BMC cancer · 2025
Biparametric MRI-based radiomics for prediction of clinically significant prostate cancer of PI-RADS category 3 lesions.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Lu Feng, Zhao Yanjun, Wang Zhongjuan, Feng Ninghan
Аннотация
PURPOSE: We aimed to investigate the diagnostic performance of biparametric MRI (bpMRI)-based radiomics in differentiating clinically significant prostate cancer (csPCa) among lesions categorized as Prostate Imaging Reporting and Data System (PI-RADS) score 3.
METHOD: Between September 2020 and October 2023, a total of 233 patients with PI-RADS category 3 lesions were identified, which were divided into training cohort (n = 160) and validation cohort (n = 73). Radiomics features were extracted from T2-weighted imaging (T2) and diffusion-weighted imaging (DWI) for csPCa prediction. The least absolute shrinkage and selection operator (LASSO) regression algorithm was used to select the most useful radiomics features. Diagnostic performance was compared using the area under the receiver operating characteristic (ROC) curve (AUC).
RESULTS: 34 robust radiomics features (incorporating 12 features from T2 and 22 features from DWI) were selected to construct the final radiomics signature. In the training group, the AUCs for prostate-specific antigen density (PSAD), radiomics, and combination were 0.658 (95% CI 0.550-0.766), 0.858 (95% CI 0.779-0.936), and 0.887 (95% CI 0.814-0.959), respectively, in the validation group were 0.690 (95% CI 0.524-0.855), 0.810 (95% CI 0.682-0.937), and 0.856 (95% CI 0.750-0.962). The combination model integrating radiomics and PSAD showed a significant improvement in diagnostic performance as compared to using these two parameters alone either in the training group (P < 0.001 and P = 0.024) or in the validation group (P = 0.024 and P = 0.048).
CONCLUSION: BpMRI-based radiomics had high diagnostic performance in predicting csPCa among PI-RADS 3 lesions, and combining it with PSAD could further improve the overall accuracy.
METHOD: Between September 2020 and October 2023, a total of 233 patients with PI-RADS category 3 lesions were identified, which were divided into training cohort (n = 160) and validation cohort (n = 73). Radiomics features were extracted from T2-weighted imaging (T2) and diffusion-weighted imaging (DWI) for csPCa prediction. The least absolute shrinkage and selection operator (LASSO) regression algorithm was used to select the most useful radiomics features. Diagnostic performance was compared using the area under the receiver operating characteristic (ROC) curve (AUC).
RESULTS: 34 robust radiomics features (incorporating 12 features from T2 and 22 features from DWI) were selected to construct the final radiomics signature. In the training group, the AUCs for prostate-specific antigen density (PSAD), radiomics, and combination were 0.658 (95% CI 0.550-0.766), 0.858 (95% CI 0.779-0.936), and 0.887 (95% CI 0.814-0.959), respectively, in the validation group were 0.690 (95% CI 0.524-0.855), 0.810 (95% CI 0.682-0.937), and 0.856 (95% CI 0.750-0.962). The combination model integrating radiomics and PSAD showed a significant improvement in diagnostic performance as compared to using these two parameters alone either in the training group (P < 0.001 and P = 0.024) or in the validation group (P = 0.024 and P = 0.048).
CONCLUSION: BpMRI-based radiomics had high diagnostic performance in predicting csPCa among PI-RADS 3 lesions, and combining it with PSAD could further improve the overall accuracy.