Abdominal Radiology最新文献

{"title":"Accurate colorectal cancer detection using a random hinge exponential distribution coupled attention network on pathological images.","authors":"E Bharath, R Vimal Raja, K Kalaivanan, Vivek Deshpande","doi":"10.1007/s00261-024-04770-2","DOIUrl":"10.1007/s00261-024-04770-2","url":null,"abstract":"<p><p>Colorectal cancer (CRC) is one of the most common and deadly forms of cancer worldwide, necessitating accurate and early detection to improve treatment outcomes. Traditional diagnostic methods often rely on manual examination of pathological images, which can be time-consuming and prone to human error. This study presents an advanced approach for colorectal cancer detection using a Random Hinge Exponential Distribution coupled Attention Network (RHED-CANet) on pathological images. The input dataset is sourced from the TCGA-CRC-DX cohort and the CRC dataset, both widely recognized for their comprehensive coverage of colorectal cancer cases. Pre-processing and feature extraction are performed using a Modified Square Root Sage-Husa Adaptive Kalman Filter combined with a Spike-Driven Transformer, enhancing noise reduction and feature clarity. Segmentation is achieved through an EfficientNetV2L Inception Transformer, ensuring precise delineation of cancerous regions. The final classification utilizes the RHED-CANet, a network tailored to handle the complexities of pathological data with high accuracy. This methodology achieved remarkable results, with an accuracy of 99.9% and a precision of 99.7%. These performance metrics underscore the method's ability to minimize false positives and enhance diagnostic accuracy. The proposed approach offers significant advantages, including a reduction in diagnostic time and a substantial improvement in detection accuracy, making it a promising tool for clinical applications. Despite its excellent accuracy, the suggested RHED-CANet technique has drawbacks, such as overfitting the TCGA-CRC-DX and CRC datasets by reducing generalizability on other datasets comprising other cancer types or image qualities. The actual application of the techniques in real-time clinical applications may be hampered by this computational load, especially in settings with limited resources, and the model's potential computational complexity due to multiple advanced processing steps. Additionally, the efficiency of training may be impacted by biased inputs, particularly for minor CRC subtypes.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"2828-2857"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142942364","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of hemodynamic changes in gastric varices using CT findings before and after vascular plug-assisted retrograde transvenous obliteration (PARTO) and evaluation of treatment outcomes.","authors":"Dong Kyu Lee, Jung Won Kwak, Sung Bum Cho","doi":"10.1007/s00261-024-04777-9","DOIUrl":"10.1007/s00261-024-04777-9","url":null,"abstract":"<p><strong>Purpose: </strong>To assess the treatment outcomes of vascular plug-assisted retrograde transvenous obliteration (PARTO) for gastric varices based on hemodynamic changes observed in pre- and post-procedural CT scans.</p><p><strong>Methods: </strong>This single-center study included 43 patients with gastric varices who underwent PARTO between June 2017 and December 2023. Peri-procedural CT and endoscopic findings were retrospectively reviewed to analyze the afferent and draining veins of gastric varices, identify any residual varices or afferent veins after treatment, assess the technical and clinical successes, and determine statistically significant variables associated with clinical success.</p><p><strong>Results: </strong>In the pre-procedural CT, the most common dominant afferent vein was the posterior gastric vein (81.4%), followed by the short gastric vein (46.5%) and the left gastric vein (37.2%). Post-procedural CT scans were available for 34 patients (34/43), revealing residual varices and afferent veins in 10 patients (10/34). Seven patients had unknown clinical outcomes due to insufficient follow-up data. The technical and clinical success rates were 93.0% (40/43) and 77.8% (28/36), respectively. In the multivariable logistic regression model, the dominant left gastric vein observed in the pre-procedural CT was a significant negative predictor of clinical success (odds ratio, 0.007; P = 0.027). The sum of the diameters of all afferent veins was also a significant negative predictor (odds ratio, 0.708; P = 0.044).</p><p><strong>Conclusion: </strong>A dominant left gastric vein and a larger sum of the diameters of all afferent veins observed in the pre-procedural CT may be associated with an increased risk of clinical failure of PARTO. Therefore, pretherapeutic CT evaluation of the hemodynamics of gastric varices, particularly the type, dominance, and diameter of the afferent veins, could be beneficial for achieving successful PARTO.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"3148-3158"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142871046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Imaging and management of complications post biliary-enteric anastomosis.","authors":"Wyanne Law, Jordan Swensson, Mackenzie Mayhew, Victor Zaydfudim, Rachita Khot","doi":"10.1007/s00261-024-04705-x","DOIUrl":"10.1007/s00261-024-04705-x","url":null,"abstract":"<p><p>Biliary-enteric anastomosis is a common surgical procedure for benign and malignant pathologies involving bile ducts, pancreas and duodenum, as well as during liver transplantation. Imaging is key in detecting potential complications. Ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), and nuclear scintigraphy provide complementary information. Ultrasound offers real-time assessment of bile duct dilation and fluid collection. CT scan, due to its wide availability, is often performed first and provides detailed cross-sectional anatomy. MRI, including MR cholangiography, excels in visualizing bile ducts and detecting subtle changes in anastomosis integrity. Common complications of BEA include bile leak, biliary anastomotic stricture, and cholangitis, each presenting with distinct imaging features. Effective imaging allows for early detection and management of these complications, improving patient outcomes. This review discusses the role of imaging in assessing post-BEA complications and emphasizes the importance of multimodal imaging approaches in the comprehensive evaluation of BEA and its complications.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"3036-3048"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142930339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Visibility of mpMRI region of interest on ultrasound during cognitive fusion targeted biopsy predicts prostate cancer detection: a prospective single-center study.","authors":"Fei Qin, Zhijian Liu, Jianguo Ma, Jingyun Wu, Qi Shen, Yi Liu, Xuesong Li","doi":"10.1007/s00261-024-04750-6","DOIUrl":"10.1007/s00261-024-04750-6","url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to evaluate the nature of ultrasound characteristics during mpMRI/TRUS cognitive fusion targeted biopsy (cTB).</p><p><strong>Methods: </strong>From 2023 to 2024, data from 502 lesions in 426 men who underwent targeted combined systematic biopsy were analyzed. All lesions had a Prostate Imaging Reporting and Data System (PI-RADS) score of ≥ 3. The primary endpoint was the detection rate of prostate cancer (PCa) according to the PI-RADS score/ultrasound characteristics, categorized as benign or invisible (Bi), hypoechoic only (Ho), and hypoechoic with microcalcification (Hm), assessed through cross-stratification. The secondary endpoints included the distribution of ultrasound characteristics across PI-RADS scores, prostate zones, and histological types. Finally, associations between ultrasound characteristics and clinically significant PCa (csPCa) were assessed using multivariate logistic regression analysis (MVA).</p><p><strong>Results: </strong>Among lesions, 233 (46%) were Bi, 210 (42%) Ho, and 59 (12%) Hm. First, Bi lesions had a 64% (103/161) non-cancer rate in PI-RADS 3, while Ho + Hm lesions showed the highest csPCa rate in PI-RADS 5 at 82% (102/124). Additionally, Ho + Hm lesions were predominantly observed in PI-RADS 5 (92% [114/124]) and in the peripheral zone (64% [179/278]). Notably, Hm lesions had a significantly higher percentage of cribriform morphology than Ho lesions (32% vs. 14%, P = 0.001). Finally, MVA confirmed Ho ([Ref Bi] OR 4.95, P < 0.001) and Hm ([Ref Bi] OR 27.7, P < 0.001) as independent predictors of csPCa.</p><p><strong>Conclusion: </strong>In cTB, the identification of Ho and Hm lesions on TRUS enhances the diagnostic yield of csPCa by facilitating more precise localization compared to Bi lesions.</p><p><strong>Clinical trial registration: </strong>No. 2023-272-002, July 14, 2023.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"3305-3312"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12181095/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142875878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Different trend of muscle wasting extracted from computed tomography in patients with severe acute pancreatitis.","authors":"Shengqi Wang, Meiping Wang, Li Jiang","doi":"10.1007/s00261-024-04741-7","DOIUrl":"10.1007/s00261-024-04741-7","url":null,"abstract":"<p><strong>Purpose: </strong>The trend of muscle wasting in patients with acute severe and moderately severe pancreatitis (AP) remains unclear. This retrospective study aimed to investigate the trend of skeletal muscle area (SMA) changes and its impact on patients with severe and moderately severe AP.</p><p><strong>Methods: </strong>Patients diagnosed with AP who had repeated CT scans after intensive care unit (ICU) admission were included. The patients were categorized into moderately severe AP or severe AP groups. The generalized additive mixed model (GAMM) was used to analyze the SMA trajectories.</p><p><strong>Results: </strong>A total of 126 patients were included. The patients in the severe AP group had more rapid muscle wasting during the first 3 weeks following ICU admission. The SMA decreased by 1.1 cm² (95% CI: 1.3 to 0.8) per day in the severe AP group, while the SMA decreased by 0.5 cm² (95% CI: 0.6 to 0.4) in the moderately severe AP group in the GAMM model. A larger change in the SMA during the first 10 days after admission was significantly associated with prolonged length of hospital stay (LOS) (β = - 0.205, P = 0.036).</p><p><strong>Conclusions: </strong>Patients with severe AP experienced more muscle wasting during the first 3 weeks after ICU admission. A larger reduction in the SMA was associated with prolonged LOS.</p><p><strong>Clinical implications: </strong>Different patterns of muscle wasting were present during the first 3 weeks after ICU admission in moderately severe and severe AP patients. Accordingly, different nutrition and rehabilitation strategies might be employed depending upon the severity of AP.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"3178-3186"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142833472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"International perspectives on LI-RADS.","authors":"Andrea S Kierans, Diego A Aguirre, Sonal Krishan, Jeong Min Lee, Maxime Ronot, Jin Wang, Elizabeth M Hecht","doi":"10.1007/s00261-024-04729-3","DOIUrl":"10.1007/s00261-024-04729-3","url":null,"abstract":"<p><p>Given the crucial role of imaging in HCC diagnosis, LI-RADS CT/MRI was developed to standardize the imaging interpretation and reporting of HCC in patients at risk for HCC and categorize hepatic observations on an ordinal scale according to the likelihood of HCC. LI-RADS has since been expanded to include 5 algorithms: LI-RADS US Surveillance, contrast-enhanced US (CEUS) LI-RADS, LI-RADS CT/MRI, and LI-RADS Treatment Response Assessment. LI-RADS has been adopted broadly in North America, however with less ubiquitous adoption outside of North America. Further elucidation of the perceived strengths and weakness of the LI-RADS algorithm, as it pertains to various geographic regions, will continue to inform a future system that may be more readily adopted globally. Therefore, the aim of this article is to summarize HCC risk factors and imaging guidelines in select geographically disparate regions, and to solicit feedback from liver imaging experts on the limitations and barriers to adoption of LI-RADS algorithms in their patient populations.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"2917-2927"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142852083","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Precise vesical wall staging of bladder cancer in the era of precision medicine: has it been fulfilled?","authors":"Mohamed Ragab Nouh, Omnia Ezz Eldin","doi":"10.1007/s00261-024-04786-8","DOIUrl":"10.1007/s00261-024-04786-8","url":null,"abstract":"<p><p>Urinary bladder cancer is a global disease that poses medical and socioeconomic challenges to patients and healthcare systems. Predicting detrusor invasiveness and pathological grade of bladder cancer by the radiologist is imperative for informed decision-making and effective patient-tailored therapy. Cystoscopy and TURBT are the current gold standard for preoperative histologic diagnosis and local pathological staging but are compromised by their intrusiveness, under-sampling, and staging inaccuracies. Over the last few decades, incredible imaging technology advancements have enabled radiologists to progress in these grading and staging tasks. MRI has become widely accepted as a noninvasive alternative. It supplements morphologic data with functional insights into the tumor microenvironment, enhancing tumor characterization and predicting the detrusor's histologic grade and invasiveness status. Radiomics is a promising field that helps radiologists achieve higher accuracies in bladder cancer staging, re-staging, and direct treating teams to potential management readjustments. Such knowledge leaps hold promise for personalized management of bladder cancer in a precision medicine era.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"3084-3091"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142891299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preoperative clinical and contrasted-enhanced CT features to predict perineural invasion in gallbladder carcinoma: focus on clinical T3-4 stage.","authors":"Lu Chen, Yang Zhou, Xun Xu, Hui Zhang, Xuan Xiao, Chang-Xian Li, Wei You, Hai-Bin Shi, Xi-Sheng Liu, Fei-Yun Wu, Xiang-Cheng Li, Fei-Peng Zhu","doi":"10.1007/s00261-024-04782-y","DOIUrl":"10.1007/s00261-024-04782-y","url":null,"abstract":"<p><strong>Purpose: </strong>To investigate the utility of combining clinical and contrasted-enhanced tomography (CECT) parameters for the preoperative evaluation of perineural invasion (PNI) in gallbladder carcinoma (GBC).</p><p><strong>Methods: </strong>A total of 134 patients with GBC (male/female, 52/82; age, 64.4 ± 9.7 years) were divided into PNI-positive (n = 63) and PNI-negative groups (n = 71). Clinical characteristics (demographic information, liver function indicators and tumor markers) and CECT parameters (tumor type, tumor size, gallbladder stone, invasion of gallbladder neck/cystic duct, clinical T stage and N stage) were collected and compared between two groups. Binary logistic regression analysis, receiver operating characteristic curves analyses and Delong test were used in further statistical analyses in clinical T3-4 stage (cT3-4) GBC patients. Overall survival (OS) rates after surgery were compared between PNI-negative group and PNI-positive group of cT3-4 GBC patients.</p><p><strong>Results: </strong>The majority of GBC patients with PNI were classified as cT3-4 (61/63, 96.8%), while only 3.2% (2/63) of PNI-positive cases were identified at cT1-2. Among cT3-4 GBC, OS was significantly lower in the PNI-positive group than the PNI-negative group after surgery (HR,1.661; 95% CI, 1.044-2.643; P = 0.032). Gender and gallbladder neck/cystic duct invasion were independent predictive factors for cT3-4 GBC patients with PNI. A combination of gender and gallbladder neck/cystic duct invasion showed the best diagnostic performance than that of individual parameters (all P < 0.05).</p><p><strong>Conclusions: </strong>Preoperative T staging using CECT enables the initial assessment of PNI status in GBC patients. A combination of gender and gallbladder neck/cystic duct invasion may effectively predict PNI in GBC, particularly in cT3-4 GBC.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"2989-2998"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142891303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel artificial intelligence segmentation model for early diagnosis of bladder tumors.","authors":"Lu Li, Lingxiao Jiang, Kun Yang, Bin Luo, Xinghuan Wang","doi":"10.1007/s00261-024-04715-9","DOIUrl":"10.1007/s00261-024-04715-9","url":null,"abstract":"<p><strong>Objective: </strong>Despite cystoscopy plays an important role in bladder tumors diagnosis, it often falls short in flat cancerous tissue and minuscule satellite lesions. It can easily lead to a missed diagnosis by the urologist, which can lead to a swift tumor regrowth following transurethral resection of the bladder tumor (TURBT). Therefore, we developed a deep learning-based intelligent diagnosis system for early bladder cancer to improve the identification rate of early bladder tumors.</p><p><strong>Methods: </strong>Video data from 273 bladder cancer patients who underwent TURBT at Zhongnan Hospital were collected. The dataset was carefully annotated by urologists to clearly define tumor boundaries. Subsequently, we developed a new bladder tumor segmentation network (BTS-Net) based on transformer to accurately diagnose early-stage bladder cancer lesions.</p><p><strong>Results: </strong>Our experiments demonstrate that the BTS-Net we developed has outperformed other method on the external B validation dataset, achieving a MPrecision of 91.39%, a MRecall of 95.71%, a MIoU of 88.18% and an F1-score of 93.18%. The BTS-Net showed high accuracy with real-time processing speed at 23 fps.</p><p><strong>Conclusion: </strong>Missed detection of satellite lesions in early bladder tumors often leads to tumor recurrence. Our BTS-Net is capable of segmenting all potential satellite lesions in surgical videos, without the need for complex professional equipment. This AI-assisted diagnosis system has the potential to improve surgical outcomes by ensuring comprehensive treatment of all tumor-related areas during TURBT.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":"3092-3099"},"PeriodicalIF":2.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142908732","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A computed tomography-based radiomics prediction model for BRAF mutation status in colorectal cancer.","authors":"Boqi Zhou, Huaqing Tan, Yuxuan Wang, Bin Huang, Zhijie Wang, Shihui Zhang, Xiaobo Zhu, Zhan Wang, Junlin Zhou, Yuntai Cao","doi":"10.1007/s00261-025-04983-z","DOIUrl":"https://doi.org/10.1007/s00261-025-04983-z","url":null,"abstract":"<p><strong>Purpose: </strong>The aim of this study was to develop and validate CT venous phase image-based radiomics to predict BRAF gene mutation status in preoperative colorectal cancer patients.</p><p><strong>Methods: </strong>In this study, 301 patients with pathologically confirmed colorectal cancer were retrospectively enrolled, comprising 225 from Centre I (73 mutant and 152 wild-type) and 76 from Centre II (36 mutant and 40 wild-type). The Centre I cohort was randomly divided into a training set (n = 158) and an internal validation set (n = 67) in a 7:3 ratio, while Centre II served as an independent external validation set (n = 76). The whole tumor region of interest was segmented, and radiomics characteristics were extracted. To explore whether tumor expansion could improve the performance of the study objectives, the tumor contour was extended by 3 mm in this study. Finally, a t-test, Pearson correlation, and LASSO regression were used to screen out features strongly associated with BRAF mutations. Based on these features, six classifiers-Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), K-Nearest Neighbors (KNN), and Extreme Gradient Boosting (XGBoost)-were constructed. The model performance and clinical utility were evaluated using receiver operating characteristic (ROC) curves, decision curve analysis, accuracy, sensitivity, and specificity.</p><p><strong>Results: </strong>Gender was an independent predictor of BRAF mutations. The unexpanded RF model, constructed using 11 imaging histologic features, demonstrated the best predictive performance. For the training cohort, it achieved an AUC of 0.814 (95% CI 0.732-0.895), an accuracy of 0.810, and a sensitivity of 0.620. For the internal validation cohort, it achieved an AUC of 0.798 (95% CI 0.690-0.907), an accuracy of 0.761, and a sensitivity of 0.609. For the external validation cohort, it achieved an AUC of 0.737 (95% CI 0.616-0.847), an accuracy of 0.658, and a sensitivity of 0.667.</p><p><strong>Conclusions: </strong>A machine learning model based on CT radiomics can effectively predict BRAF mutations in patients with colorectal cancer. The unexpanded RF model demonstrated optimal predictive performance.</p>","PeriodicalId":7126,"journal":{"name":"Abdominal Radiology","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144075235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}