 0.789 (0.621 – 0.907) and 0.787 (0.616 – 0.907) in the test and validation datasets, respectively. However, the performance of the combined model covering clinical features and ultrasomics signatures improved significantly. Additionally, the AUC (95% CI), sensitivity, specificity, and accuracy were 0.867 (0.712 – 0.957), 0.750, 0.875, 0.861, and 0.862 (0.703 – 0.955), 0.833, 0.862, and 0.857 in the test dataset and external validation dataset, respectively.

Conclusion

Ultrasomics signatures could be used to predict the expression of CK19 in HCC patients. The combination of clinical features and ultrasomics signatures showed excellent effects, which significantly improved prediction accuracy and robustness.

Usefulness of computed tomography textural analysis in renal cell carcinoma nuclear grading

Purpose: To evaluate the usefulness of computed tomography (CT) texture descriptors integrated with machine-learning (ML) models in the identification of clear cell renal cell carcinoma (ccRCC) and for the first time papillary renal cell carcinoma (pRCC) tumor nuclear grades [World Health Organization (WHO)/International Society of Urologic Pathologists (ISUP) 1, 2, 3, and 4]. Approach: A total of 143 ccRCC and 21 pRCC patients were analyzed in this study. Texture features were extracted from late arterial phase CT images. A complete separation of training/validation and testing subsets from the beginning to the end of the pipeline was adopted. Feature dimension was reduced by collinearity analysis and Gini impurity-based feature selection. The synthetic minority over-sampling technique was employed for imbalanced datasets. The ML classifiers were logistic regression, SVM, RF, multi-layer perceptron, and K -NN. The differentiation between low grades/ high grades, grade 1/grade 2, grade 3/grade 4, and between all grades was assessed for ccRCC and pRCC datasets. The classification performance was assessed and compared by certain metrics. Results: Textures-based classifiers were able to efficiently identify ccRCC and pRCC grades. An accuracy and area under the characteristic operating curve (AUC) up to 91%/0.9, 91%/0.9, 90%/0.9, and 88%/1 were reached when discriminating ccRCC low grades/ high grades, grade 1/grade 2, grade 3/grade 4, and all grades, respectively. An accuracy and AUC up to 96%/1, 81%/0.8, 86%/0.9, and 88%/0.9 were found when differentiating pRCC low grades/ high grades, grade 1/grade 2, grade 3/grade 4, and all grades, respectively. Conclusion: CT texture-based ML models can be used to assist radiologist in predicting the WHO/ISUP grade of ccRCC and pRCC pre-operatively.

The utility of texture analysis of kidney MRI for evaluating renal dysfunction with multiclass classification model

We evaluated a multiclass classification model to predict estimated glomerular filtration rate (eGFR) groups in chronic kidney disease (CKD) patients using magnetic resonance imaging (MRI) texture analysis (TA). We identified 166 CKD patients who underwent MRI comprising Dixon-based T1-weighted in-phase (IP)/opposed-phase (OP)/water-only (WO) images, apparent diffusion coefficient (ADC) maps, and T2* maps. The patients were divided into severe, moderate, and control groups based on eGFR borderlines of 30 and 60 mL/min/1.73 m². After extracting 93 texture features (TFs), dimension reduction was performed using inter-observer reproducibility analysis and sequential feature selection (SFS) algorithm. Models were created using linear discriminant analysis (LDA); support vector machine (SVM) with linear, rbf, and sigmoid kernels; decision tree (DT); and random forest (RF) classifiers, with synthetic minority oversampling technique (SMOTE). Models underwent 100-time repeat nested cross-validation. Overall performances of our classification models were modest, and TA based on T1-weighted IP/OP/WO images provided better performance than those based on ADC and T2* maps. The most favorable result was observed in the T1-weighted WO image using RF classifier and the combination model was derived from all T1-weighted images using SVM classifier with rbf kernel. Among the selected TFs, total energy and energy had weak correlations with eGFR.

Machine Learning for Renal Pathologies: An Updated Survey

Within the literature concerning modern machine learning techniques applied to the medical field, there is a growing interest in the application of these technologies to the nephrological area, especially regarding the study of renal pathologies, because they are very common and widespread in our society, afflicting a high percentage of the population and leading to various complications, up to death in some cases. For these reasons, the authors have considered it appropriate to collect, using one of the major bibliographic databases available, and analyze the studies carried out until February 2022 on the use of machine learning techniques in the nephrological field, grouping them according to the addressed pathologies: renal masses, acute kidney injury, chronic kidney disease, kidney stone, glomerular disease, kidney transplant, and others less widespread. Of a total of 224 studies, 59 were analyzed according to inclusion and exclusion criteria in this review, considering the method used and the type of data available. Based on the study conducted, it is possible to see a growing trend and interest in the use of machine learning applications in nephrology, becoming an additional tool for physicians, which can enable them to make more accurate and faster diagnoses, although there remains a major limitation given the difficulty in creating public databases that can be used by the scientific community to corroborate and eventually make a positive contribution in this area.

Diagnostic utility of a conventional MRI-based analysis and texture analysis for discriminating between ovarian thecoma-fibroma groups and ovarian granulosa cell tumors

Objective

To evaluate the diagnostic utility of conventional magnetic resonance imaging (MRI)-based characteristics and a texture analysis (TA) for discriminating between ovarian thecoma-fibroma groups (OTFGs) and ovarian granulosa cell tumors (OGCTs).

Methods

This retrospective multicenter study enrolled 52 patients with 32 OGCTs and 21 OTFGs, which were dissected and pathologically diagnosed between January 2008 and December 2019.

MRI-based features (MBFs) and texture features (TFs) were evaluated and compared between OTFGs and OGCTs. A least absolute shrinkage and selection operator (LASSO) regression analysis was performed to select features and construct the discriminating model. ROC analyses were conducted on MBFs, TFs, and their combination to discriminate between the two diseases.

Results

We selected 3 features with the highest absolute value of the LASSO regression coefficient for each model: the apparent diffusion coefficient (ADC), peripheral cystic area, and contrast enhancement in the venous phase (VCE) for the MRI-based model; the 10th percentile, difference variance, and maximal correlation coefficient for the TA-based model; and ADC, VCE, and the difference variance for the combination model. The areas under the curves of the constructed models were 0.938, 0.817, and 0.941, respectively. The diagnostic performance of the MRI-based and combination models was similar (p = 0.38), but significantly better than that of the TA-based model (p < 0.05).

Conclusions

The conventional MRI-based analysis has potential as a method to differentiate OTFGs from OGCTs. TA did not appear to be of any additional benefit. Further studies are needed on the use of these methods for a preoperative differential diagnosis of these two diseases.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13048-022-00989-z.

Artificial Intelligence in Kidney Cancer

Artificial intelligence is rapidly expanding into nearly all facets of life, particularly within the field of medicine. The diagnosis, characterization, management, and treatment of kidney cancer is ripe with areas for improvement that may be met with the promises of artificial intelligence. Here, we explore the impact of current research work in artificial intelligence for clinicians caring for patients with renal cancer, with a focus on the perspectives of radiologists, pathologists, and urologists. Promising preliminary results indicate that artificial intelligence may assist in the diagnosis and risk stratification of newly discovered renal masses and help guide the clinical treatment of patients with kidney cancer. However, much of the work in this field is still in its early stages, limited in its broader applicability, and hampered by small datasets, the varied appearance and presentation of kidney cancers, and the intrinsic limitations of the rigidly structured tasks artificial intelligence algorithms are trained to complete. Nonetheless, the continued exploration of artificial intelligence holds promise toward improving the clinical care of patients with kidney cancer.

Radiomics analysis for differentiating of cervical lymphadenopathy between cancer of unknown primary and malignant lymphoma on unenhanced computed tomography

To investigate the usefulness of texture analysis to discriminate between cervical lymph node (LN) metastasis from cancer of unknown primary (CUP) and cervical LN involvement of malignant lymphoma (ML) on unenhanced computed tomography (CT). Cervical LN metastases in 17 patients with CUP and cervical LN involvement in 17 patients with ML were assessed by ¹⁸F-FDG PET/CT. The texture features were obtained in the total cross-sectional area (CSA) of the targeted LN, following the contour of the largest cervical LN on unenhanced CT. Values for the max standardized uptake value (SUVmax) and the mean SUV value (SUVmean), and 34 texture features were compared using a Mann-Whitney U test. The diagnostic accuracy and area under the curve (AUC) of the combination of the texture features were evaluated by support vector machine (SVM) with nested cross-validation. The SUVmax and SUVmean did not differ significantly between cervical LN metastases from CUP and cervical LN involvement from ML. However, significant differences of 9 texture features of the total CSA were observed (p = 0.001 - 0.05). The best AUC value of 0.851 for the texture feature of the total CSA were obtained from the correlation in the gray-level co-occurrence matrix features. SVM had the best AUC and diagnostic accuracy of 0.930 and 84.8%. Radiomics analysis appears to be useful for differentiating cervical LN metastasis from CUP and cervical LN involvement of ML on unenhanced CT.

Prediction Models for Prognosis of Femoral Neck–Fracture Patients 6 Months after Total Hip Arthroplasty

Purpose

To establish prediction models for 6-month prognosis in femoral neck–fracture patients receiving total hip arthroplasty (THA).

Patients and Methods

In total, 182 computed tomography image pairs from 85 patients were collected and divided into a training set (n=127) and testing set (n=55). Least absolute shrinkage–selection operator regression was used for selecting optimal predictors. A random-forest algorithm was used to establish the prediction models, which were evaluated for accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC).

Results

The best model in this study was constructed based on demographic data, preoperative laboratory indicators, and three preoperative radiomic features. In the random-forest model, activated partial thromboplastin time, a preoperative radiomic feature (maximum diameter), and fibrinogen were important variables correlating with patient outcomes. The AUC, sensitivity, specificity, PPV, NPV, and accuracy in the training set were 0.986 (95% CI 0.971–1), 0.925 (95% CI 0.862–0.988), 0.983 (95% CI 0.951–1.016), 0.984 (95% CI 0.953–1.014), 0.922 (95% CI 0.856–0.988), and 0.953 (95% CI 0.916–0.990), respectively. The AUC, sensitivity, specificity, PPV, NPV, and accuracy in the testing set were 0.949 (95% CI 0.885–1), 0.767 (95% CI 0.615–0.918), 1 (95% CI 1–1), 1 (95% CI 1–1), 0.781 (95% CI 0.638–0.924), and 0.873 (95% CI 0.785–0.961), respectively.

Conclusion

The model based on demographic, preoperative clinical, and preoperative radiomic data showed the best predictive ability for 6-month prognosis in the femoral neck–fracture patients receiving THA.

Development and Validation of a CT-Based Radiomics Nomogram for Predicting Postoperative Progression-Free Survival in Stage I-III Renal Cell Carcinoma

Background

Many patients experience recurrence of renal cell carcinoma (RCC) after radical and partial nephrectomy. Radiomics nomogram is a newly used noninvasive tool that could predict tumor phenotypes.

Objective

To investigate Radiomics Features (RFs) associated with progression-free survival (PFS) of RCC, assessing its incremental value over clinical factors, and to develop a visual nomogram in order to provide reference for individualized treatment.

Methods

The RFs and clinicopathological data of 175 patients (125 in the training set and 50 in the validation set) with clear cell RCC (ccRCC) were retrospectively analyzed. In the training set, RFs were extracted from multiphase enhanced CT tumor volume and selected using the stability LASSO feature selection algorithm. A radiomics nomogram final model was developed that incorporated the RFs weighted sum and selected clinical predictors based on the multivariate Cox proportional hazard regression. The performances of a clinical variables-only model, RFs-only model, and the final model were compared by receiver operator characteristic (ROC) analysis and DeLong test. Nomogram performance was determined and validated with respect to its discrimination, calibration, reclassification, and clinical usefulness.

Results

The radiomics nomogram included age, clinical stage, KPS score, and RFs weighted sum, which consisted of 6 selected RFs. The final model showed good discrimination, with a C-index of 0.836 and 0.706 in training and validation, and good calibration. In the training set, the C-index of the final model was significantly larger than the clinical-only model (DeLong test, p = 0.008). From the clinical variables-only model to the final model, the reclassification of net reclassification improvement was 18.03%, and the integrated discrimination improvement was 19.08%. Decision curve analysis demonstrated the clinical usefulness of the radiomics nomogram.

Conclusion

The CT-based RF is an improvement factor for clinical variables-only model. The radiomics nomogram provides individualized risk assessment of postoperative PFS for patients with RCC.

MRI-based radiomics model can improve the predictive performance of postlaminar optic nerve invasion in retinoblastoma

OBJECTIVES

To develop an MRI-based radiomics model to predict postlaminar optic nerve invasion (PLONI) in retinoblastoma (RB) and compare its predictive performance with subjective radiologists' assessment.

METHODS

We retrospectively enrolled 124 patients with pathologically proven RB (90 in training set and 34 in validation set) who had MRI scans before surgery. A radiomics model for predicting PLONI was developed by extracting quantitative imaging features from axial T2W images and contrast-enhanced T1W images in the training set. The Kruskal-Wallis test, least absolute shrinkage and selection operator regression, and recursive feature elimination were used for feature selection, where upon a radiomics model was built with a logistic regression (LR) classifier. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve and the accuracy were assessed to evaluate the predictive performance in the training and validation set. The performance of the radiomics model was compared to radiologists' assessment by DeLong test.

RESULTS

The AUC of the radiomics model for the prediction of PLONI was 0.928 in the training set and 0.841 in the validation set. Radiomics model produced better sensitivity than radiologists' assessment (81.1% vs  43.2% in training set, 82.4vs 52.9% in validation set). In all 124 patients, the AUC of the radiomics model was 0.897, while that of radiologists' assessment was 0.674 (p < 0.001, DeLong test).

CONCLUSION

MRI-based radiomics model to predict PLONI in RB patients was shown to be superior to visual assessment with improved sensitivity and AUC, and may serve as a potential tool to guide personalized treatment.

A Combined Model to Improve the Prediction of Local Control for Lung Cancer Patients Undergoing Stereotactic Body Radiotherapy Based on Radiomic Signature Plus Clinical and Dosimetric Parameters

PURPOSE

Stereotactic body radiotherapy (SBRT) is an important treatment modality for lung cancer patients, however, tumor local recurrence rate remains some challenge and there is no reliable prediction tool. This study aims to develop a prediction model of local control for lung cancer patients undergoing SBRT based on radiomics signature combining with clinical and dosimetric parameters.

METHODS

The radiomics model, clinical model and combined model were developed by radiomics features, incorporating clinical and dosimetric parameters and radiomics signatures plus clinical and dosimetric parameters, respectively. Three models were established by logistic regression (LR), decision tree (DT) or support vector machine (SVM). The performance of models was assessed by receiver operating characteristic curve (ROC) and DeLong test. Furthermore, a nomogram was built and was assessed by calibration curve, Hosmer-Lemeshow and decision curve.

RESULTS

The LR method was selected for model establishment. The radiomics model, clinical model and combined model showed favorite performance and calibration (Area under the ROC curve (AUC) 0.811, 0.845 and 0.911 in the training group, 0.702, 0.786 and 0.818 in the validation group, respectively). The performance of combined model was significantly superior than the other two models. In addition, Calibration curve and Hosmer-Lemeshow (training group: P = 0.898, validation group: P = 0.891) showed good calibration of combined nomogram and decision curve proved its clinical utility.

CONCLUSIONS

The combined model based on radiomics features plus clinical and dosimetric parameters can improve the prediction of 1-year local control for lung cancer patients undergoing SBRT.

Machine learning principles applied to CT radiomics to predict mucinous pancreatic cysts

PURPOSE

Current diagnostic and treatment modalities for pancreatic cysts (PCs) are invasive and are associated with patient morbidity. The purpose of this study is to develop and evaluate machine learning algorithms to delineate mucinous from non-mucinous PCs using non-invasive CT-based radiomics.

METHODS

A retrospective, single-institution analysis of patients with non-pseudocystic PCs, contrast-enhanced computed tomography scans within 1 year of resection, and available surgical pathology were included. A quantitative imaging software platform was used to extract radiomics. An extreme gradient boosting (XGBoost) machine learning algorithm was used to create mucinous classifiers using texture features only, or radiomic/radiologic and clinical combined models. Classifiers were compared using performance scoring metrics. Shapely additive explanation (SHAP) analyses were conducted to identify variables most important in model construction.

RESULTS

Overall, 99 patients and 103 PCs were included in the analyses. Eighty (78%) patients had mucinous PCs on surgical pathology. Using multiple fivefold cross validations, the texture features only and combined XGBoost mucinous classifiers demonstrated an area under the curve of 0.72 ± 0.14 and 0.73 ± 0.14, respectively. By SHAP analysis, root mean square, mean attenuation, and kurtosis were the most predictive features in the texture features only model. Root mean square, cyst location, and mean attenuation were the most predictive features in the combined model.

CONCLUSION

Machine learning principles can be applied to PC texture features to create a mucinous phenotype classifier. Model performance did not improve with the combined model. However, specific radiomic, radiologic, and clinical features most predictive in our models can be identified using SHAP analysis.

Classification of malignant tumors by a non-sequential recurrent ensemble of deep neural network model

Many significant efforts have so far been made to classify malignant tumors by using various machine learning methods. Most of the studies have considered a particular tumor genre categorized according to its originating organ. This has enriched the domain-specific knowledge of malignant tumor prediction, we are devoid of an efficient model that may predict the stages of tumors irrespective of their origin. Thus, there is ample opportunity to study if a heterogeneous collection of tumor images can be classified according to their respective stages. The present research work has prepared a heterogeneous tumor dataset comprising eight different datasets from The Cancer Imaging Archives and classified them according to their respective stages, as suggested by the American Joint Committee on Cancer. The proposed model has been used for classifying 717 subjects comprising different imaging modalities and varied Tumor-Node-Metastasis stages. A new non-sequential deep hybrid model ensemble has been developed by exploiting branched and re-injected layers, followed by bidirectional recurrent layers to classify tumor images. Results have been compared with standard sequential deep learning models and notable recent studies. The training and validation accuracy along with the ROC-AUC scores have been found satisfactory over the existing models. No model or method in the literature could ever classify such a diversified mix of tumor images with such high accuracy. The proposed model may help radiologists by acting as an auxiliary decision support system and speed up the tumor diagnosis process.

Accuracy of CT texture analysis for differentiating low-grade and high-grade renal cell carcinoma: systematic review and meta-analysis

OBJECTIVES

This study aimed to assess the accuracy of CT texture analysis (CTTA) for differentiating low-grade and high-grade renal cell carcinoma (RCC).

DESIGN

Systematic review and meta-analysis.

DATA SOURCES

PubMed, Cochrane Library, Embase, Web of Science, OVID Medline, Science Direct and Springer were searched to identify the included studies.

ELIGIBILITY CRITERIA FOR INCLUDING STUDIES

Clinical studies that report about the accuracy of CTTA in differentiating low-grade and high-grade RCC.

METHODS

Multiple databases were searched to identify studies from their inception to 20 October 2021. Two radiologists independently extracted data from the primary studies. The pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR) and diagnostic OR (DOR) were calculated to assess CTTA performance. The summary receiver operating characteristic (SROC) curve was plotted, and the area under the curve (AUC) was calculated to evaluate the accuracy of CTTA in grading RCC.

RESULTS

This meta-analysis included 11 studies, with 1603 lesions observed in 1601 patients. Values of the pooled sensitivity, specificity, PLR, NLR, DOR were 0.79 (95% CI 0.73 to 0.84), 0.84 (95% CI 0.81 to 0.87), 5.1 (95% CI 4.0 to 6.4), 0.24 (95% CI 0.19 to 0.32) and 21 (95% CI 13 to 33), respectively. The SROC curve showed that the AUC was 0.88 (95% CI 0.84 to 0.90). Deeks' test found no significant publication bias among the studies (p=0.42).

CONCLUSIONS

The findings of this meta-analysis suggest that CTTA has a high accuracy in differentiating low-grade and high-grade RCC. A standardised methodology and large sample-based study are necessary to certain the diagnostic accuracy of CTTA in RCC grading for clinical decision making.

Association between nuclear grade of renal cell carcinoma and the aorta-lesion-attenuation-difference

INTRODUCTION AND BACKGROUND

Several features noted on renal mass biopsy (RMB) can influence treatment selection including tumor histology and nuclear grade. However, there is poor concordance between renal cell carcinoma (RCC) nuclear grade on RMB compared to nephrectomy specimens. Here, we evaluate the association of nuclear grade with aorta-lesion-attenuation-difference (ALAD) values determined on preoperative CT scan.

METHODS AND MATERIALS

A retrospective review of preoperative CT scans and surgical pathology was performed on patients undergoing nephrectomy for solid renal masses. ALAD was calculated by measuring the difference in Hounsfield units (HU) between the aorta and the lesion of interest on the same image slice on preoperative CT scan. The discriminative ability of ALAD to differentiate low-grade (nuclear grade 1 and 2) and high-grade (nuclear grade 3 and 4) tumors was evaluated by sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under curve (AUC) using ROC analysis. Sub-group analysis by histologic sub-type was also performed.

RESULTS

A total of 368 preoperative CT scans in patients with RCC on nephrectomy specimen were reviewed. Median patient age was 61 years (IQR 52-68). The majority of patients were male, 66% (243/368). Tumor histology was chromophobe RCC in 7.6%, papillary RCC in 15.5%, and clear cell RCC in 76.9%. The majority, 69.3% (253/365) of tumors, were stage T1a. Nuclear grade was grade 1 in 5.46% (19/348), grade 2 in 64.7% (225/348), grade 3 in 26.2% (91/348), and grade 4 in 3.2% (11/348). Nephrographic ALAD values for grade 1, 2, 3, and 4 were 73.7, 46.5, 36.4, and 43.1, respectively (p = 0.0043). Nephrographic ALAD was able to differentiate low-grade from high-grade RCC with a sensitivity of 32%, specificity of 89%, PPV of 86%, and NPV of 36%. ROC analysis demonstrated the predictive utility of nephrographic ALAD to predict high- versus low-grade RCC with an AUC of 0.60 (95% CI 0.51-0.69).

CONCLUSION

ALAD was significantly associated with nuclear grade in our nephrectomy series. Strong specificity and PPV for the nephrographic phrase demonstrate a potential role for ALAD in the pre-operative setting that may augment RMB findings in assessing nuclear grade of RCC. Although this association was statistically significant, the clinical utility is limited at this time given the results of the statistical analysis (relatively poor ROC analysis). Sub-group analysis by histologic subtype yielded very similar diagnostic performance and limitations of ALAD. Further studies are necessary to evaluate this relationship further.

A systematic review and quality of reporting checklist for repeatability and reproducibility of radiomic features

•

Main factors impacting feature stability: Image acquisition, reconstruction, tumor segmentation, and interpolation.

•

Textural features are less robust than morphological or statistical features.

•

A checklist is provided including items that should be reported in a radiomic study.

Purpose

Although quantitative image biomarkers (radiomics) show promising value for cancer diagnosis, prognosis, and treatment assessment, these biomarkers still lack reproducibility. In this systematic review, we aimed to assess the progress in radiomics reproducibility and repeatability in the recent years.

Methods and materials

Four hundred fifty-one abstracts were retrieved according to the original PubMed search pattern with the publication dates ranging from 2017/05/01 to 2020/12/01. Each abstract including the keywords was independently screened by four observers. Forty-two full-text articles were selected for further analysis. Patient population data, radiomic feature classes, feature extraction software, image preprocessing, and reproducibility results were extracted from each article. To support the community with a standardized reporting strategy, we propose a specific reporting checklist to evaluate the feasibility to reproduce each study.

Results

Many studies continue to under-report essential reproducibility information: all but one clinical and all but two phantom studies missed to report at least one important item reporting image acquisition. The studies included in this review indicate that all radiomic features are sensitive to image acquisition, reconstruction, tumor segmentation, and interpolation. However, the amount of sensitivity is feature dependent, for instance, textural features were, in general, less robust than statistical features.

Conclusions

Radiomics repeatability, reproducibility, and reporting quality can substantially be improved regarding feature extraction software and settings, image preprocessing and acquisition, cutoff values for stable feature selection. Our proposed radiomics reporting checklist can serve to simplify and improve the reporting and, eventually, guarantee the possibility to fully replicate and validate radiomic studies.

CT-based radiomics stratification of tumor grade and TNM stage of clear cell renal cell carcinoma

OBJECTIVES

To evaluate the utility of CT-based radiomics signatures in discriminating low-grade (grades 1-2) clear cell renal cell carcinomas (ccRCC) from high-grade (grades 3-4) and low TNM stage (stages I-II) ccRCC from high TNM stage (stages III-IV).

METHODS

A total of 587 subjects (mean age 60.2 years ± 12.2; range 22-88.7 years) with ccRCC were included. A total of 255 tumors were high grade and 153 were high stage. For each subject, one dominant tumor was delineated as the region of interest (ROI). Our institutional radiomics pipeline was then used to extract 2824 radiomics features across 12 texture families from the manually segmented volumes of interest. Separate iterations of the machine learning models using all extracted features (full model) as well as only a subset of previously identified robust metrics (robust model) were developed. Variable of importance (VOI) analysis was performed using the out-of-bag Gini index to identify the top 10 radiomics metrics driving each classifier. Model performance was reported using area under the receiver operating curve (AUC).

RESULTS

The highest AUC to distinguish between low- and high-grade ccRCC was 0.70 (95% CI 0.62-0.78) and the highest AUC to distinguish between low- and high-stage ccRCC was 0.80 (95% CI 0.74-0.86). Comparable AUCs of 0.73 (95% CI 0.65-0.8) and 0.77 (95% CI 0.7-0.84) were reported using the robust model for grade and stage classification, respectively. VOI analysis revealed the importance of neighborhood operation-based methods, including GLCM, GLDM, and GLRLM, in driving the performance of the robust models for both grade and stage classification.

CONCLUSION

Post-validation, CT-based radiomics signatures may prove to be useful tools to assess ccRCC grade and stage and could potentially add to current prognostic models. Multiphase CT-based radiomics signatures have potential to serve as a non-invasive stratification schema for distinguishing between low- and high-grade as well as low- and high-stage ccRCC.

KEY POINTS

• Radiomics signatures derived from clinical multiphase CT images were able to stratify low- from high-grade ccRCC, with an AUC of 0.70 (95% CI 0.62-0.78). • Radiomics signatures derived from multiphase CT images yielded discriminative power to stratify low from high TNM stage in ccRCC, with an AUC of 0.80 (95% CI 0.74-0.86). • Models created using only robust radiomics features achieved comparable AUCs of 0.73 (95% CI 0.65-0.80) and 0.77 (95% CI 0.70-0.84) to the model with all radiomics features in classifying ccRCC grade and stage, respectively.

MRI-Based Grading of Clear Cell Renal Cell Carcinoma Using a Machine Learning Classifier

Objective

To develop a machine learning (ML)-based classifier for discriminating between low-grade (ISUP I-II) and high-grade (ISUP III-IV) clear cell renal cell carcinomas (ccRCCs) using MRI textures.

Materials and Methods

We retrospectively evaluated a total of 99 patients (with 61 low-grade and 38 high-grade ccRCCs), who were randomly divided into a training set (n = 70) and a validation set (n = 29). Regions of interest (ROIs) of all tumors were manually drawn three times by a radiologist at the maximum lesion level of the cross-sectional CMP sequence images. The quantitative texture analysis software, MaZda, was used to extract texture features, including histograms, co-occurrence matrixes, run-length matrixes, gradient models, and autoregressive models. Reproducibility of the texture features was assessed with the intra-class correlation coefficient (ICC). Features were chosen based on their importance coefficients in a random forest model, while the multi-layer perceptron algorithm was used to build a classifier on the training set, which was later evaluated with the validation set.

Results

The ICCs of 257 texture features were equal to or higher than 0.80 (0.828–0.998. Six features, namely Kurtosis, 135dr_RLNonUni, Horzl_GLevNonU, 135dr_GLevNonU, S(4,4)Entropy, and S(0,5)SumEntrp, were chosen to develop the multi-layer perceptron classifier. A three-layer perceptron model, which has 229 nodes in the hidden layer, was trained on the training set. The accuracy of the model was 95.7% with the training set and 86.2% with the validation set. The areas under the receiver operating curves were 0.997 and 0.758 for the training and validation sets, respectively.

Conclusions

A machine learning-based grading model was developed that can aid in the clinical diagnosis of clear cell renal cell carcinoma using MRI images.

Fuhrman nuclear grade prediction of clear cell renal cell carcinoma: influence of volume of interest delineation strategies on machine learning-based dynamic enhanced CT radiomics analysis

OBJECTIVE

To investigate the influence of different volume of interest (VOI) delineation strategies on machine learning-based predictive models for discrimination between low and high nuclear grade clear cell renal cell carcinoma (ccRCC) on dynamic contrast-enhanced CT.

METHODS

This study retrospectively collected 177 patients with pathologically proven ccRCC (124 low-grade; 53 high-grade). Tumor VOI was manually segmented, followed by artificially introducing uncertainties as: (i) contour-focused VOI, (ii) margin erosion of 2 or 4 mm, and (iii) margin dilation (2, 4, or 6 mm) inclusive of perirenal fat, peritumoral renal parenchyma, or both. Radiomics features were extracted from four-phase CT images (unenhanced phase (UP), corticomedullary phase (CMP), nephrographic phase (NP), excretory phase (EP)). Different combinations of four-phasic features for each VOI delineation strategy were used to build 176 classification models. The best VOI delineation strategy and superior CT phase were identified and the top-ranked features were analyzed.

RESULTS

Features extracted from UP and EP outperformed features from other single/combined phase(s). Shape features and first-order statistics features exhibited superior discrimination. The best performance (ACC 81%, SEN 67%, SPE 87%, AUC 0.87) was achieved with radiomics features extracted from UP and EP based on contour-focused VOI.

CONCLUSION

Shape and first-order features extracted from UP + EP images are better feature representations. Contour-focused VOI erosion by 2 mm or dilation by 4 mm within peritumor renal parenchyma exerts limited impact on discriminative performance. It provides a reference for segmentation tolerance in radiomics-based modeling for ccRCC nuclear grading.

KEY POINTS

• Lesion delineation uncertainties are tolerated within a VOI erosion range of 2 mm or dilation range of 4 mm within peritumor renal parenchyma for radiomics-based ccRCC nuclear grading. • Radiomics features extracted from unenhanced phase and excretory phase are superior to other single/combined phase(s) at differentiating high vs low nuclear grade ccRCC. • Shape features and first-order statistics features showed superior discriminative capability compared to texture features.

Evaluation of radiomics and machine learning in identification of aggressive tumor features in renal cell carcinoma (RCC)

PURPOSE

The purpose of this study was to evaluate the use of CT radiomics features and machine learning analysis to identify aggressive tumor features, including high nuclear grade (NG) and sarcomatoid (sarc) features, in large renal cell carcinomas (RCCs).

METHODS

CT-based volumetric radiomics analysis was performed on non-contrast (NC) and portal venous (PV) phase multidetector computed tomography images of large (> 7 cm) untreated RCCs in 141 patients (46W/95M, mean age 60 years). Machine learning analysis was applied to the extracted radiomics data to evaluate for association with high NG (grade 3-4), with multichannel analysis for NG performed in a subset of patients (n = 80). A similar analysis was performed in a sarcomatoid rich cohort (n = 43, 31M/12F, mean age 63.7 years) using size-matched non-sarcomatoid controls (n = 49) for identification of sarcomatoid change.

RESULTS

The XG Boost Model performed best on the tested data. After manual and machine feature extraction, models consisted of 3, 7, 5, 10 radiomics features for NC sarc, PV sarc, NC NG and PV NG, respectively. The area under the receiver operating characteristic curve (AUC) for these models was 0.59, 0.65, 0.69 and 0.58 respectively. The multichannel NG model extracted 6 radiomic features using the feature selection strategy and showed an AUC of 0.67.

CONCLUSIONS

Statistically significant but weak associations between aggressive tumor features (high nuclear grade, sarcomatoid features) in large RCC were identified using 3D radiomics and machine learning analysis.

Can texture analysis based on single unenhanced CT accurately predict the WHO/ISUP grading of localized clear cell renal cell carcinoma?

OBJECTIVE

The purpose was to investigate the value of texture analysis in predicting the World Health Organization (WHO)/International Society of Urological Pathology (ISUP) grading of localized clear cell renal cell carcinoma (ccRCC) based on unenhanced CT (UECT).

MATERIALS AND METHODS

Pathologically confirmed subjects (n = 104) with localized ccRCC who received UECT scanning were collected retrospectively for this study. All cases were classified into low grade (n = 53) and high grade (n = 51) according to the WHO/ISUP grading and were randomly divided into training set and test set as a ratio of 7:3. Using 3D-ROI segmentation on UECT images and extracted ninety-three texture features (first-order, gray-level co-occurrence matrix [GLCM], gray-level run length matrix [GLRLM], gray-level size zone matrix [GLSZM], neighboring gray tone difference matrix [NGTDM] and gray-level dependence matrix [GLDM] features). Univariate analysis and the least absolute shrinkage selection operator (LASSO) regression were used for feature dimension reduction, and logistic regression classifier was used to develop the prediction model. Using receiver operating characteristic (ROC) curve, bar chart and calibration curve to evaluate the performance of the prediction model.

RESULTS

Dimension reduction screened out eight optimal texture features (maximum, median, dependence variance [DV], long run emphasis [LRE], run entropy [RE], gray-level non-uniformity [GLN], gray-level variance [GLV] and large area low gray-level emphasis [LALGLE]), and then the prediction model was developed according to the linear combination of these features. The accuracy, sensitivity, specificity, and AUC of the model in training set were 86.1%, 91.4%, 81.1%, and 0.937, respectively. The accuracy, sensitivity, specificity, and AUC of the model in test set were 81.2%, 81.2%, 81.2%, and 0.844, respectively. The calibration curves showed good calibration both in training set and test set (P > 0.05).

CONCLUSION

This study has demonstrated that the radiomics model based on UECT texture analysis could accurately evaluate the WHO/ISUP grading of localized ccRCC.

Differentiating Benign from Malignant Renal Tumors Using T2- and Diffusion-Weighted Images: A Comparison of Deep Learning and Radiomics Models Versus Assessment from Radiologists

BACKGROUND

Differentiating benign from malignant renal tumors is important for selection of the most effective treatment.

PURPOSE

To develop magnetic resonance imaging (MRI)-based deep learning (DL) models for differentiation of benign and malignant renal tumors and to compare their discrimination performance with the performance of radiomics models and assessment by radiologists.

STUDY TYPE

Retrospective.

POPULATION

A total of 217 patients were randomly assigned to a training cohort (N = 173) or a testing cohort (N = 44).

FIELD STRENGTH/SEQUENCE

Diffusion-weighted imaging (DWI) and fast spin-echo sequence T2-weighted imaging (T2WI) at 3.0T.

ASSESSMENT

A radiologist manually labeled the region of interest (ROI) on each image. Three DL models using ResNet-18 architecture and three radiomics models using random forest were developed using T2WI alone, DWI alone, and a combination of the two image sets to discriminate between benign and malignant renal tumors. The diagnostic performance of two radiologists was assessed based on professional experience. We also compared the performance of each model and the radiologists.

STATISTICAL TESTS

The area under the receiver operating characteristic (ROC) curve (AUC) was used to assess the performance of each model and the radiologists. P < 0.05 indicated statistical significance.

RESULTS

The AUC of the DL models based on T2WI, DWI, and the combination was 0.906, 0.846, and 0.925 in the testing cohorts, respectively. The AUC of the combination DL model was significantly better than that of the models based on individual sequences (0.925 > 0.906, 0.925 > 0.846). The AUC of the radiomics models based on T2WI, DWI, and the combination was 0.824, 0.742, and 0.826 in the testing cohorts, respectively. The AUC of two radiologists was 0.724 and 0.667 in the testing cohorts.

CONCLUSION

Thus, the MRI-based DL model is useful for differentiating benign from malignant renal tumors in clinic, and the DL model based on T2WI + DWI had the best performance.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2.

MRI Radiomics for the Prediction of Fuhrman Grade in Clear Cell Renal Cell Carcinoma: a Machine Learning Exploratory Study

The Fuhrman nuclear grade is a recognized prognostic factor for patients with clear cell renal cell carcinoma (CCRCC) and its pre-treatment evaluation significantly affects decision-making in terms of management. In this study, we aimed to assess the feasibility of a combined approach of radiomics and machine learning based on MR images for a non-invasive prediction of Fuhrman grade, specifically differentiation of high- from low-grade tumor and grade assessment. Images acquired on a 3-Tesla scanner (T2-weighted and post-contrast) from 32 patients (20 with low-grade and 12 with high-grade tumor) were annotated to generate volumes of interest enclosing CCRCC lesions. After image resampling, normalization, and filtering, 2438 features were extracted. A two-step feature reduction process was used to between 1 and 7 features depending on the algorithm employed. A J48 decision tree alone and in combination with ensemble learning methods were used. In the differentiation between high- and low-grade tumors, all the ensemble methods achieved an accuracy greater than 90%. On the other end, the best results in terms of accuracy (84.4%) in the assessment of tumor grade were achieved by the random forest. These evidences support the hypothesis that a combined radiomic and machine learning approach based on MR images could represent a feasible tool for the prediction of Fuhrman grade in patients affected by CCRCC.

Radiomics to better characterize small renal masses

PURPOSE

Radiomics is a specific field of medical research that uses programmable recognition tools to extract objective information from standard images to combine with clinical data, with the aim of improving diagnostic, prognostic, and predictive accuracy beyond standard visual interpretation. We performed a narrative review of radiomic applications that may support improved characterization of small renal masses (SRM). The main focus of the review was to identify and discuss methods which may accurately differentiate benign from malignant renal masses, specifically between renal cell carcinoma (RCC) subtypes and from angiomyolipoma without visible fat (fat-poor AML) and oncocytoma. Furthermore, prediction of grade, sarcomatoid features, and gene mutations would be of importance in terms of potential clinical utility in prognostic stratification and selecting personalised patient management strategies.

METHODS

A detailed search of original articles was performed using the PubMed-MEDLINE database until 20 September 2020 to identify the English literature relevant to radiomics applications in renal tumour assessment. In total, 42 articles were included in the analysis in 3 main categories related to SRM: prediction of benign versus malignant SRM, subtypes, and nuclear grade, and other features of aggressiveness.

CONCLUSION

Overall, studies reported the superiority of radiomics over expert radiological assessment, but were mainly of retrospective design and therefore of low-quality evidence. However, it is clear that radiomics is an attractive modality that has the potential to improve the non-invasive diagnostic accuracy of SRM imaging and prediction of its natural behaviour. Further prospective validation studies of radiomics are needed to augment management algorithms of SRM.

The role of MRI texture analysis based on susceptibility-weighted imaging in predicting Fuhrman grade of clear cell renal cell carcinoma

BACKGROUND

The Fuhrman nuclear grade system is one of the most important independent indicators in patients with clear cell renal cell carcinoma (ccRCC) for aggressiveness and prognosis. Preoperative assessment of tumor aggressiveness is important for surgical decision-making.

PURPOSE

To explore the role of magnetic resonance imaging (MRI) texture analysis based on susceptibility-weighted imaging (SWI) in predicting Fuhrman grade of ccRCC.

MATERIAL AND METHODS

A total of 45 patients with SWI and surgically proven ccRCC were divided into two groups: the low-grade group (Fuhrman I/II, n = 29) and the high-grade group (Fuhrman III/IV, n = 16). Texture features were extracted from SWI images. Feature selection was performed, and multivariable logistic regression analysis was performed to develop the SWI-based texture model for grading ccRCCs. Receiver operating characteristic (ROC) curve analysis and leave-group-out cross-validation (LGOCV) were performed to test the reliability of the model.

RESULTS

A total of 396 SWI-based texture features were extracted from each SWI image. The SWI-based texture model developed by multivariable logistic regression analysis was: SWIscore = -0.59 + 1.60 * ZonePercentage. The area under the ROC curve of the SWI-based texture model for differentiating high-grade ccRCC from low-grade ccRCC was 0.81 (95% confidence interval 0.67-0.94), with 80% accuracy, 56.25% sensitivity, and 93.10% specificity. After 100 LGOCVs, the mean accuracy, sensitivity, and specificity were 90.91%, 91.83%, and 89.89% for the training sets, and 77.29%, 80.52%, and 71.44% for the test sets, respectively.

CONCLUSION

SWI-based texture analysis might be a reliable quantitative approach for differentiating high-grade ccRCC from low-grade ccRCC.

Development and Validation of a Radiomic Nomogram for Predicting the Prognosis of Kidney Renal Clear Cell Carcinoma

Purpose

The present study aims to comprehensively investigate the prognostic value of a radiomic nomogram that integrates contrast-enhanced computed tomography (CECT) radiomic signature and clinicopathological parameters in kidney renal clear cell carcinoma (KIRC).

Methods

A total of 136 and 78 KIRC patients from the training and validation cohorts were included in the retrospective study. The intraclass correlation coefficient (ICC) was used to assess reproducibility of radiomic feature extraction. Univariate Cox analysis and least absolute shrinkage and selection operator (LASSO) as well as multivariate Cox analysis were utilized to construct radiomic signature and clinical signature in the training cohort. A prognostic nomogram was established containing a radiomic signature and clinicopathological parameters by using a multivariate Cox analysis. The predictive ability of the nomogram [relative operating characteristic curve (ROC), concordance index (C-index), Hosmer–Lemeshow test, and calibration curve] was evaluated in the training cohort and validated in the validation cohort. Patients were split into high- and low-risk groups, and the Kaplan–Meier (KM) method was conducted to identify the forecasting ability of the established models. In addition, genes related with the radiomic risk score were determined by weighted correlation network analysis (WGCNA) and were used to conduct functional analysis.

Results

A total of 2,944 radiomic features were acquired from the tumor volumes of interest (VOIs) of CECT images. The radiomic signature, including ten selected features, and the clinical signature, including three selected clinical variables, showed good performance in the training and validation cohorts [area under the curve (AUC), 0.897 and 0.712 for the radiomic signature; 0.827 and 0.822 for the clinical signature, respectively]. The radiomic prognostic nomogram showed favorable performance and calibration in the training cohort (AUC, 0.896, C-index, 0.846), which was verified in the validation cohort (AUC, 0.768). KM curves indicated that the progression-free interval (PFI) time was dramatically shorter in the high-risk group than in the low-risk group. The functional analysis indicated that radiomic signature was significantly associated with T cell activation.

Conclusions

The nomogram combined with CECT radiomic and clinicopathological signatures exhibits excellent power in predicting the PFI of KIRC patients, which may aid in clinical management and prognostic evaluation of cancer patients.