BMC Medical Imaging最新文献

{"title":"Correction: ES-UNet: efficient 3D medical image segmentation with enhanced skip connections in 3D UNet.","authors":"Minyoung Park, Seungtaek Oh, Junyoung Park, Taikyeong Jeong, Sungwook Yu","doi":"10.1186/s12880-025-01891-y","DOIUrl":"10.1186/s12880-025-01891-y","url":null,"abstract":"","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"365"},"PeriodicalIF":3.2,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12403535/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941958","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":"Sliding volume-based streak artifact reduction network (S-STAR Net) for ultra-sparse-view computed tomography.","authors":"Shiang Zhang, Yibo Hu, Ziheng Deng, Yujie Wang, Jun Zhao, Jianqi Sun","doi":"10.1186/s12880-025-01910-y","DOIUrl":"10.1186/s12880-025-01910-y","url":null,"abstract":"","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"364"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12403527/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941757","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":"CT-based deep learning radiomics model for predicting proliferative hepatocellular carcinoma: application in transarterial chemoembolization and radiofrequency ablation.","authors":"Hengtao Zhang, Zhaogang Zhang, Kun Zhang, Zhongsong Gao, Zhiwei Shen, Wen Shen","doi":"10.1186/s12880-025-01913-9","DOIUrl":"10.1186/s12880-025-01913-9","url":null,"abstract":"","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"363"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12403943/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941976","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":"Evaluating machine learning models for post-surgery treatment response assessment in glioblastoma multiforme: a comparative study of gray level co-occurrence matrix (GLCM), curvelet, and combined radiomics features selected by multiple algorithms.","authors":"Sanaz Alibabaei, Mohammad Yousefipour, Masoumeh Rahmani, Samira Raminfard, Marziyeh Tahmasbi","doi":"10.1186/s12880-025-01906-8","DOIUrl":"10.1186/s12880-025-01906-8","url":null,"abstract":"<p><strong>Background: </strong>Developing quantitative methods to assess post-surgery treatment response in Glioblastoma Multiforme (GBM) is critical for improving patient outcomes and refining current subjective approaches. This study analyzes the performance of machine learning models trained on radiomic datasets derived from magnetic resonance imaging (MRI) scans of GBM patients.</p><p><strong>Methods: </strong>MRI scans from 143 GBM patients receiving adjuvant therapy post-surgery were acquired and preprocessed. A total of 92 radiomic features, including 68 Gy-level co-occurrence matrix (GLCM)-based features calculated in four directions (0°, 45°, 90°, and 135°) and 24 Curvelet coefficient-based features, were extracted from each patient's segmented tumor cavity. Machine learning classifiers, including Support Vector Machine (SVM), Random Forest, K-Nearest Neighbors (KNN), AdaBoost, CatBoost, LightGBM, XGBoost, Gaussian Naïve Bayes (GNB), and Logistic Regression (LR), were trained on the extracted radiomics selected using sequential feature selection, LASSO, and PCA. Validation was performed with 10-fold cross-validation.</p><p><strong>Results: </strong>The proposed pipeline achieved an accuracy of 87% in classifying post-surgery treatment responses in GBM patients. This accuracy was achieved with the SVM trained on a combination of GLCM and Curvelet-based radiomics selected via forward sequential algorithm-8, and with KNN trained on GLCM and Curvelet radiomics combination selected using LASSO (alpha = 0.01). The LR model trained on Curvelet-based LASSO-selected radiomics (alpha = 0.01) also showed strong performance.</p><p><strong>Conclusion: </strong>The results demonstrate that MRI-based radiomics, specifically GLCM and Curvelet features, can effectively train machine learning models to quantitatively assess GBM treatment response. These models serve as valuable tools to complement qualitative evaluations, enhancing accuracy and objectivity in post-surgery outcome assessment.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"362"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12403642/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941636","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":"Deep learning model for predicting lymph node metastasis around rectal cancer based on rectal tumor core area and mesangial imaging features.","authors":"Lili Guo, Kuang Fu, Wenjia Wang, Li Zhou, Lu Chen, Miaomiao Jiang","doi":"10.1186/s12880-025-01878-9","DOIUrl":"10.1186/s12880-025-01878-9","url":null,"abstract":"","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"361"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12400577/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941105","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":"Diagnostic values of diffusion-weighted imaging and dynamic contrast-enhanced MRI in the pathological grading of adenoid cystic carcinoma.","authors":"Peng An, Nan Jiang, Jinyan Li, Wei Li, Kun Zhou, Jiaxiang Xin, Peihang Jing, Lixin Sun","doi":"10.1186/s12880-025-01898-5","DOIUrl":"https://doi.org/10.1186/s12880-025-01898-5","url":null,"abstract":"<p><strong>Objectives: </strong>The combination of dynamic-contrast-enhanced-magnetic-resonance-imaging (DCE-MRI) with the apparent-diffusion-coefficient (ADC) holds significant value for predicting tumor pathological outcomes. This study pioneers the combined application of DCE-MRI and ADC parameters to evaluate their utility in differentiating histopathological grades of adenoid cystic carcinoma (ACC).</p><p><strong>Methods: </strong>Retrospective diagnosis of 23 ear and temporal ACC patients was confirmed based on surgical pathology from March 2020 to April 2024. All patients underwent routine MRI, DWI, and DCE-MRI scans within one week before surgery. The lesion ADC values and DCE-MRI perfusion parameters, including Ve, Kep, Ktrans, and iAUC, were measured. Consistency tests were conducted on the measurements done by two physicians. The ADC values and DCE-MRI perfusion parameters between different pathological grades were compared. The correlation among all parameters and ACC pathological grading were analyzed, and the diagnostic accuracy of each parameter was assessed using receiver-operating-characteristic (ROC) curves.</p><p><strong>Results: </strong>The measurements from the two observers showed high consistency (ICC > 0.9). Ktrans, iAUC, and ADC values demonstrated significant differences between different pathological grades (P < .01, P < .05, P < .05). Correlation analysis indicated that Ktrans, Kep, Ve, and iAUC were positively correlated with ACC pathological grading, and Ktrans demonstrated the most robust correlation (correlation coefficient r = .578, P < .01). In contrast, ADC values were markedly and negatively correlated with pathological grading (r=-.470, P < .05). In ROC analysis, the area-under-the-curve (AUC) for Ktrans, iAUC, and ADC were 0.841, 0.790, and 0.778, respectively, all higher than those for Kep and Ve. Ktrans showed the best diagnostic performance.</p><p><strong>Conclusion: </strong>Combining DCE-MRI perfusion parameters with ADC values provides a non-invasive and effective method for preoperative grading of ACC, with Ktrans, iAUC, and ADC showing strong diagnostic potential. These findings support more accurate tumor characterization and personalized treatment planning, warranting further validation in larger prospective studies.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"359"},"PeriodicalIF":3.2,"publicationDate":"2025-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12398149/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941590","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":"MSFE-GallNet-X: a multi-scale feature extraction-based CNN Model for gallbladder disease analysis with enhanced explainability.","authors":"Hadiur Rahman Nabil, Istyak Ahmed, Aritra Das, M F Mridha, Md Mohsin Kabir, Zeyar Aung","doi":"10.1186/s12880-025-01902-y","DOIUrl":"https://doi.org/10.1186/s12880-025-01902-y","url":null,"abstract":"","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"360"},"PeriodicalIF":3.2,"publicationDate":"2025-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12399013/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941667","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":"Association between perivascular fat density on CT angiography and abdominal aortic aneurysm progression.","authors":"Shuai Zhang, Na Chang, Xinxin Yu, Bing Kang, Ru Tan","doi":"10.1186/s12880-025-01895-8","DOIUrl":"https://doi.org/10.1186/s12880-025-01895-8","url":null,"abstract":"<p><strong>Background: </strong>Perivascular adipose tissue has been shown to play a role in cardiovascular disease. This provides evidences that perivascular fat density (PFD) may have a correlation with abdominal aortic aneurysm (AAA). The aim of study was to investigate the association between PFD on computed tomography angiography (CTA) and AAA expanding rate.</p><p><strong>Methods: </strong>A total of 144 patients with AAA who underwent at least two computed tomography angiography (CTA) examinations at intervals of ≥ 6 months between January 2014 and June 2023 were included. The patients were divided into slowly and rapidly expanding AAA groups according to the median value of AAA expansion rate. The clinical and CTA characteristics of both groups were compared. The relationships between AAA diameter, AAA volume, expansion rate, and PFD were tested using the pearson coefficient.</p><p><strong>Results: </strong>Compared with the slowly expanding group, patients with rapidly expanding AAA had a significantly higher prevalence of hypertension (77.8% versus 55.6%; P = 0.005), a significantly lower prevalence of diabetes (26.4% versus 47.2%; P < 0.010), and a higher PFD at baseline (-72.6 ± 9.7 HU vs. -81.2 ± 7.9 HU; P < 0.001). In the whole group, slowly expanding group, and rapidly expanding group, PFD at baseline was positively correlated with AAA diameter at baseline (P < 0.001), AAA volume at baseline (P < 0.001), and expansion rate (P < 0.001). A positive correlation was observed between increasing PFD and expansion rate (P < 0.05).</p><p><strong>Conclusions: </strong>A higher PFD on CTA was found to be related to a rapidly expanding AAA. Thus, PFD may become a non-invasive and potential image marker for predicting and treating AAA progression.</p><p><strong>Clinical trial number: </strong>Not applicable. This research is a retrospective analysis.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"357"},"PeriodicalIF":3.2,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12395999/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941927","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":"The value of contrast-enhanced ultrasound in the follow-up of Bosniak IIF cystic renal lesions.","authors":"Ákos Járay, Péter István Farkas, Dávid Semjén, Bálint Botz","doi":"10.1186/s12880-025-01905-9","DOIUrl":"https://doi.org/10.1186/s12880-025-01905-9","url":null,"abstract":"<p><strong>Background: </strong>Contrast-enhanced ultrasound (CEUS) is increasingly used in the characterization of cystic renal lesions. Bosniak IIF lesions warrant follow-up, and their reported progression rate remains variable.</p><p><strong>Methods: </strong>In this single-center retrospective study we assessed renal CEUS exams (SonoVue^®) with a diagnosis of Bosniak IIF lesion, conducted between 2015 and 2020. 56 patients (59 lesions) met inclusion criteria. Patient demographics, lesion morphology, follow-up adherence, and outcomes were evaluated.</p><p><strong>Results: </strong>Significant (p = 0.037) positive correlation was found between patient age and lesion size. 33.9% of patients were immediately lost to follow-up, and they tended to be younger, albeit not significantly (p = 0.09). Recommendation for follow-up imaging was indicated in 66.1% of the initial radiological reports. Follow-up adherence was not significantly lower for lesions with absent recommendation (55% vs. 70.27%, p = 0.26). Fewer (52%) female vs. male (74.19%) patients had a follow-up recommendation (p = 0.1, not significant). 10.8% of the followed lesions demonstrated progression within 5 years. Lesion reevaluation according to the 2020 European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB) criteria resulted in 77.36% agreement with 22.2% of lesions being downgraded, and a single lesion being upgraded (p = 0.0015, significant).</p><p><strong>Conclusions: </strong>Follow-up adherence of Bosniak IIF cystic renal lesions was found to be suboptimal, with potential gender disparities. Standardization of follow-up recommendation in the report is an unmet need. Progression rate remains low, but is expected to change with the adoption of novel CEUS-specific criteria. The EFSUMB criteria in particular can improve selection of truly indeterminate lesions.</p><p><strong>Trial registration: </strong>Retrospectively registered.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"358"},"PeriodicalIF":3.2,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12395875/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941695","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":"Hydration-induced hydronephrosis in healthy adults: a diagnostic pitfall in renal ultrasound imaging.","authors":"Liqa A Rousan, Khaled J Zaitoun, Sofyan Freihat, Mohammad A Albatayneh, Abdel Rahman M Al Serhan, Aghyad Alsalkhadi, Saad Tayyem, Ali M Abdel Kareem","doi":"10.1186/s12880-025-01900-0","DOIUrl":"https://doi.org/10.1186/s12880-025-01900-0","url":null,"abstract":"","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"354"},"PeriodicalIF":3.2,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12381993/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144941655","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}