Multi-sequence MRI-based radiomics model to preoperatively predict the WHO/ISUP grade of clear Cell Renal Cell Carcinoma: a two-center study.

OBJECTIVES: To develop radiomics models based on multi-sequence MRI from two centers for the preoperative prediction of the WHO/ISUP grade of Clear Cell Renal Cell Carcinoma (ccRCC). METHODS: This retrospective study included 334 ccRCC patients from two centers. Significant clinical factors were identified through univariate and multivariate analyses. MRI sequences included Dynamic contrast-enhanced MRI, axial fat-suppressed T2-weighted imaging, diffusion-weighted imaging, and in-phase/out-of-phase images. Feature selection methods and logistic regression (LR) were used to construct clinical and radiomics models, and a combined model was developed using the Rad-score and significant clinical factors. Additionally, seven classifiers were used to construct the combined model and different folds LR was used to construct the combined model to evaluate its performance. Models were evaluated using receiver operating characteristic (ROC) curves, area under the curve (AUC), and decision curve analysis (DCA). The Delong test compared ROC performance, with p < 0.050 considered significant. RESULTS: Multivariate analysis identified intra-tumoral vessels as an independent predictor of high-grade ccRCC. In the external validation set, the radiomics model (AUC = 0.834) outperformed the clinical model (AUC = 0.762), with the combined model achieving the highest AUC (0.855) and significantly outperforming the clinical model (p = 0.003). DCA showed that the combined model had a higher net benefit within the 0.04-0.54 risk threshold range than clinical model. Additionally, the combined model constructed using logistic regression has a higher priority compared to other classifiers. Additionally, 10-fold cross-validation with LR for the combined model showed consistent AUC values (0.849-0.856) across different folds. CONCLUSION: The radiomics models based on multi-sequence MRI might be a noninvasive and effective tool, demonstrating good efficacy in preoperatively predicting the WHO/ISUP grade of ccRCC.

Multi-parametric MRI-based machine learning model for prediction of pathological grade of renal injury in a rat kidney cold ischemia-reperfusion injury model.

BACKGROUND: Renal cold ischemia-reperfusion injury (CIRI), a pathological process during kidney transplantation, may result in delayed graft function and negatively impact graft survival and function. There is a lack of an accurate and non-invasive tool for evaluating the degree of CIRI. Multi-parametric MRI has been widely used to detect and evaluate kidney injury. The machine learning algorithms introduced the opportunity to combine biomarkers from different MRI metrics into a single classifier. OBJECTIVE: To evaluate the performance of multi-parametric magnetic resonance imaging for grading renal injury in a rat model of renal cold ischemia-reperfusion injury using a machine learning approach. METHODS: Eighty male SD rats were selected to establish a renal cold ischemia -reperfusion model, and all performed multiparametric MRI scans (DWI, IVIM, DKI, BOLD, T1mapping and ASL), followed by pathological analysis. A total of 25 parameters of renal cortex and medulla were analyzed as features. The pathology scores were divided into 3 groups using K-means clustering method. Lasso regression was applied for the initial selecting of features. The optimal features and the best techniques for pathological grading were obtained. Multiple classifiers were used to construct models to evaluate the predictive value for pathology grading. RESULTS: All rats were categorized into mild, moderate, and severe injury group according the pathologic scores. The 8 features that correlated better with the pathologic classification were medullary and cortical Dp, cortical T2*, cortical Fp, medullary T2*, ∆T1, cortical RBF, medullary T1. The accuracy(0.83, 0.850, 0.81, respectively) and AUC (0.95, 0.93, 0.90, respectively) for pathologic classification of the logistic regression, SVM, and RF are significantly higher than other classifiers. For the logistic model and combining logistic, RF and SVM model of different techniques for pathology grading, the stable and perform are both well. Based on logistic regression, IVIM has the highest AUC (0.93) for pathological grading, followed by BOLD(0.90). CONCLUSION: The multi-parametric MRI-based machine learning model could be valuable for noninvasive assessment of the degree of renal injury.

CT-based radiomics for predicting pathological grade in hepatocellular carcinoma.

Objective: To construct and validate radiomics models for hepatocellular carcinoma (HCC) grade predictions based on contrast-enhanced CT (CECT). Methods: Patients with pathologically confirmed HCC after surgery and underwent CECT at our institution between January 2016 and December 2020 were enrolled and randomly divided into training and validation datasets. With tumor segmentation and feature extraction, radiomic models were constructed using univariate analysis, followed by least absolute shrinkage and selection operator (LASSO) regression. In addition, combined models with clinical factors and radiomics scores (Radscore) were constructed using logistic regression. Finally, all models were evaluated using the receiver operating characteristic (ROC) curve with the area under the curve (AUC), calibration curve, and decision curve analysis (DCA). Results: In total 242 patients were enrolled in this study, of whom 170 and 72 formed the training and validation datasets, respectively. The arterial phase and portal venous phase (AP+VP) radiomics model were evaluated as the best for predicting HCC pathological grade among all the models built in our study (AUC = 0.981 in the training dataset; AUC = 0.842 in the validation dataset) and was used to build a nomogram. Furthermore, the calibration curve and DCA indicated that the AP+VP radiomics model had a satisfactory prediction efficiency. Conclusions: Low- and high-grade HCC can be distinguished with good diagnostic performance using a CECT-based radiomics model.

Effect of Intraoperative Opioid Dose on Perioperative Neutrophil-to-Lymphocyte Ratio and Lymphocyte-to-Monocyte Ratio in Glioma.

Background: The neutrophil-to-lymphocyte ratio (NLR) and lymphocyte-to-monocyte ratio (LMR) are inflammatory biomarkers. Until now, it is unknown the impact of opioid dosage on perioperative immunity in glioma patients. The aim of this study was to explore the effect of intraoperative opioid dosage on perioperative immune perturbations using NLR and LMR as inflammatory biomarkers and evaluate the correlation between inflammatory biomarkers and pathological grade of glioma. Methods: The study included 208 patients with primary glioma who underwent glioma resection from February 2012 to November 2019 at Harbin Medical University Cancer Hospital. Complete blood count (CBC) was collected at 3 time points: one week before surgery, and 24 hours and one week after surgery. Patients were divided into high-dose and low-dose groups, based on the median value of intraoperative opioid dose. The relationships between perioperative NLR, LMR and intraoperative opioid dosage were analyzed using repeated measurement analysis of variance (ANOVA). Correlations between preoperative various factors and pathological grade were analyzed by Spearman analysis. Receiver operating characteristic (ROC) curves were performed to assess the predictive performance of the NLR and LMR for pathological grade. Results: The NLR (P=0.020) and lower LMR (P=0.037) were statistically significant different between high-dose and low-dose groups one week after surgery. The area under the curve (AUC) of the NLR to identify poor diagnosis was 0.685, which was superior to the LMR (AUC: 0.607) and indicated a correlation between the NLR with pathological grade. The preoperative NLR (P=0.000), LMR (P=0.009), age (P=0.000) and tumor size (P=0.001) exhibited a significant correlation with the pathological grade of glioma. Conclusion: Intraoperative opioids in the high-dose group were associated with higher NLR and lower LMR in postoperative glioma patients. The preoperative NLR and LMR demonstrated predictive value for distinguishing between high-grade and low-grade gliomas.

Comparison of Machine Learning Models Using Diffusion-Weighted Images for Pathological Grade of Intrahepatic Mass-Forming Cholangiocarcinoma.

Is the radiomic approach, utilizing diffusion-weighted imaging (DWI), capable of predicting the various pathological grades of intrahepatic mass-forming cholangiocarcinoma (IMCC)? Furthermore, which model demonstrates superior performance among the diverse algorithms currently available? The objective of our study is to develop DWI radiomic models based on different machine learning algorithms and identify the optimal prediction model. We undertook a retrospective analysis of the DWI data of 77 patients with IMCC confirmed by pathological testing. Fifty-seven patients initially included in the study were randomly assigned to either the training set or the validation set in a ratio of 7:3. We established four different classifier models, namely random forest (RF), support vector machines (SVM), logistic regression (LR), and gradient boosting decision tree (GBDT), by manually contouring the region of interest and extracting prominent radiomic features. An external validation of the model was performed with the DWI data of 20 patients with IMCC who were subsequently included in the study. The area under the receiver operating curve (AUC), accuracy (ACC), precision (PRE), sensitivity (REC), and F1 score were used to evaluate the diagnostic performance of the model. Following the process of feature selection, a total of nine features were retained, with skewness being the most crucial radiomic feature demonstrating the highest diagnostic performance, followed by Gray Level Co-occurrence Matrix lmc1 (glcm-lmc1) and kurtosis, whose diagnostic performances were slightly inferior to skewness. Skewness and kurtosis showed a negative correlation with the pathological grading of IMCC, while glcm-lmc1 exhibited a positive correlation with the IMCC pathological grade. Compared with the other three models, the SVM radiomic model had the best diagnostic performance with an AUC of 0.957, an accuracy of 88.2%, a sensitivity of 85.7%, a precision of 85.7%, and an F1 score of 85.7% in the training set, as well as an AUC of 0.829, an accuracy of 76.5%, a sensitivity of 71.4%, a precision of 71.4%, and an F1 score of 71.4% in the external validation set. The DWI-based radiomic model proved to be efficacious in predicting the pathological grade of IMCC. The model with the SVM classifier algorithm had the best prediction efficiency and robustness. Consequently, this SVM-based model can be further explored as an option for a non-invasive preoperative prediction method in clinical practice.

Prognostic factors among patients with pathological Grade Group 5 prostate cancer based on robot-associated radical prostatectomy specimens from a large Japanese cohort (MSUG94).

PURPOSE: There are no definitive prognostic factors for patients with pathological Grade Group 5 (pGG 5) prostate cancer (PCa) undergoing robot-associated radical prostatectomy (RARP). This study aimed to explore the prognostic factors among patients with pGG 5 PCa in a large Japanese cohort (MSUG94). METHODS: This retrospective, multi-institutional cohort study was conducted between 2012 and 2021 at ten centers in Japan and included 3195 patients. Patients with clinically metastatic PCa (cN1 or cM1) and those receiving neoadjuvant and/or adjuvant therapy were excluded. Finally, 217 patients with pGG5 PCa were analyzed. RESULTS: The median follow-up period was 28.0 months. The 3- and 5-year biochemical recurrence-free survival (BCRFS) rates of the overall population were 66.1% and 57.7%, respectively. The optimal threshold value (47.2%) for the percentage of positive cancer cores (PPCC) with any GG by systematic biopsy was chosen based on receiver operating characteristic curve analysis. Univariate analysis revealed that the prostate-specific antigen level at diagnosis, pT, pN, positive surgical margins (PSMs), lymphovascular invasion, and PPCC were independent prognostic factors for BCRFS. A multivariate analysis revealed that PSMs and PPCC were independent prognostic factors for BCRFS. Using these two predictors, we stratified BCRFS, metastasis-free survival (MFS), and castration-resistant PCa-free survival (CRPC-FS) among patients with pGG 5 PCa. CONCLUSION: The combination of PSMs and PPCC may be an important predictor of BCRFS, MFS, and CRPC-FS in patients with pGG 5 PCa undergoing RARP.

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.

A radiomics-based interpretable model to predict the pathological grade of pancreatic neuroendocrine tumors.

OBJECTIVES: To develop a computed tomography (CT) radiomics-based interpretable machine learning (ML) model to predict the pathological grade of pancreatic neuroendocrine tumors (pNETs) in a non-invasive manner. METHODS: Patients with pNETs who underwent contrast-enhanced abdominal CT between 2010 and 2022 were included in this retrospective study. Radiomics features were extracted, and five radiomics-based ML models, namely logistic regression (LR), random forest (RF), support vector machine (SVM), XGBoost, and GaussianNB, were developed. The performance of these models was evaluated using a time-independent testing set, and metrics such as sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (AUC) were calculated. The accuracy of the radiomics model was compared to that of needle biopsy. The Shapley Additive Explanation (SHAP) tool and the correlation between radiomics and biological features were employed to explore the interpretability of the model. RESULTS: A total of 122 patients (mean age: 50 ± 14 years; 53 male) were included in the training set, whereas 100 patients (mean age: 48 ± 13 years; 50 male) were included in the testing set. The AUCs for LR, SVM, RF, XGBoost, and GaussianNB were 0.758, 0.742, 0.779, 0.744, and 0.745, respectively, with corresponding accuracies of 73.0%, 70.0%, 77.0%, 71.9%, and 72.9%. The SHAP tool identified two features of the venous phase as the most significant, which showed significant differences among the Ki-67 index or mitotic count subgroups (p < 0.001). CONCLUSIONS: An interpretable radiomics-based RF model can effectively differentiate between G1 and G2/3 of pNETs, demonstrating favorable interpretability. CLINICAL RELEVANCE STATEMENT: The radiomics-based interpretable model developed in this study has significant clinical relevance as it offers a non-invasive method for assessing the pathological grade of pancreatic neuroendocrine tumors and holds promise as an important complementary tool to traditional tissue biopsy. KEY POINTS: • A radiomics-based interpretable model was developed to predict the pathological grade of pNETs and compared with preoperative needle biopsy in terms of accuracy. • The model, based on CT radiomics, demonstrated favorable interpretability. • The radiomics model holds potential as a valuable complementary technique to preoperative needle biopsy; however, it should not be considered a replacement for biopsy.

Diagnostic value of 18F-PSMA-1007 PET/CT for predicting the pathological grade of prostate cancer.

This study was designed to evaluate the diagnostic efficacy of relevant parameters of 18F-prostate-specific membrane antigen (PSMA)-1007 PET/CT in predicting the pathological grade of primary prostate cancer. Briefly, a prospective analysis was performed on 53 patients diagnosed with prostate cancer by systematic puncture biopsy, followed by 18F-PSMA-1007 PET/CT examination prior to treatment within 10 d. The patients were grouped in accordance with the Gleason grading system revised by the International Association of Urology Pathology (ISUP). They were divided into high-grade group (ISUP 4-5 group) and low-grade group (ISUP 1-3 group). The differences in maximum standardized uptake value (SUVmax), tumor-to-background ratio (TBR), intraprostatic PSMA-derived tumor volume (iPSMA-TV), and intraprostatic total lesion PSMA (iTL-PSMA) between the high- and low-grade group were statistically significant (p < .001). No significant difference was found for mean standardized uptake value (SUVmean) between the high- and low-grade groups (Z = -1.131, p = .258). Besides, binary multivariate logistic regression analysis showed that only iPSMA-TV and iTL-PSMA were independent predictors of the pathological grading, for which the odds ratios were 18.821 [95% confidence interval (CI): 2.040-173.614, p = .010] and 0.758 (95% CI: 0.613-0.938, p = .011), respectively. The area under the ROC of this regression model was 0.983 (95% CI: 0.958-1.00, p < .001). Only iTL-PSMA was a significant parameter for distinguishing ISUP-4 and ISUP-5 groups (Z = -2.043, p = .041). In a nutshell, 18F-PSMA-1007 PET/CT has good application value in predicting the histopathological grade of primary prostate cancer. Three-dimensional volume metabolism parameters iPSMA-TV and iTL-PSMA were found to be independent predictors for pathological grade.

Spectral CT - a new supplementary method for preoperative assessment of pathological grades of esophageal squamous cell carcinoma.

BACKGROUND: Spectral CT imaging parameters have been reported to be useful in the differentiation of pathological grades in different malignancies. This study aims to investigate the value of spectral CT in the quantitative assessment of esophageal squamous cell carcinoma (ESCC) with different degrees of differentiation. METHODS: There were 191 patients with proven ESCC who underwent enhanced spectral CT from June 2018 to March 2020 retrospectively enrolled. These patients were divided into three groups based on pathological results: well differentiated ESCC, moderately differentiated ESCC, and poorly differentiated ESCC. Virtual monoenergetic 40 keV-equivalent image (VMI40keV), iodine concentration (IC), water concentration (WC), effective atomic number (Eff-Z), and the slope of the spectral curve(λHU) of the arterial phase (AP) and venous phase (VP) were measured or calculated. The quantitative parameters of the three groups were compared by using one-way ANOVA and pairwise comparisons were performed with LSD. Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance of these parameters in poorly differentiated groups and non-poorly differentiated groups. RESULTS: There were significant differences in VMI40keV, IC, Eff-Z, and λHU in AP and VP among the three groups (all p < 0.05) except for WC (p > 0.05). The VMI40keV, IC, Eff-Z, and λHU in the poorly differentiated group were significantly higher than those in the other groups both in AP and VP (all p < 0.05). In the ROC analysis, IC performed the best in the identification of the poorly differentiated group and non-poorly differentiated group in VP (AUC = 0.729, Sensitivity = 0.829, and Specificity = 0.569 under the threshold of 21.08 mg/ml). CONCLUSIONS: Quantitative parameters of spectral CT could offer supplemental information for the preoperative differential diagnosis of ESCC with different degrees of differentiation.

Risk due to Elevated Uric Acid Levels in Children With Henoch-Schonlein Purpura.

PURPOSE: To compare uric acid levels in children with Henoch-Schonlein purpura (HSP)without nephritis and with renal damage, and at different pathological grades. METHODS: A total of 451 children were enrolled in this study, including 64 with HSP without nephritis and 387 HSP with kidney damage. Age, gender, uric acid, urea, creatinine and cystatin C levels were reviewed. Pathological findings of those with renal impairment were also reviewed. RESULTS: Among the HSP children with renal damage, 44 were grade I, 167 were grade II and 176 were grade III. There were significant differences in age, uric acid, urea, creatinine and cystatin C levels between the two groups (p<0.05, all). Correlation analysis showed that uric acid levels in children with HSP without nephritis were positively correlated with urea and creatinine levels (p<0.05). Uric acid levels in HSP children with renal damage was positively correlated with age, urea, creatinine and cystatin C levels (p<0.05, all). Regression analysis found that, without adding any correction factors, there were significant differences in uric acid levels between the two groups; however, after adjusting for pathological grade, there was no longer a significant difference. CONCLUSIONS: There were significant differences of uric acid levels in children with HSP without nephritis and with renal impairment. Uric acid levels in the renal impairment group were significantly higher than that in the HSP without nephritis group. Uric acid levels were related to only the presence or absence of renal damage, not to the pathological grade.

Machine learning models combining computed tomography semantic features and selected clinical variables for accurate prediction of the pathological grade of bladder cancer.

Objective: The purpose of this research was to develop a radiomics model that combines several clinical features for preoperative prediction of the pathological grade of bladder cancer (BCa) using non-enhanced computed tomography (NE-CT) scanning images. Materials and methods: The computed tomography (CT), clinical, and pathological data of 105 BCa patients attending our hospital between January 2017 and August 2022 were retrospectively evaluated. The study cohort comprised 44 low-grade BCa and 61 high-grade BCa patients. The subjects were randomly divided into training (n = 73) and validation (n = 32) cohorts at a ratio of 7:3. Radiomic features were extracted from NE-CT images. A total of 15 representative features were screened using the least absolute shrinkage and selection operator (LASSO) algorithm. Based on these characteristics, six models for predicting BCa pathological grade, including support vector machine (SVM), k-nearest neighbor (KNN), gradient boosting decision tree (GBDT), logical regression (LR), random forest (RF), and extreme gradient boosting (XGBOOST) were constructed. The model combining radiomics score and clinical factors was further constructed. The predictive performance of the models was evaluated based on the area under the receiver operating characteristic (ROC) curve, DeLong test, and decision curve analysis (DCA). Results: The selected clinical factors for the model included age and tumor size. LASSO regression analysis identified 15 features most linked to BCa grade, which were included in the machine learning model. The SVM analysis revealed that the highest AUC of the model was 0.842. A nomogram combining the radiomics signature and selected clinical variables showed accurate prediction of the pathological grade of BCa preoperatively. The AUC of the training cohort was 0.919, whereas that of the validation cohort was 0.854. The clinical value of the combined radiomics nomogram was validated using calibration curve and DCA. Conclusion: Machine learning models combining CT semantic features and the selected clinical variables can accurately predict the pathological grade of BCa, offering a non-invasive and accurate approach for predicting the pathological grade of BCa preoperatively.

Dual-layer spectral detector computed tomography parameters can improve diagnostic efficiency of lung adenocarcinoma grading.

Background: It is difficult to distinguish the pathological grade of lung adenocarcinoma (LUAD) with traditional computed tomography (CT). The aim of this study was to assess tumor differentiation by dual-layer spectral detector CT combined with morphological parameters. Methods: In this prospective study, a total of 67 patients with pathologically diagnosed LUAD were enrolled: 39 patients in the well- and moderately-differentiated group (14 and 25 patients, respectively) and 28 patients in the poorly-differentiated group. Morphological parameters, non-enhanced CT number, double-enhanced CT number, effective atomic number, monoenergetic CT images (40 and 70 keV), iodine density, and thoracic aorta iodine density of tumors were measured. The slope of the spectral curve and normalized iodine density were calculated. The diagnostic efficiency of the spectral parameters alone, and the combined spectral and morphological parameters were obtained by statistical analysis. Results: The morphological signs of LUAD (the vessel convergence sign, bronchus encapsulated air sign, and liquefactive necrosis) were higher in the poorly-differentiated group than in the well-moderately-differentiated group (57.1% vs. 30.8%, 40.0%; 60.7% vs. 28.2%, 32.0%; 64.3% vs. 28.2%, 24.0%; all P<0.05). There were significant differences in normalized iodine density (arterial phase: 0.10±0.04 vs. 0.12±0.05, 0.13±0.04; venous phase: 0.31±0.07 vs. 0.39±0.17, 0.40±0.17) among the poorly-differentiated group and moderately-differentiated group as well as the well-differentiated group (all P<0.05). Receiver operating characteristic (ROC) curves of the poorly-differentiated group and well-moderately-differentiated group showed that the normalized iodine density had the best diagnostic efficiency in the arterial phase, with an area under the curve (AUC) of 0.817, sensitivity of 92.9%, and specificity of 82.1% (P<0.001). The AUC increased to 0.916 when the morphological parameters were included, and sensitivity and specificity were 96.4% and 82.1% (P<0.001), respectively. Conclusions: The parameters of dual-layer spectral detector CT can help discriminate the pathological grade of LUAD. Among the spectral parameters, the normalized iodine density in the arterial phase has the best diagnostic efficiency, and the combination of spectral and morphological parameters improves the pathological grading of LUAD.

Preoperative determination of pathological grades of primary single HCC: development and validation of a scoring model.

PURPOSE: This study aimed to establish a reliable diagnostic score model for the preoperative determination of pathological grade in HCC based on gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) enhanced MRI and biochemical indicators. METHODS: In this retrospective study, we analyzed 139 patients with HCC who underwent Gd-EOB-DTPA MRI between 2014 and 2020, including an establishment cohort of 76 patients and a validation cohort of 63 patients. Based on the imaging features demonstrated on Gd-EOB-DTPA MRI images and biochemical indicators of the establishment cohort, a scoring model based on logistic regression was developed, and compared with postoperative pathological findings in terms of effective determination of pathological grade. The validity of the scoring model was assessed by ROC curves and an independent external validation cohort. RESULTS: Three parameters related to pathological grades were identified, including maximum diameter of the tumor, peritumoral hypointensity in the hepatobiliary phase, and [alkaline phosphatase (U/L) + gamma glutamyl transpeptidase (U/L)]/ lymphocyte count (× 109/L) (AGLR) ratios. Based on these three parameters, a scoring model was developed. ROC curve showed that a score of > 5 was set as the threshold for determining pathological grades with accuracy, sensitivity, specificity, PPV, and NPV of 89.5%, 75.0%, 95.1%, 85.7%, and 90.7%, respectively. CONCLUSION: The study provided the groundwork for a promising and easily implementable scoring model for preoperative determination of HCC pathological grades, for which further validation should be pursued.

The application value of multi-parameter cystoscope in improving the accuracy of preoperative bladder cancer grading.

PURPOSE: To develop and validate a preoperative cystoscopic-based predictive model for predicting postoperative high-grade bladder cancer (BCa), which could be used to guide the surgical selection and postoperative treatment strategies. MATERIALS AND METHODS: We retrospectively recruited 366 patients with cystoscopy biopsy for pathology and morphology evaluation between October 2010 and January 2021. A binary logistic regression model was used to assess the risk factors for postoperative high-grade BCa. Diagnostic performance was analyzed by plotting receiver operating characteristic curve and calculating area under the curve (AUC), sensitivity, specificity. From January 2021 to July 2021, we collected 105 BCa prospectively to validate the model's accuracy. RESULTS: A total of 366 individuals who underwent transurethral resection of bladder tumor (TURBT) or radical cystectomy following cystoscopy biopsy were included for analysis. 261 (71.3%) had a biopsy pathology grade that was consistent with postoperative pathology grade. We discovered five cystoscopic parameters, including tumor diameter, site, non-pedicled, high-grade biopsy pathology, morphology, were associated with high-grade BCa. The established multi-parameter logistic regression model ("JSPH" model) revealed AUC was 0.917 (P < 0.001). Sensitivity and specificity were 86.2% and 84.0%, respectively. And the consistency of pre- and post-operative high-grade pathology was improved from biopsy-based 70.5% to JSPH model-based 85.2%. In a 105-patients prospective validation cohort, the consistency of pre- and post-operative high-grade pathology was increased from 63.1 to 84.2% after incorporation into JSPH model for prediction. CONCLUSION: The cystoscopic parameters based "JSPH model" is accurate at predicting postoperative pathological high-grade tumors prior to operations.

Values of MRI Imaging Presentations in the Hepatobiliary Phase, DWI and T2WI Sequences in Predicting Pathological Grades of Intrahepatic Mass-Forming Cholangiocarcinoma.

Objective: To retrospectively investigate the value of various MRI image menifestations in the hepatobiliary phase (HBP), DWI and T2WI sequences in predicting the pathological grades of intrahepatic mass-forming cholangiocarcinoma (IMCC). Materials and Methods: Forty-three patients of IMCCs confirmed by pathology were enrolled including 25 cases in well- or moderately-differentiated group and 18 cases in poorly-differentiated group. All patients underwent DWI, T2WI and HBP scan. The Chi square test was used to compare the differences in the general information. Logistic regression analysis was used to analyze the risk factors in predicting the pathological grade of IMCCs. Results: The maximal diameter of the IMCC lesion was < 3 cm in 11 patients, between 3 cm and 6 cm in 15, and > 6 cm in 17. Sixteen cases had intrahepatic metastasis, including 5 in the well- or moderately-differentiated group and 11 in the poorly-differentiated group. Seventeen (39.5%) patients presented with target signs in the DWI sequence, including 9 in the well- or moderately-differentiated group and 8 in the poorly-differentiated group. Twenty (46.5%) patients presented with target signs in the T2WI sequence, including 8 in the well- or moderately-differentiated group and 12 in the poorly-differentiated group. Nineteen cases (54.3%) had a complete hypointense signal ring, including 13 in the well- or moderately-differentiated group and 6 in the poorly-differentiated group. Sixteen (45.7%) cases had an incomplete hypointense signal ring, including 5 in the well- or moderately-differentiated group and 11 in the poorly-differentiated group. The lesion size, intrahepatic metastasis, T2WI signal, and integrity of a hypointense signal ring in HBP were statistically significantly different between two gourps. T2WI signal, presence or non-presence of intrahepatic metastasis, and integrity of hypointense signal ring were the independent influencing factors for pathological grade of IMCC. Conclusion: Target sign in T2WI sequence, presence of intrahepatic metastasis and an incomplete hypointense-signal ring in HBP are more likely to be present in poorly-differentiated IMCCs.

Machine Learning Based on Diffusion Kurtosis Imaging Histogram Parameters for Glioma Grading.

Glioma grading plays an important role in surgical resection. We investigated the ability of different feature reduction methods in support vector machine (SVM)-based diffusion kurtosis imaging (DKI) histogram parameters to distinguish glioma grades. A total of 161 glioma patients who underwent magnetic resonance imaging (MRI) from January 2017 to January 2020 were included retrospectively. The patients were divided into low-grade (n = 61) and high-grade (n = 100) groups. Parametric DKI maps were derived, and 45 features from the DKI maps were extracted semi-automatically for analysis. Three feature selection methods [principal component analysis (PCA), recursive feature elimination (RFE) and least absolute shrinkage and selection operator (LASSO)] were used to establish the glioma grading model with an SVM classifier. To evaluate the performance of SVM models, the receiver operating characteristic (ROC) curves of SVM models for distinguishing glioma grades were compared with those of conventional statistical methods. The conventional ROC analysis showed that mean diffusivity (MD) variance, MD skewness and mean kurtosis (MK) C50 could effectively distinguish glioma grades, particularly MD variance. The highest classification distinguishing AUC was found using LASSO at 0.904 ± 0.069. In comparison, classification AUC by PCA was 0.866 ± 0.061, and 0.899 ± 0.079 by RFE. The SVM-PCA model with the lowest AUC among the SVM models was significantly better than the conventional ROC analysis (z = 1.947, p = 0.013). These findings demonstrate the superiority of DKI histogram parameters by LASSO analysis and SVM for distinguishing glioma grades.

Presence of a Ground-Glass Opacity Component Is the True Prognostic Determinant in Clinical Stage I NSCLC.

Introduction: Recent studies have suggested that including presence or absence of ground-glass opacity (GGO) may improve the tumor descriptor (T descriptor) classification in clinical stage I NSCLC. In this study, we analyzed prognostic implications of presence or absence of GGO, size of the solid component, and predominant histology to identify the true prognostic determinant for early-stage NSCLC. Methods: We retrospectively examined 384 patients with clinical stage I NSCLC (solid: 242, part solid: 142) who underwent complete resection between 2009 and 2013. Results: Survival curves of the whole cohort revealed good separation using the current TNM classification. Nevertheless, the part-solid group had a favorable prognosis irrespective of solid component size. Conversely, patients in the solid tumor group with tumors between 3 and 4 cm had a worse prognosis than patients whose tumors were less than or equal to 3 cm. Thus, we propose the following novel T descriptor classification: IA, part-solid tumors; IB, solid tumors less than or equal to 3 cm; and IC, solid tumors between 3 and 4 cm. This novel classification system stratified patient prognosis better than the current classification. On pathologic evaluation, the part-solid group always had better prognoses than the solid group in each subgroup divided by pathologic grade. Conclusions: These results suggest that presence of GGO is the true prognostic determinant of stage I NSCLC, irrespective of the size of the solid component. Our novel T descriptor classification system could more accurately predict prognoses of clinical stage I NSCLC cases.

Human leukocyte antigen-G (HLA-G) expression plays an important role in the diagnosis and grading of endometrial cancer.

The research aimed to investigate the expression of human leukocyte antigen G (HLA-G) in cancer tissues and normal endometrium and the expression of HLA-G in the three different grades of Endometrial cancer, to determine if HLA-G expression is related with the diagnosis and grading of endometrial cancer. The expression of HLA-G protein was analysed in the primary tumour in 97 tissue samples obtained from endometrial cancer, in which 30 samples were at pathological Grade 1; 37 samples were at Grade 2; 27 samples were at Grade 3; and the other 5 samples were obtained from normal endometrium. The HLA-G protein level was measured by immunohistochemical method and analysed according to the clinicopathological parameters of patients. A statistically significant difference (p < .05) was observed in HLA-G expression between the cancerous tissue and the normal endometrium (p = .0007), and the histochemistry score (H-score) of the negative control was 0.05 ± 0.03 (mean ± SD). Statistically significant correlations were also observed between samples of pathological Grade 1 and Grade 2 (p = .0126), Grade 2 and Grade 3 (p = .0359), Grade 1 and Grade 3 (p = .0001). Endometrial cancer cells express higher levels of HLA-G probably to escape immune surveillance, and HLA-G expression level is related with the pathological grade of endometrial cancer. Therefore, HLA-G detecting and quantifying could possibly help diagnosing, grading and treatment of endometrial cancer.Impact statementWhat is already known on this subject? The expression of a member of the non-classical HLA antigens, HLA-G, is one of the main ways for tumour immune escape and progression. The significance of HLA-G in tumour biology has been intensively investigated (Carosella et al. 2015), and now it is widely acknowledged that HLA-G expression in tumours is highly linked with immune suppressive microenvironments, advanced tumour stage, poor therapeutic responses and prognosis (Lin and Yan, 2018). However, to our knowledge, no research has been conducted on the correlation between HLA-G expression and pathological grades of endometrial cancer.What do the results of this study add? Our study demonstrated that the expression of HLA-G plays an important role in the pathological grading of endometrial cancer.What are the implications of these findings for clinical practice and/or further research? Measuring the level of HLA-G expression to help pathological grading of endometrial cancer is important in determining the treatment of patients with endometrial cancer and studying the underlying mechanisms of the development of endometrial cancer, while proving or finding new targeted therapies inhibiting or modifying these processes still requires further investigation.

A Novel MRI-Based Risk Stratification Algorithm for Predicting Postoperative Recurrence of Meningioma: More Benefits to Patients.

Pathological grading of meningioma is insufficient to predict recurrence after resection and to guide individualized treatment strategies. One hundred and thirty-three patients with meningiomas who underwent total resection were enrolled in this retrospective study. Univariate analyses were conducted to evaluate the association between factors and recurrence. Least absolute shrinkage and selection operator (Lasso) was used to further select variables to build a logistic model. The predictive efficiency of the model and WHO grade was compared by using receiver operating characteristic curve (ROC), decision curve analysis (DCA), and net reclassification improvement (NRI). Patients were given a new risk layer based on a nomogram. The recurrence of meningioma in different groups was observed through the Kaplan-Meier curve. Univariate analysis demonstrated that 11 risk factors were associated with prognosis (P < 0.05). The result of ROC proved that the quantified risk-scoring system (AUC = 0.853) had a higher benefit than pathological grade (AUC = 0.689, P = 0.011). The incidence of recurrence of the high risk cohort (69%) was significantly higher than that of the low risk cohort (9%) by Kaplan-Meier analysis (P < 0.001). And all patients who did not relapse in the high risk group received adjuvant radiotherapy. The novel risk stratification algorithm has a significant value for the recurrence of meningioma and can help in optimizing the individualized design of clinical therapy.