Rationale and Objectives: Perineural invasion (PNI) is an important prognostic biomarker for prostate cancer (PCa). This study aimed to develop and validate a predictive model integrating biparametric MRI-based deep learning radiomics and clinical characteristics for the non-invasive prediction of PNI in patients with PCa. Materials and Methods: In this prospective study, 557 PCa patients who underwent preoperative MRI and radical prostatectomy were recruited and randomly divided into the training and the validation cohorts at a ratio of 7:3. Clinical model for predicting PNI was constructed by univariate and multivariate regression analyses on various clinical indicators, followed by logistic regression. Radiomics and deep learning methods were used to develop different MRI-based radiomics and deep learning models. Subsequently, the clinical, radiomics, and deep learning signatures were combined to develop the integrated deep learning-radiomics-clinical model (DLRC). The performance of the models was assessed by plotting the receiver operating characteristic (ROC) curves and precision-recall (PR) curves, as well as calculating the area under the ROC and PR curves (ROC-AUC and PR-AUC). The calibration curve and decision curve were used to evaluate the model's goodness of fit and clinical benefit. Results: The DLRC model demonstrated the highest performance in both the training and the validation cohorts, with ROC-AUCs of 0.914 and 0.848, respectively, and PR-AUCs of 0.948 and 0.926, respectively. The DLRC model showed good calibration and clinical benefit in both cohorts. Conclusion: The DLRC model, which integrated clinical, radiomics, and deep learning signatures, can serve as a robust tool for predicting PNI in patients with PCa, thus aiding in developing effective treatment strategies. (c) 2024 The Association of University Radiologists. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Objective This study aimed to develop and evaluate a predictive model for Human Epidermal Growth Factor Receptor 2 (HER2) expression levels in bladder cancer patients using clinical data and computed tomography (CT) radiomic features across various imaging phases. Methods The investigation involved: (1) compiling clinical data from bladder cancer patients; (2) performing HER2 immunohistochemistry (IHC) assessments post-surgery using the Hercep Test scoring system; (3) delineating tumor regions on CT images to extract radiomic features; (4) utilizing T-tests and Least Absolute Shrinkage and Selection Operator (LASSO) regression to identify the most predictive radiomic features of HER2 status. Decision trees and random forest algorithms were then employed to construct radiomic models. Each model's predictive accuracy, sensitivity, specificity, and area under the curve (AUC) were evaluated through cross-validation, identifying the model with the highest AUC as the optimal radiomic predictor. Results The study included 84 bladder cancer patients, with 53 classified as HER2-negative and 31 as HER2-positive via IHC. Radiomic features that correlated with HER2 status were identified, with three, eight, and two features selected from non-contrast, arterial, and venous phase CTs, respectively. Models based solely on arterial phase features exhibited modest predictive capacity (AUC = 0.44), which improved slightly with the inclusion of clinical data (AUC = 0.48). However, a model integrating features from all three CT phases (totaling 13 features) significantly enhanced performance, achieving an AUC of 0.78, which further improved to 0.83 when combined with clinical variables. Conclusion CT-based radiomics is a viable method for predicting HER2 expression in bladder cancer. The comprehensive model, incorporating features from non-contrast, arterial, and venous CT phases alongside clinical data, demonstrated superior predictive efficacy.

Rationale and Objectives To establish a multimodal deep learning nomogram for predicting clinically significant prostate cancer in patients with gray-zone PSA levels. Methods This retrospective study enrolled 303 patients with pathological results between January 2018 and December 2022. Clinical variables and the PI-RADS v2.1 score were used to construct a clinical model. Radiomics and deep learning features from bp-MRI were used to develop a radiomics model with SVM and a deep learning model, respectively. A hybrid fusion approach was used to integrate the multimodal data and construct combined models (Comb.Rad.model and Comb.DL.model). The robustness of the radiomics model with XGBoost was validated and compared. Model efficacy was assessed through ROC curve and decision curve analysis. A nomogram was developed based on the best-performing model. Results The clinical model had AUCs of 0.845 and 0.779 in the training and testing set. The radiomics model with SVM and the deep learning model achieved AUCs of 0.825 and 0.933 in the training set and 0.811 and 0.907 in the testing set, respectively. The diagnostic performance of the combined models was significantly improved, with Comb.DL.model having a higher AUC than Comb.Rad.model in both the training (0.986 vs. 0.924, P = 0.008) and testing (0.965 vs. 0.859, P = 0.005) set. The diagnostic efficiency of both the radiomics model and Comb.Rad.model with XGBoost were comparable to that of SVM, confirming the robustness of the established model. Conclusion The integrated nomogram combining deep learning features, PI-RADS score, and clinical variables significantly outperformed the traditional radiomics and clinical models.

Background To investigate the role of native T1 mapping in the non-invasive quantitative assessment of renal function and renal fibrosis (RF) in chronic kidney disease (CKD) patients. Methods A prospective analysis of 71 consecutive patients [no RF (0%): 9 cases; mild RF (<25%): 36 cases; moderate RF (25–50%): 17 cases; severe RF (>50%): 9 cases] who were clinically diagnosed with CKD that was pathologically confirmed and who underwent magnetic resonance imaging (MRI) examination between October 2021 and September 2022 was performed. T1-C (mean cortical T1 value), T1-M (mean medullary T1 value), ΔT1 (mean corticomedullary difference) and T1% (mean corticomedullary ratio) values were compared. Correlations between T1 parameters and clinical and histopathological values were analyzed. Regression analysis was performed to determine independent predictors of RF. The areas under the receiver operating characteristic curve (AUC) were calculated to assess the diagnostic value of RF. Results The T1-C, ΔT1 and T1% values (P<0.05) were significantly different in the CKD group, but T1-M was not (P>0.05). The ΔT1 and T1% values showed significant differences in pairwise comparisons among CKD subgroups (P<0.05) except for CKD 2 and 3. ΔT1 and T1% were moderately correlated with the estimated glomerular filtration rate (ΔT1: rs=−0.561; T1%: r=−0.602), serum creatinine (ΔT1: rs=0.591; T1%: rs=0.563), blood urea nitrogen (ΔT1: rs=0.433; T1%: rs=0.435) and histopathological score (ΔT1: rs=0.630; T1%: rs=0.658). ΔT1 and T1%, but not T1-C, were independent predictors of RF (P<0.05). ΔT1 and T1% were set as −410.07 ms and 0.8222 with great specificity [ΔT1: 91.7% (77.5–98.2%); T1%: 97.2% (85.5–99.9%)] to identify mild RF and moderate-severe RF. The optimal cutoff values for differentiating severe RF from mild-moderate RF were −343.81 ms (ΔT1) and 0.8359 (T1%) with high sensitivity [both 100% (66.4–100%)] and specificity [ΔT1: 90.6% (79.3–96.9%); T1%: 94.3% (84.3–98.8%)]. Conclusions ΔT1 and T1% overwhelm T1-C for assessment of renal function and RF in CKD patients. ΔT1 and T1% identify patients with <25% and >50% fibrosis, which can guide clinical decision-making and help to avoid biopsy-related bleeding.

Objective:To explore the consistency of MRI-based ovarian-adnexal report and data system (O-RADS) score and its diagnostic value for ovarian adnexal masses.Methods:The MRI data of 309 patients with ovarian adnexal masses confirmed by pathology were retrospectively collected from January 2017 to August 2021 in the Second Affiliated Hospital of Soochow University, including 327 lesions consisted of 250 benign lesions, 21 borderline lesions, and 56 malignant lesions confirmed by pathology. Borderline and malignant lesions were classified into the malignant group ( n=77) and benign lesions were classified as benign group ( n=250). Two radiologists scored all lesions according to the MRI-based O-RADS, and scored again after 6 months. The proportion of borderline/malignant lesions in each MRI-based O-RADS score was calculated. The weighted Kappa test was used to assess the intra-reader and inter-reader consistency of the image interpretation results. The receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic efficacy of MRI-based O-RADS classification for distinguishing benign and malignant ovarian adnexal masses. Results:The weighted Kappa value of the MRI-based O-RADS score between the two radiologists was 0.810 (95%CI 0.764-0.855), and the weighted Kappa values of the two radiologists′ scores at different times were 0.848 (95%CI 0.806-0.889) and 0.875 (95%CI 0.835-0.914), respectively. The borderline/malignant lesions accounted for 0/16, 0.8% (1/127), 10.1% (10/99), 76.0% (57/75), 9/10 and 0/17, 0 (0/122), 8.0% (8/100), 76.2% (48/63), and 84.0% (21/25) of the lesions in the two radiologists based on the MRI O-RADS score of 1, 2, 3, 4, and 5, respectively. When adopting O-RADS score>3 as a cut-off value, the area under the ROC curve of the two radiologists for distinguishing benign and malignant ovarian adnexal masses was 0.928 (95%CI 0.895-0.954) and 0.942 (95%CI 0.911-0.965), respectively. The sensitivity was 0.857 and 0.896, the specificity was 0.924 and 0.924, and the accuracy was 0.908 and 0.917 respectively.Conclusion:The MRI-based O-RADS yields high diagnostic efficiency in the evaluation of benign and malignant ovarian adnexal masses, and the intra-reader and inter-reader consistency of the image interpretation is strong.

目的 建立基于多参数MRI影像组学结合PI-RADS v2.1 和临床指标的新型列线图,评价其预测临床显著性前列腺癌(csPCa)的价值.方法 回顾性分析204 例患者的资料,进行PI-RADS v2.1 评分和影像组学分析.应用受试者工作特征曲线和临床决策曲线评估临床模型、PI-RADS模型、影像组学模型及各联合模型诊断csPCa的效能和临床获益,基于效能最优模型建立列线图并验证.结果 影像组学模型诊断效能显著优于临床模型和PI-RADS评分模型,差异具有统计学意义(P<0.05).在临床模型或PI-RADS模型中增加影像组学的特征,其联合诊断效能会显著提高(P<0.05).结论 基于多参数MRI影像组学结合PI-RADS v2.1 和临床指标的联合模型所建立的列线图为术前预测csPCa提供了一种无创性的新方法.

目的 比较分析多参数磁共振成像(mp-MRI)和4种简化双参数磁共振成像(bp-MRI)评分方式对前列腺癌(PCa)的诊断效能.方法 回顾性分析197例有病理诊断结果的前列腺病变患者的影像资料.运用Cohen's Kappa检验评估2名医师采用5种评分方式(方式1为mp-MRI,方式2～5为bp-MRI)诊断结果间的一致性.运用受试者工作特征(ROC)曲线计算5种方式诊断PCa和临床显著性前列腺癌(csPCa)的诊断准确性,并加以比较.结果 2名医师采用5种评分方式诊断结果间均有较好的一致性,其中方式2和方式1评分方式相当(k值为0.69)且高于方式3、4、5.诊断PCa的曲线下面积(AUC)值从大到小依次为:方式1(0.898)>方式2(0.889)>方式5(0.882)>方式4(0.875)>方式3(0.872),且方式2和方式1,方式5和方式1间诊断AUC值均无统计学差异(P>0.05).诊断csPCa的AUC值从大到小同样为:方式1(0.940)>方式2(0.928)>方式5(0.911)>方式4(0.908)>方式3(0.897),仅方式2和方式1间诊断AUC值无统计学差异(P>0.05).结论 运用基于bp-MRI的评分方式(方式2)诊断PCa及csPCa的效能与mp-MRI基本相当,值得临床推广应用.

目的 探讨钆塞酸二钠增强MRI肝胆期肝脏与脾脏信号强度(SI)比(LSC)、肝脏与门静脉SI比(LPC)及门静脉与脾脏SI比(PSC)评价肝硬化患者肝功能的临床价值.方法 选取接受钆塞酸二钠增强MRI检查的128例患者,依据肝功能分为对照组(41例)、Child-Pugh A组(53例)、Child-Pugh B组(26例)和Child-Pugh C组(8例).记录肝胆期肝脏、门静脉及脾脏SI,并计算LSC、LPC、PSC.采用单因素方差分析比较各组诸SI比值间的差异;采用Spearman线性相关分析比较SI比值与肝功能评分的相关性.结果 LSC及LPC从对照组到Child-Pugh C组逐渐减低(P＜0.001),PSC各组间差异无统计学意义.Child-Pugh评分与LSC及LPC呈中度负相关(r值分别为-0.514,-0.530,P均＜0.001),MELD评分与LSC及LPC呈中度负相关(r值分别为-0.614,-0.620,P均＜0.001).结论 钆塞酸二钠增强MRI肝胆期LSC及LPC能够反映肝硬化患者肝功能损害的严重程度,可以作为评价肝功能的影像学指标,PSC不能反映肝功能情况.

目的基于双参数MRI图像纹理构建一个影像组学模型,并探讨其对临床显著性前列腺癌(clinically significant prostate cancer,csPCa)的诊断价值.材料与方法回顾性分析381例(非csPCa组239例,csPCa组142例)患者临床、病理及影像资料.通过图像预处理与分割,特征提取与选择,建立影像组学模型,评估模型对csPCa的诊断价值.结果基于双参数MRI图像所提取的影像组学特征在观察者内及观察者间均具有良好的一致性,构建的影像组学模型对csPCa具有较高的诊断价值,训练组和测试组的曲线下面积(area under the curve,AUC)值分别为0.991、0.983.结论双参数MRI是检出csPCa的有效方法,经训练并测试所构建的影像组学模型对csPCa具有较高的诊断价值,且相对客观、准确,可作为临床诊断csPCa的辅助方法,为临床制订患者诊疗决策提供重要参考依据.

Abstract Background To investigate the value of 18F-FDG PET/CT molecular radiomics combined with a clinical model in predicting thoracic lymph node metastasis (LNM) in invasive lung adenocarcinoma (≤ 3 cm). Methods A total of 528 lung adenocarcinoma patients were enrolled in this retrospective study. Five models were developed for the prediction of thoracic LNM, including PET radiomics, CT radiomics, PET/CT radiomics, clinical and integrated PET/CT radiomics-clinical models. Ten PET/CT radiomics features and two clinical characteristics were selected for the construction of the integrated PET/CT radiomics-clinical model. The predictive performance of all models was examined by receiver operating characteristic (ROC) curve analysis, and clinical utility was validated by nomogram analysis and decision curve analysis (DCA). Results According to ROC curve analysis, the integrated PET/CT molecular radiomics-clinical model outperformed the clinical model and the three other radiomics models, and the area under the curve (AUC) values of the integrated model were 0.95 (95% CI: 0.93–0.97) in the training group and 0.94 (95% CI: 0.89–0.97) in the test group. The nomogram analysis and DCA confirmed the clinical application value of this integrated model in predicting thoracic LNM. Conclusions The integrated PET/CT molecular radiomics-clinical model proposed in this study can ensure a higher level of accuracy in predicting the thoracic LNM of clinical invasive lung adenocarcinoma (≤ 3 cm) compared with the radiomics model or clinical model alone.

PurposeTo compare the performance of radiomics to that of the Prostate Imaging Reporting and Data System (PI-RADS) v2.1 scoring system in the detection of clinically significant prostate cancer (csPCa) based on biparametric magnetic resonance imaging (bpMRI) vs. multiparametric MRI (mpMRI).MethodsA total of 204 patients with pathological results were enrolled between January 2018 and December 2019, with 142 patients in the training cohort and 62 patients in the testing cohort. The radiomics model was compared with the PI-RADS v2.1 for the diagnosis of csPCa based on bpMRI and mpMRI by using receiver operating characteristic (ROC) curve analysis.ResultsThe radiomics model based on bpMRI and mpMRI signatures showed high predictive efficiency but with no significant differences (AUC = 0.975 vs 0.981, p=0.687 in the training cohort, and 0.953 vs 0.968, p=0.287 in the testing cohort, respectively). In addition, the radiomics model outperformed the PI-RADS v2.1 in the diagnosis of csPCa regardless of whether bpMRI (AUC = 0.975 vs. 0.871, p= 0.030 for the training cohort and AUC = 0.953 vs. 0.853, P = 0.024 for the testing cohort) or mpMRI (AUC = 0.981 vs. 0.880, p= 0.030 for the training cohort and AUC = 0.968 vs. 0.863, P = 0.016 for the testing cohort) was incorporated.ConclusionsOur study suggests the performance of bpMRI- and mpMRI-based radiomics models show no significant difference, which indicates that omitting DCE imaging in radiomics can simplify the process of analysis. Adding radiomics to PI-RADS v2.1 may improve the performance to predict csPCa.

目的 探讨基于2.1版前列腺影像报告与数据系统(PI-RADS v2.1)的双参数磁共振成像(bp-MRI)联合临床相关指标对前列腺特异性抗原(PSA)灰区中临床显著性前列腺癌(csPCa)的诊断价值.方法 回顾性分析211例PSA灰区(4～10 ng/mL)患者的临床、影像及病理资料,其中csPCa组35例,非csPCa组176例.根据PI-RADS v2.1评分标准对前列腺主病灶进行bp-MRI评分.对年龄、总前列腺特异性抗原(tPSA)、游离前列腺特异性抗原(fPSA)、游离与总前列腺特异性抗原比值(f/tPSA)、前列腺体积(PV)、前列腺特异性抗原密度(PSAD)及bp-MRI评分进行单因素和多因素分析,确定csPCa独立预测因子,并建立联合预测模型.运用受试者工作特征(ROC)曲线评估各独立预测因子及联合预测模型对csPCa的诊断效能,并通过Z检验对曲线下面积(AUC)进行两两比较.结果 f/tPSA、PV、PSAD及bp-MRI评分在csPCa组和非csPCa组间存在统计学差异(均P<0.05).PV和bp-MRI评分为csPCa的独立预测因子(OR=0.974,P=0.024;OR=4.206;P<0.001).PV、bp-MRI评分和二者联合预测模型诊断csPCa的AUC值分别为0.684、0.856、0.878,且PV和bp-MRI评分间,PV和联合预测模型间,bp-MRI评分和联合预测模型间AUC值差异均有统计学意义(Z=3.416,P=0.001;Z=4.562,P<0.001;Z=2.059,P=0.040).结论 Bp-MRI有助于检出PSA灰区csPCa,与PV联合应用后,可进一步提高对csPCa的检出效能,减少患者不必要的穿刺.

Background This study attempted to develop a nomogram for predicting clinically significant prostate cancer (cs-PCa) in the transition zone (TZ) with the Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1) score based on biparametric magnetic resonance imaging (bp-MRI) and clinical indicators. Methods We retrospectively reviewed 383 patients with suspicious prostate lesions in the TZ as a training cohort and 128 patients as the validation cohort from January 2015 to March 2020. Multivariable logistic regression analysis was performed to determine independent predictors for building a nomogram, and the performance of the nomogram was assessed by the area under the receiver operating characteristic curve (AUC), the calibration curve and decision curve. Results The PI-RADS v2.1 score and prostate-specific antigen density (PSAD) were independent predictors of TZ cs-PCa. The prediction model had a significantly higher AUC (0.936) than the individual predictors (0.914 for PI-RADS v2.1 score, P=0.045, 0.842 for PSAD, P<0.001). The nomogram showed good discrimination (AUC of 0.936 in the training cohort and 0.963 in the validation cohort) and favorable calibration. When the PI-RADS v2.1 score was combined with PSAD, the diagnostic sensitivity and specificity were 80.7% and 93.8%, respectively, which were better than those of the PI-RADS v2.1 score (sensitivity, 74.2%; specificity, 92.5%) and PSAD (sensitivity, 66.1%; specificity, 88.2%). Conclusions The newly constructed nomogram exhibits satisfactory predictive accuracy and consistency for TZ cs-PCa. PI-RADS v2.1 based on bp-MRI is a strong predictor in the detection of TZ cs-PCa. Adding PSAD to PI-RADS v2.1 could improve its diagnostic performance, thereby avoiding unnecessary biopsies.

Objective:To compare the diagnostic value between prostate imaging reporting and data system version 2 (PI-RADS V2) and version 2.1 (PI-RADS V2.1) for clinically significant prostate cancer (csPCa).Methods:The imaging, pathological and clinical data of 837 patients with prostatic multiparametric MRI in Second Affiliated Hospital of Soochow University from May 2015 to August 2019 were retrospectively analyzed. According to the pathological results of systematic biopsy, the prostate cancer with Gleason score (GS) ≥3+4 was csPCa. A total of 25% of the patients (209 cases) were selected using a simple random sampling, and the index lesions were scored by 2 radiologists with PI-RADS V2 and V2.1, respectively. The weighted Kappa test was used to evaluate the consistency of the scores interpreted between the 2 radiologists. The remaining cases were scored by one of the radiologists using the 2 scoring system respectively. The ROC curve was used to evaluate the diagnostic performance of the 2 scoring system for csPCa in total lesions, peripheral lesions and transitional lesions. Z test was used to investigate whether there was any difference in the detection efficiency between the 2 scoring system. Results:There were 251 patients with csPCa, including 163 patients in peripheral zone and 88 patients in transitional zone. The weighted Kappa value of total lesions, transitional lesions, peripheral lesions was 0.757, 0.653, 0.748 for PI-RADS V2 and 0.794, 0.707, 0.759 for PI-RADS V2.1, respectively. In total lesions, transitional lesions and peripheral lesions, the area under the ROC curve of csPCa detected by PI-RADS V2.1 was 0.922, 0.932, 0.854 and 0.902, 0.905, 0.817 by PI-RADS V2, respectively, and all the differences were statistically significant ( Z=4.104, P<0.001; Z=2.538, P=0.011; Z=3.350, P<0.001). Conclusion:PI-RADS V2.1 has a slightly higher consistent weighted Kappa value in evaluating prostate lesions than PI-RADS V2, and the detection efficiency of csPCa was higher than PI-RADS V2.

OBJECTIVES:Anaplastic lymphoma kinase (ALK) rearrangement status examination has been widely used in clinic for non-small cell lung cancer (NSCLC) patients in order to find patients that can be treated with targeted ALK inhibitors. This study intended to non-invasively predict the ALK rearrangement status in lung adenocarcinomas by developing a machine learning model that combines PET/CT radiomic features and clinical characteristics.METHODS:Five hundred twenty-six patients of lung adenocarcinoma with PET/CT scan examination were enrolled, including 109 positive and 417 negative patients for ALK rearrangements from February 2016 to March 2019. The Artificial Intelligence Kit software was used to extract radiomic features of PET/CT images. The maximum relevance minimum redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO) logistic regression were further employed to select the most distinguishable radiomic features to construct predictive models. The mRMR is a feature selection method, which selects the features with high correlation to the pathological results (maximum correlation), meanwhile retain the features with minimum correlation between them (minimum redundancy). LASSO is a statistical formula whose main purpose is the feature selection and regularization of data model. LASSO method regularizes model parameters by shrinking the regression coefficients, reducing some of them to zero. The feature selection phase occurs after the shrinkage, where every non-zero value is selected to be used in the model. Receiver operating characteristic (ROC) analysis was used to evaluate the performance of the models, and the performance of different models was compared by the DeLong test.RESULTS:A total of 22 radiomic features were extracted from PET/CT images for constructing the PET/CT radiomic model, and majority of these features used were based on CT features (20 out of 22), only 2 PET features were included (PET percentile 10 and PET difference entropy). Moreover, three clinical features associated with ALK mutation (age, burr and pleural effusion) were also employed to construct a combined model of PET/CT and clinical model. We found that this combined model PET/CT-clinical model has a significant advantage to predict the ALK mutation status in the training group (AUC = 0.87) and the testing group (AUC = 0.88) compared with the clinical model alone in the training group (AUC = 0.76) and the testing group (AUC = 0.74) respectively. However, there is no significant difference between the combined model and PET/CT radiomic model.CONCLUSIONS:This study demonstrated that PET/CT radiomics-based machine learning model has potential to be used as a non-invasive diagnostic method to help diagnose ALK mutation status for lung adenocarcinoma patients in the clinic.

为了从多参数磁共振（mp-MRI）的前列腺区域中自动提取前列腺癌病灶区域,提出新的深度卷积神经网络模型SE-Mask-RCNN.在特征图上搜索定位包含病灶的候选区域,基于候选区域实现病灶的精细分割.为了利用mp-MRI中的互补信息,通过2个并行卷积网络分别提取表观扩散系数（ADC）和T2加权（T2W）图像的特征图后进行融合,使用挤压与激励块自动提升融合特征图中的有效特征并抑制无效特征.在收集得到的140例数据上进行实验.结果表明,使用SE-Mask-RCNN得到前列腺癌病灶分割Dice系数为0.654,敏感度为0.695,特异度为0.970,阳性预测值为0.685.与U-net、V-net、Resnet50-U-net和Mask-RCNN等模型相比,SE-Mask-RCNN能够有效提升mpMRI中前列腺癌病灶区域的分割精度.

目的:探讨基于2.1版前列腺影像报告与数据系统(PI-RADS v2.1)双参数磁共振成像(bp-MRI)联合临床常用指标对临床显著性前列腺癌(csPCa)的诊断效能.方法:回顾性分析2018年3月-2020年4月本院515例经病理证实且行前列腺bp-MRI检查(T2 WI+DWI)患者的相关资料,单因素分析患者bp-MRI PI-RADS v2.1评分、年龄、前列腺特异性抗原(PSA)、游离前列腺特异性抗原(fP-SA)、前列腺特异性抗原密度(PSAD)以及前列腺体积(PV)在csPCa组和非csPCa组间的差异,多因素logistic回归分析上述有统计学差异的指标得出csPCa的独立预测因素.利用约登指数确定各独立预测因素的最佳诊断阈值,通过受试者工作特性曲线(ROC)评估各因素单独及联合诊断csPCa的效能,并通过Z检验加以比较.结果:515例患者,年龄47～93岁,平均70±8岁,其中csPCa116例,非csP-Ca399例(低危癌,即Gleason评分≤3+338例,非PCa 361例).单因素分析显示PI-RADS v2.1评分和上述临床指标在csPCa组和非csPCa组间均有统计学差异(P<0.05).多因素logistic回归分析显示PI-RADS v2.1评分(OR=7.015,95％CI=4.776,10.302)和PSAD(OR=5.545,95％CI=2.364,13.007)为csPCa的独立预测因素(P<0.001).当PI-RADS v2.1评分≥4、PSAD≥0.25μg/L·mL时约登指数最大,ROC曲线表明两者联合诊断csPCa的准确性高于bp-MRI和PSAD单独应用(AUC分别为0.939、0.887、0.777),差异有统计学差异(Z=2.259、5.573,均P<0.05).结论:基于PI-RADS v2.1的bp-MRI联合PSAD对csPCa的诊断有一定价值,当PI-RADS v2.1评分≥4、PSAD≥0.25μg/L·mL时,两者联合显著提高对csPCa的诊断效能,优于单独应用.

OBJECTIVES:This study aims to develop a clinically practical model to predict EGFR mutation in lung adenocarcinoma patients according to radiomics signatures based on PET/CT and clinical risk factors.METHODS:This retrospective study included 583 lung adenocarcinoma patients, including 295 (50.60%) patients with EGFR mutation and 288 (49.40%) patients without EGFR mutation. The clinical risk factors associated with lung adenocarcinoma were collected at the same time. We developed PET/CT, CT, and PET radiomics models for the prediction of EGFR mutation using multivariate logistic regression analysis, respectively. We also constructed a combined PET/CT radiomics-clinical model by nomogram analysis. The diagnostic performance and clinical net benefit of this risk-scoring model were examined via receiver operating characteristic (ROC) curve analysis while the clinical usefulness of this model was evaluated by decision curve analysis (DCA).RESULTS:The ROC analysis showed predictive performance for the PET/CT radiomics model (AUC = 0.76), better than the PET model (AUC = 0.71, Delong test: Z = 3.03, p value = 0.002) and the CT model (AUC = 0.74, Delong test: Z = 1.66, p value = 0.098). Also, the PET/CT radiomics-clinical combined model has a better performance (AUC = 0.84) to predict EGFR mutation than the PET/CT radiomics model (AUC = 0.76, Delong test: D = 2.70, df = 790.81, p value < 0.001) or the clinical model (AUC = 0.81, Delong test: Z = 3.46, p value < 0.001).CONCLUSIONS:We demonstrated that the combined PET/CT radiomics-clinical model has an advantage to predict EGFR mutation in lung adenocarcinoma.KEY POINTS:• Radiomics from lung tumor increase the efficiency of the prediction for EGFR mutation in clinical lung adenocarcinoma on PET/CT. • A radiomic nomogram was developed to predict EGFR mutation. • Combining PET/CT radiomics-clinical model has an advantage to predict EGFR mutation.

BACKGROUND. PI-RADS version 2.1 (v2.1) introduced a number of key changes to the assessment of transition zone (TZ) lesions. OBJECTIVE. The purpose of this study was to evaluate interobserver agreement and diagnostic accuracy for detecting TZ prostate cancer (PCa) and clinically significant PCa (csPCa) by use of PI-RADS v2 and PI-RADS v2.1 among radiologists with different levels of experience. METHODS. This retrospective study included 355 biopsy-naïve patients who from January 2017 to March 2020 underwent prostate MRI that showed a TZ lesion and underwent subsequent biopsy. PCa was diagnosed in 93 patients (International Society of Urological Pathology [ISUP] grade group 1, n = 34; ISUP grade group ≥ 2, n = 59) and non-cancerous lesions in 262 patients. Five radiologists with varying experience in prostate MRI scored lesions using PI-RADS v2 and PI-RADS v2.1 in sessions separated by at least 4 weeks. Interobserver agreement was evaluated with kappa and Kendall W statistics. ROC curve analysis was used to evaluate performance in detection of TZ PCa and csPCa. RESULTS. Interobserver agreement among all readers was higher for PI-RADS v2.1 than for PI-RADS v2 (mean weighted κ = 0.700 vs 0.622; Kendall W = 0.805 vs 0.728; p = .03). The pooled AUC values for detecting TZ PCa and csPCa were higher among all readers using PI-RADS v2.1 (0.866 vs 0.827 for TZ PCa; 0.929 vs 0.899 for TZ csPCa; p < .001). For detecting TZ PCa, the pooled sensitivity, specificity, and accuracy were 86.9%, 79.4%, and 75.4% among all readers for PI-RADS v2.1 compared with 79.4%, 71.8%, and 73.8% for PI-RADS v2. For detecting TZ csPCa, the pooled sensitivity, specificity, and accuracy were 84.8%, 90.9%, and 89.9% among all readers for PI-RADS v2.1 compared with 81.4%, 89.9%, and 88.5% for PI-RADS v2. Reader 1, who had the least experience, had the lowest sensitivity, specificity, and accuracy (78.0%, 89.2%, and 87.3%). Reader 5, who had the most experience, had the highest sensitivity, specificity, and accuracy (88.1%, 92.9%, and 92.1%) in detecting csPCa. CONCLUSION. PI-RADS v2.1 had better interobserver agreement and diagnostic accuracy than PI-RADS v2 for evaluating TZ lesions. Reader experience continues to affect the performance of prostate MRI interpretation with PI-RADS v2.1. CLINICAL IMPACT. PI-RADS v2.1 is more accurate and reproducible than PI-RADS v2 for the diagnosis of TZ PCa.

Background: To evaluate the potential of clinical-based model, a biparametric MRI-based radiomics model and a clinical-radiomics combined model for predicting clinically significant prostate cancer (PCa). Methods: In total, 381 patients with clinically suspicious PCa were included in this retrospective study; of those, 199 patients did not have PCa upon biopsy, while 182 patients had PCa. All patients underwent 3.0-T MRI examinations with the same acquisition parameters, and clinical risk factors associated with PCa (age, prostate volume, serum PSA, etc.) were collected. We randomly stratified the training and test sets using a 6:4 ratio. The radiomic features included gradient-based histogram features, grey-level co-occurrence matrix (GLCM), run-length matrix (RLM), and grey-level size zone matrix (GLSZM). Three models were developed using multivariate logistic regression analysis to predict clinically significant PCa: a clinical model, a radiomics model and a clinical-radiomics combined model. The diagnostic performance and clinical net benefit of each model were compared via receiver operating characteristic (ROC) curve analysis and decision curves, respectively. Results: Both the radiomics model (AUC: 0.98) and the clinical-radiomics combined model (AUC: 0.98) achieved greater predictive efficacy than the clinical model (AUC: 0.79). The decision curve analysis also showed that the radiomics model and combined model had higher net benefits than the clinical model. Conclusions: Compared with the evaluation of clinical risk factors associated with PCa only, the radiomics-based machine learning model can improve the predictive accuracy for clinically significant PCa, in terms of both diagnostic performance and clinical net benefit.