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.

Avoid non-diagnostic EUS-FNA: a DNN model as a possible gatekeeper to distinguish pancreatic lesions prone to inconclusive biopsy.

OBJECTIVE: This work aimed to explore the utility of CT radiomics with machine learning for distinguishing the pancreatic lesions prone to non-diagnostic ultrasound-guided fine-needle aspiration (EUS-FNA). METHODS: 498 patients with pancreatic EUS-FNA were retrospectively reviewed [Development cohort: 147 pancreatic ductal adenocarcinoma (PDAC); Validation cohort: 37 PDAC]. Pancreatic lesions not PDAC were also tested exploratively. Radiomics extracted from contrast-enhanced CT was integrated with deep neural networks (DNN) after dimension reduction. The receiver operating characteristic (ROC) curve, and decision curve analysis (DCA) were performed for model evaluation. And, the explainability of the DNN model was analyzed by integrated gradients. RESULTS: The DNN model was effective in distinguishing PDAC lesions prone to non-diagnostic EUS-FNA (Development cohort: AUC = 0.821, 95% CI: 0.742-0.900; Validation cohort: AUC = 0.745, 95% CI: 0.534-0.956). In all cohorts, the DNN model showed better utility than the logistic model based on traditional lesion characteristics with NRI >0 (p < 0.05). And, the DNN model had net benefits of 21.6% at the risk threshold of 0.60 in the validation cohort. As for the model explainability, gray-level co-occurrence matrix (GLCM) features contributed the most averagely and the first-order features were the most important in the sum attribution. CONCLUSION: The CT radiomics-based DNN model can be a useful auxiliary tool for distinguishing the pancreatic lesions prone to nondiagnostic EUS-FNA and provide alerts for endoscopists preoperatively to reduce unnecessary EUS-FNA. ADVANCES IN KNOWLEDGE: This is the first investigation into the utility of CT radiomics-based machine learning in avoiding non-diagnostic EUS-FNA for patients with pancreatic masses and providing potential pre-operative assistance for endoscopists.

Is the radiomics-clinical combined model helpful in distinguishing between pancreatic cancer and mass-forming pancreatitis?

PURPOSE: To develop CT-based radiomics models for distinguishing between resectable PDAC and mass-forming pancreatitis (MFP) and to provide a non-invasive tool for cases of equivocal imaging findings with EUS-FNA needed. METHODS: A total of 201 patients with resectable PDAC and 54 patients with MFP were included. Development cohort: patients without preoperative EUS-FNA (175 PDAC cases, 38 MFP cases); validation cohort: patients with EUS-FNA (26 PDAC cases, 16 MFP cases). Two radiomic signatures (LASSOscore, PCAscore) were developed based on the LASSO model and principal component analysis. LASSOCli and PCACli prediction models were established by combining clinical features with CT radiomic features. ROC analysis and decision curve analysis (DCA) were performed to evaluate the utility of the model versus EUS-FNA in the validation cohort. RESULTS: In the validation cohort, the radiomic signatures (LASSOscore, PCAscore) were both effective in distinguishing between resectable PDAC and MFP (AUCLASSO = 0.743, 95% CI: 0.590-0.896; AUCPCA = 0.788, 95% CI: 0.639-0.938) and improved the diagnostic accuracy of the baseline onlyCli model (AUConlyCli = 0.760, 95% CI: 0.614-0.960) after combination with variables including age, CA19-9, and the double-duct sign (AUCPCACli = 0.880, 95% CI: 0.776-0.983; AUCLASSOCli = 0.825, 95% CI: 0.694-0.955). The PCACli model showed comparable performance to FNA (AUCFNA = 0.810, 95% CI: 0.685-0.935). In DCA, the net benefit of the PCACli model was superior to that of EUS-FNA, avoiding biopsies in 70 per 1000 patients at a risk threshold of 35%. CONCLUSIONS: The PCACli model showed comparable performance with EUS-FNA in discriminating resectable PDAC from MFP.

Noninvasive assessment of renal function and fibrosis in CKD patients using histogram analysis based on diffusion kurtosis imaging.

PURPOSE: To investigate the potential of histogram analysis based on diffusion kurtosis imaging (DKI) in evaluating renal function and fibrosis associated with chronic kidney disease (CKD). MATERIALS AND METHODS: Thirty-six CKD patients were enrolled, and DKI was performed in all patients before the renal biopsy. The histogram parameters of diffusivity (D) and kurtosis (K) were obtained using FireVoxel. The histogram parameters between the stable [estimated glomerular filtration rate (eGFR) ≥ 60 ml/min/1.73 m2] and impaired (eGFR < 60 ml/min/1.73 m2) eGFR group were compared. Besides, patients were classified into mild, moderate, and severe fibrosis group using a semi-quantitative standard. The correlations of histogram parameters with eGFR and fibrosis scores were investigated and the diagnostic performances of histogram parameters in assessing renal dysfunction and fibrosis were analyzed. The added value of combination of most significant parameter with 24 h urinary protein (24 h-UPRO) in evaluating fibrosis was also explored. RESULTS: Seven D histogram parameters in cortex (mean, median, 10th, 25th, 75th, 90th percentiles and entropy), two D histogram parameters in medulla (75th, 90th percentiles), seven K histogram parameters in cortex (mean, min, median, 10th, 25th, 75th, 90th percentiles) and three K histogram parameters in medulla (mean, median, 25th percentile) were significantly different between the two groups. The Dmean of cortex was the most relevant parameter to eGFR (r = 0.648, P < 0.001) and had the largest area under the curve (AUC) for differentiating the stable from impaired eGFR group [AUC = 0.889; 95% confidence interval (CI) 0.728-0.970]. The K90th of cortex presented the strongest correlation with fibrosis scores (r = 0.575, P < 0.001) and achieved the largest AUC for distinguishing the mild from moderate to severe fibrosis group (AUC = 0.849, 95% CI 0.706-0.993). Combining the K90th in cortex with 24 h-UPRO gained statistically higher AUC value (AUC = 0.880, 95% CI 0.763-0.996). CONCLUSION: Histogram analysis based on DKI is practicable for the noninvasive assessment of renal function and fibrosis in CKD patients.

CT-Based Radiomics Can Predict the Efficacy of Anlotinib in Advanced Non-Small-Cell Lung Cancer.

Anlotinib is a small-molecule RTK inhibitor that has achieved certain results in further-line treatment, but many patients do not respond to this drug and lack effective methods for identification. Although radiomics has been widely used in lung cancer, very few studies have been conducted in the field of antiangiogenic drugs. This study aims to develop a new model to predict the efficacy of patients receiving anlotinib by combining pretreatment computed tomography (CT) radiomic characters with clinical characters, in order to assist precision medicine of pulmonary cancer. 254 patients from seven institutions were involved in the study. Lesions were selected according to the RECIST 1.1 criteria, and the corresponding radiomic features were obtained. We constructed prediction models based on clinical, NCE-CT, and CE-CT radiomic features, respectively, and evaluated the prediction performance of the models for training sets, internal validation sets, and external validation sets. In the RAD score only model, the area under curve(AUC) of the NCE-CT cohort was 0.740 (95% CI: 0.622, 0.857) for the training set, 0.711 (95% CI: 0.480, 0.942) for the internal validation set, and 0.633(95% CI: 0.479, 0.787) for the external validation set, while that of the CE-CT cohort was 0.815 (95% CI: 0.705, 0.926) for the training set, 0.771 (95% CI: 0.539, 1.000) for the internal validation set, and 0.701 (95% CI: 0.489, 0.913) for the external validation set. In the RAD score-combined model, the AUC of the NCE-CT cohort was 0.796 (95% CI: 0.691, 0.901) for the training set, 0.579 (95% CI: 0.309, 0.848) for the internal validation set, and 0.590 (95% CI: 0.427, 0.753) for the external validation set, while that of the CE-CT cohort was 0.902 (95% CI: 0.828, 0.977) for the training set, 0.865 (95% CI: 0.696, 1.000) for the internal validation set, and 0.837 (95% CI: 0.682, 0.992) for the external validation set. In conclusion, radiomics has accurate predictions for the efficacy of anlotinib. CE-CT-based radiomic models have the best predictive potential in predicting the efficacy of anlotinib, and model predictions become better when they are combined with clinical characteristics.