Usefulness of multidetector computed tomography to differentiate between renal cell carcinoma and oncocytoma. A model validation

Artificial Intelligence-Based Non-invasive Differentiation of Distinct Histologic Subtypes of Renal Tumors With Multiphasic Multidetector Computed Tomography

INTRODUCTION

With rising cases of renal cell carcinoma (RCC), precise identification of tumor subtypes is essential, particularly for detecting small, heterogenous lesions often overlooked in traditional histopathological examinations. This study demonstrates the non-invasive use of deep learning for Histopathological differentiation of renal tumors through quadriphasic multidetector computed tomography (MDCT).

PATIENTS AND METHODS

This prospective longitudinal study includes 50 subjects (32 males, 18 females) with suspected renal tumors. A deep neural network (DNN) is developed to predict RCC subtypes using peak attenuation values measured in Hounsfield Units (HUs) obtained from quadriphasic MDCT scans. The network then generates confidence scores for each of the four primary subtypes of renal tumors, effectively distinguishing between benign oncocytoma and various malignant subtypes.

RESULTS

Our neural network accurately distinguishes Renal tumor subtypes, including clear cell, papillary, chromophobe, and benign oncocytoma, with a confidence score of 68% with the network's diagnosis aligning with Histopathological examinations. Our network was also able to accurately classify RCC subtypes on a synthetically generated dataset with 20,000 samples.

CONCLUSION

We developed an artificial intelligence-based RCC subtype classification technique. Our approach is non-invasive and has the potential to transform the methodology in Renal oncology by providing accurate and timely diagnostic information and enhancing clinical decisions.

Quantitative Measurement on Contrast-Enhanced CT Distinguishes Small Clear Cell Renal Cell Carcinoma From Benign Renal Tumors: A Multicenter Study

RATIONALE AND OBJECTIVES

To evaluate the potential of quantitative measurements on contrast-enhanced CT (CECT) in differentiating small (≤4 cm) clear cell renal cell carcinoma (ccRCC) from benign renal tumors, including fat-poor angiomyolipoma (fpAML) and renal oncocytoma (RO).

MATERIALS AND METHODS

244 patients with pathologically confirmed ccRCC (n = 184) and benign renal tumors (fpAML, n = 50; RO, n = 10) were randomly assigned into training cohort (n = 193) and test cohort 1 (n = 51), while external test cohort 2 (n = 50) was from another hospital. Quantitative parameters were obtained from CECT (unenhanced phase, UP; corticomedullary phase, CMP; nephrographic phase, NP; excretory phase, EP) by measuring attenuation of renal mass and cortex and subsequently calculated. Univariable and multivariable logistic regression analyses were performed to evaluate the association between these parameters and ccRCC. Finally, the constructed models were compared with radiologists' diagnoses.

RESULTS

In univariable analysis, UP-related parameters, particularly UPC-T (cortex minus tumor attenuation on UP), demonstrated AUC of 0.766 in training cohort, 0.901 in test cohort 1, 0.805 in test cohort 2. The heterogeneity-related parameter SD (standard deviation) showed AUC of 0.781, 0.834, and 0.875 respectively. In multivariable analysis, model 1 incorporating UPC-T, NPC-T (cortex minus tumor attenuation on NP), CMPT-UPT (tumor attenuation on CMP minus UP), and SD yielded AUC of 0.866, 0.923, and 0.949 respectively. When compared with radiologists, multivariate models demonstrated higher accuracy (0.800-0.860) and sensitivity (0.794-0.971) than radiologists' assessments (accuracy: 0.700-0.720, sensitivity: 0.588-0.706).

CONCLUSION

Quantitative measurements on CECT, particularly UP- and heterogeneity-related parameters, have potential to discriminate ccRCC and benign renal tumors (fpAML, RO).

Four-phase computed tomography helps differentiation of renal oncocytoma with central hypodense areas from clear cell renal cell carcinoma

PURPOSE

To explore the utility of four-phase computed tomography (CT) in distinguishing renal oncocytoma with central hypodense areas from clear cell renal cell carcinoma (ccRCC).

METHODS

Eighteen patients with oncocytoma and 63 patients with ccRCC presenting with central hypodense areas were included in this study. All patients underwent four-phase CT imaging including the excretory phases later than 20 min after contrast injection. Two blinded experienced radiologists visually reviewed the enhancement features of the central hypodense areas in the excretory phase images and selected the area demonstrating the greatest degree of enhancement of the tumor in the corticomedullary phase images. Regions of interest (ROIs) were placed in the same location in each of the three contrast-enhanced imaging phases. Additionally, ROIs were placed in the adjacent normal renal cortex for normalization. The ratio of the lesion to cortex attenuation (L/C) for the three contrast-enhanced imaging phases and absolute de-enhancement were calculated. The receiver operating characteristic curve was used to obtain the cut-off values.

RESULTS

Complete enhancement inversion of the central areas was observed in 12 oncocytomas (66.67%) and 16 ccRCCs (25.40%) (P = 0.003). Complete enhancement inversion combined with L/C in the corticomedullary phase lower than 1.0 (P < 0.001) or absolute de-enhancement lower than 42.5 HU (P < 0.001) provided 86.42% and 85.19% accuracy, 61.11% and 55.56% sensitivity, 93.65% and 93.65% specificity, 73.33% and 71.43% positive predictive value (PPV), and 89.39% and 88.06% negative predictive value (NPV), respectively, for the diagnosis of oncocytomas. Combined with complete enhancement inversion, L/C in the corticomedullary phase lower than 1.0 and absolute de-enhancement lower than 42.5 HU provided 87.65%, 55.56%, 96.83%, 83.33%, and 88.41% of accuracy, sensitivity, specificity, PPV, and NPV, respectively, for the diagnosis of oncocytomas.

CONCLUSION

The combination of enhancement features of the central hypodense areas and the peripheral tumor parenchyma can help distinguish oncocytoma with central hypodense areas from ccRCC.

Effectiveness of Radiomic ZOT Features in the Automated Discrimination of Oncocytoma from Clear Cell Renal Cancer

Background: Benign renal tumors, such as renal oncocytoma (RO), can be erroneously diagnosed as malignant renal cell carcinomas (RCC), because of their similar imaging features. Computer-aided systems leveraging radiomic features can be used to better discriminate benign renal tumors from the malignant ones. The purpose of this work was to build a machine learning model to distinguish RO from clear cell RCC (ccRCC). Method: We collected CT images of 77 patients, with 30 cases of RO (39%) and 47 cases of ccRCC (61%). Radiomic features were extracted both from the tumor volumes identified by the clinicians and from the tumor’s zone of transition (ZOT). We used a genetic algorithm to perform feature selection, identifying the most descriptive set of features for the tumor classification. We built a decision tree classifier to distinguish between ROs and ccRCCs. We proposed two versions of the pipeline: in the first one, the feature selection was performed before the splitting of the data, while in the second one, the feature selection was performed after, i.e., on the training data only. We evaluated the efficiency of the two pipelines in cancer classification. Results: The ZOT features were found to be the most predictive by the genetic algorithm. The pipeline with the feature selection performed on the whole dataset obtained an average ROC AUC score of 0.87 ± 0.09. The second pipeline, in which the feature selection was performed on the training data only, obtained an average ROC AUC score of 0.62 ± 0.17. Conclusions: The obtained results confirm the efficiency of ZOT radiomic features in capturing the renal tumor characteristics. We showed that there is a significant difference in the performances of the two proposed pipelines, highlighting how some already published radiomic analyses could be too optimistic about the real generalization capabilities of the models.

CT differentiation of the oncocytoma and renal cell carcinoma based on peripheral tumor parenchyma and central hypodense area characterisation

BACKGROUND

Although the central scar is an essential imaging characteristic of renal oncocytoma (RO), its utility in distinguishing RO from renal cell carcinoma (RCC) has not been well explored. The study aimed to evaluate whether the combination of CT characteristics of the peripheral tumor parenchyma (PTP) and central hypodense area (CHA) can differentiate typical RO with CHA from RCC.

METHODS

A total of 132 tumors on the initial dataset were retrospectively evaluated using four-phase CT. The excretory phases were performed more than 20 min after the contrast injection. In corticomedullary phase (CMP) images, all tumors had CHAs. These tumors were categorized into RO (n = 23), clear cell RCC (ccRCC) (n = 85), and non-ccRCC (n = 24) groups. The differences in these qualitative and quantitative CT features of CHA and PTP between ROs and ccRCCs/non-ccRCCs were statistically examined. Logistic regression filters the main factors for separating ROs from ccRCCs/non-ccRCCs. The prediction models omitting and incorporating CHA features were constructed and evaluated, respectively. The effectiveness of the prediction models including CHA characteristics was then confirmed through a validation dataset (8 ROs, 35 ccRCCs, and 10 non-ccRCCs).

RESULTS

The findings indicate that for differentiating ROs from ccRCCs and non-ccRCCs, prediction models with CHA characteristics surpassed models without CHA, with the corresponding areas under the curve (AUC) being 0.962 and 0.914 versus 0.952 and 0.839 respectively. In the prediction models that included CHA parameters, the relative enhancement ratio (RER) in CMP and enhancement inversion, as well as RER in nephrographic phase and enhancement inversion were the primary drivers for differentiating ROs from ccRCCs and non-ccRCCs, respectively. The prediction models with CHA characteristics had the comparable diagnostic ability on the validation dataset, with respective AUC values of 0.936 and 0.938 for differentiating ROs from ccRCCs and non-ccRCCs.

CONCLUSION

The prediction models with CHA characteristics can help better differentiate typical ROs from RCCs. When a mass with CHA is discovered, particularly if RO is suspected, EP images with longer delay scanning periods should be acquired to evaluate the enhancement inversion characteristics of CHA.

Use of specific contrast-enhanced CT regions of interest to differentiate renal oncocytomas from small clear cell and chromophobe renal cell carcinomas

PURPOSE We aimed to examine the usefulness of utilizing a specific contrast-enhanced computed tomog raphy (CT) region of interest (ROI) to differentiate renal oncocytoma (RO) from small clear cell renal cell carcinoma (ccRCC) and chromophobe renal cell carcinoma (chRCC). METHODS A retrospective analysis of pre-contrast phase (PCP), corticomedullary phase (CMP), and nephro graphic phase (NP) contrast-enhanced CT images of the histopathologically confirmed initial cohort (27 ROs, 74 ccRCCs, and 36 chRCCs) was conducted. Small, medium, large, and whole ROIs (S-ROI, M-ROI, L-ROI, and W-ROI, respectively) were utilized for CT attenuation value of tumor (AVT), lesion-to-cortex attenuation (L/C), and heterogeneous degree of tumor (HDT) calcula tions. Differences in these parameters were then compared between RO and ccRCC/chRCC, with receiver operating characteristic (ROC) curves being utilized to gauge the diagnostic utility of the statistically significant parameters. Logistic regression analyses were employed to identify key factors capable of differentiating RO and ccRCC/chRCC, with predictive models further being established. A validation cohort (6 ROs, 30 ccRCCs, and 12 chRCCs) was then employed to vali date the performance of the predictive models. RESULTS Of the parameters evaluated using different ROIs, L/C-CMP (S-ROI) (0.88 ± 0.15 vs. 1.13 ± 0.25, P < .001) and HDT-CMP (W-ROI) (23.02 (12.00-51.21) vs. 37.81 (16.09-89.45), P < .001) were best suited to differentiating RO and ccRCC, yielding respective area under the curve (AUC) values of 0.803 and 0.834. AVT-NP (S-ROI) (122.85 ± 18.87 vs. 86.50 ± 18.65, P < .001) and AVT-NP (M-ROI) (119 (86-167) vs. 81.5 (53-142), P < .001) were better able to differentiate RO and chRCC, yielding respective AUC values of 0.918 and 0.906. Logistic regression analyses revealed that L/C-CMP (S-ROI) and HDT-PCP, as well as AVT-NP (S-ROI) and HDT-CMP, were the primary factors capable of differentiating RO from ccRCC and chRCC, respectively. The predictive model developed to dif ferentiate between RO and ccRCC exhibited a sensitivity of 66.67% and 55.14% in the initial and validation cohorts, respectively, with corresponding specificity of 94.59% and 93.55%, accuracy of 87.13% and 86.84%, and AUC of 0.908 and 0.876. The predictive model developed to differ entiate between RO and chRCC exhibited a sensitivity of 85.19% and 100.00% in the initial and validation cohorts, respectively, with corresponding specificity of 94.59% and 92.86%, accuracy of 87.30% and 95.24%, and AUC of 0.944 and 0.959. CONCLUSION These data demonstrate that a combination of quantitative parameters measured with particu lar ROIs can enable the efficient and reliable differentiation of RO from ccRCC and chRCC for use in routine patient differential diagnosis.

Contrast-enhanced ultrasound (CEUS) of benign and malignant renal tumors: Distinguishing CEUS features differ with tumor size

BACKGROUND

Contrast-enhanced ultrasound (CEUS) is now a guideline-recommended strategy for diagnosing renal lesions. Tumor size is related to the risk of the treatment and prognosis in renal tumors. Thus, we aim to analyze the CEUS features of solid renal tumors in relation to tumor size.

METHODS

The CEUS appearance of 156 pathologically diagnosed solid renal tumors were retrospectively analyzed. Three groups were stratified according to the tumor size (≤2 cm [group I], 2.1-4 cm [group II] and 4.1-7 cm [group III]). For each group, the features of wash-in type, enhancement degree, enhancement homogeneity, and the presence of a pseudocapsule sign were compared between benign and malignant tumors.

RESULTS

All 156 included lesions were detected by CEUS. The proportion of benign tumors in three size groups was 37.1%, 19.4%, and 20.4%, respectively. The proportion of malignant tumors was highest (80.6%) in group II, followed by group III (79.6%) and group I (62.9%). In group I, malignant and benign tumors differed significantly in the presence of a pseudocapsule sign (p = 0.015) and homogeneity (p = 0.007). In group II, the degree of enhancement differed (p = 0.02) between tumor types. In group III, the two tumor types differed in both the wash-in pattern (p = 0.015) and enhancement degree (p = 0.024). The weighted and Cohen's kappa values for the concordance between inter-observer agreement ranged from 0.31 (95% CI: 0.36-0.57) to 0.90 (95% CI: 0.77-1.00).

CONCLUSIONS

CEUS features of malignant and benign renal tumors change along with the tumor size. The use of CEUS features in the diagnosis of benign and malignant tumors requires consideration of tumor size.

The multispeciality approach to the management of localised kidney cancer

Historically, kidney cancer was approached in a siloed single-speciality way, with urological surgeons managing the localised stages of the disease and medical oncologists caring for patients if metastases developed. However, improvements in the management of localised kidney cancer have occurred rapidly over the past two decades with greater understanding of the disease biology, diagnostic options, and innovations in curative treatments. These developments are favourable for patients but provide a substantially more complex landscape for patients and clinicians to navigate, with associated challenging decisions about who to treat, how, and when. As such, the skill sets needed to manage the various aspects of the disease and guide patients appropriately outstrips the capabilities of one particular specialist, and the evolution of a multispeciality approach to the management of kidney cancer is now essential. In this Review, we summarise the current best multispeciality practice for the management of localised kidney cancer and the areas in need of further research and development.

A CT-based radiomics nomogram for differentiation of small masses (< 4 cm) of renal oncocytoma from clear cell renal cell carcinoma

PURPOSE

Renal oncocytoma (RO) is the most commonly resected benign renal tumor because of misdiagnosis as renal cell carcinoma. This misdiagnosis is generally owing to overlapping imaging features. This study describes the building of a radiomics nomogram based on clinical data and radiomics signature for the preoperative differentiation of RO from clear cell renal cell carcinoma (ccRCC) on tri-phasic contrast-enhanced CT.

METHODS

A total of 122 patients (85 in training set and 37 in external validation set) with ROs (n = 46) or ccRCCs (n = 76) were enrolled. Patient characteristics and tri-phasic contrast-enhanced CT imaging features were evaluated to build a clinical factors model. A radiomics signature was constructed by extracting radiomics features from tri-phasic contrast-enhanced CT images and a radiomics score (Rad-score) was calculated. A radiomics nomogram was then built by incorporating the Rad-score and significant clinical factors according to a multivariate logistic regression analysis. The diagnostic performance of the above three models was evaluated in training and validation sets.

RESULTS

Central stellate area and perirenal fascia thickening were selected to build the clinical factors model. Eleven radiomics features were combined to construct the radiomics signature. The AUCs of the radiomics nomogram, which was based on the selected clinical factors and Rad-score, were 0.960 and 0.898 in the training and validation sets, respectively. The decision curves of the radiomics nomogram and radiomics signature in the validation set indicated an overall net benefit over the clinical factors model.

CONCLUSION

Our radiomics nomogram can effectively predict the preoperative diagnosis of ROs and may therefore be of assistance in sparing unnecessary surgery and tailoring precise therapy. The ROC curves of the clinical model, the radiomics signature and the radiomics nomogram for the validation set. RO = Renal oncocytoma; ccRCC = Clear cell renal cell carcinoma.