BMC Medical Imaging最新文献

{"title":"MSA-Net: multiple self-attention mechanism for 3D lung nodule classification in CT images.","authors":"Jiating Pan, Lishi Liang, Peng Sun, Yongbo Liang, Jianming Zhu, Zhencheng Chen","doi":"10.1186/s12880-025-01725-x","DOIUrl":"10.1186/s12880-025-01725-x","url":null,"abstract":"<p><strong>Purpose: </strong>Lung cancer is a life-threatening disease that poses a significant risk to human health. Accurate differentiation between benign and malignant lung nodules, based on computed tomography (CT), is crucial to assess lung health. Developing an automated computer-aided diagnostic method for this differentiation is essential. We introduced a streamlined 3D model structure to solve the problems of 2D models cannot extract spatial information effectively and 3D models have high complexity and large occupation of computing resources.</p><p><strong>Methods: </strong>We proposed an MSA (multiple self-attention-based) model to address the limitations of 2D models in extracting spatial information effectively and the high complexity associated with 3D models. Our approach introduced the 3D RTConvBlock, which employed multiple self-attention mechanisms for the extraction of spatial features. This enabled the extraction of specific spatial feature information by combining local features, global information, and dependencies between features.</p><p><strong>Results: </strong>The MSA model demonstrates exceptional performance with an accuracy of 0.953, a sensitivity of 0.963, and an AUC (area under curve) of 0.993 in the LUNA16 dataset, which is higher than state-of-the-art methods. Compared with existing 2D models, we extract spatial information features better, resulting in higher accuracy.</p><p><strong>Conclusion: </strong>These results have significant implications for enhancing the accuracy and reliability of lung nodule classification, providing robust auxiliary support for physicians diagnosing lung diseases.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"193"},"PeriodicalIF":2.9,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12117915/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144156824","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":"Magnetic resonance imaging features of intrahepatic bile duct adenoma: a 10-year retrospective study.","authors":"Ge Sun, Jiamei Yao, Huai Chen, Mengsu Zeng, Mingliang Wang","doi":"10.1186/s12880-025-01733-x","DOIUrl":"10.1186/s12880-025-01733-x","url":null,"abstract":"<p><strong>Background: </strong>Intrahepatic bile duct adenoma (BDA) is a rare tumor with limited understanding of its magnetic resonance imaging (MRI) features and clinical characteristics. This study aimed to analyze the MRI characteristics of BDA.</p><p><strong>Methods: </strong>This retrospective study analyzed MRI findings and clinical profiles of 33 patients diagnosed with bile duct adenomas (BDA) at Zhongshan Hospital Fudan University from January 2014 to January 2024. MRI features and clinical data were reviewed and analyzed.</p><p><strong>Results: </strong>A total of 36 lesions were identified among 33 patients, with 31 cases presenting as solitary lesions. The average diameter was 9.2 ± 3.1 mm, predominantly subcapsular, located near the liver capsule, with the majority exhibiting well-defined margins. On T1-weighted imaging (T1WI), lesions displayed hypointensity, while T2-weighted imaging (T2WI) was slightly hypointense in most cases, enhancing the visibility of the lesions. Apparent diffusion coefficient (ADC) values averaged (1.93 ± 0.51)×10⁻³ mm²/s, significantly higher than surrounding liver tissue (P < 0.001), suggesting unique tissue properties. Notably, BDA, as a hypervascular tumor, displayed rim and non-rim enhancement patterns, along with a tendency for persistent enhancement.</p><p><strong>Conclusion: </strong>The MRI features of BDA included small lesions near the liver capsule, characterized by distinct morphology and enhancement patterns, alongside elevated ADC values that distinguish them from malignant hepatic lesions. The findings emphasize the importance of MRI in the accurate diagnosis and management of BDA.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"188"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12107894/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149090","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":"Detecting microcephaly and macrocephaly from ultrasound images using artificial intelligence.","authors":"Abraham Keffale Mengistu, Bayou Tilahun Assaye, Addisu Baye Flatie, Zewdie Mossie","doi":"10.1186/s12880-025-01709-x","DOIUrl":"10.1186/s12880-025-01709-x","url":null,"abstract":"<p><strong>Background: </strong>Microcephaly and macrocephaly, which are abnormal congenital markers, are associated with developmental and neurologic deficits. Hence, there is a medically imperative need to conduct ultrasound imaging early on. However, resource-limited countries such as Ethiopia are confronted with inadequacies such that access to trained personnel and diagnostic machines inhibits the exact and continuous diagnosis from being met.</p><p><strong>Objective: </strong>This study aims to develop a fetal head abnormality detection model from ultrasound images via deep learning.</p><p><strong>Methods: </strong>Data were collected from three Ethiopian healthcare facilities to increase model generalizability. The recruitment period for this study started on November 9, 2024, and ended on November 30, 2024. Several preprocessing techniques have been performed, such as augmentation, noise reduction, and normalization. SegNet, UNet, FCN, MobileNetV2, and EfficientNet-B0 were applied to segment and measure fetal head structures using ultrasound images. The measurements were classified as microcephaly, macrocephaly, or normal using WHO guidelines for gestational age, and then the model performance was compared with that of existing industry experts. The metrics used for evaluation included accuracy, precision, recall, the F1 score, and the Dice coefficient.</p><p><strong>Results: </strong>This study was able to demonstrate the feasibility of using SegNet for automatic segmentation, measurement of abnormalities of the fetal head, and classification of macrocephaly and microcephaly, with an accuracy of 98% and a Dice coefficient of 0.97. Compared with industry experts, the model achieved accuracies of 92.5% and 91.2% for the BPD and HC measurements, respectively.</p><p><strong>Conclusion: </strong>Deep learning models can enhance prenatal diagnosis workflows, especially in resource-constrained settings. Future work needs to be done on optimizing model performance, trying complex models, and expanding datasets to improve generalizability. If these technologies are adopted, they can be used in prenatal care delivery.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"183"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12105205/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149088","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":"Age-and gender-related variations of liver diffusion metrics apparent diffusion coefficient (ADC) and diffusion derived vessel density (DDVD), and explanations with the known physiological T2 relaxation time variations among different volunteers' groups.","authors":"Ying-Ying Deng, Ming-Hua Sun, Hua Huang, Yì Xiáng J Wáng","doi":"10.1186/s12880-025-01730-0","DOIUrl":"10.1186/s12880-025-01730-0","url":null,"abstract":"<p><strong>Background: </strong>Age-related liver diffusion metrics changes have been described. We aim to further clarify these questions: 1) whether an age-related reduction of liver perfusion can be observed by DDVD (diffusion derived vessel density) in older males; 2) whether there is a male female difference in liver perfusion; 3) whether liver ADC values and spleen ADC values are correlated. It is known that, physiologically, males' liver has a higher iron level (thus a shorter T2) than females' liver; pre-menopausal females have a lower liver iron level (thus a longer T2) than post-menopausal females. The observations of this study will be interpreted with the recently gained knowledge of the T2 contribution to diffusion metrics.</p><p><strong>Methods: </strong>Included in this healthy volunteer's study were 68 males (mean age:50.22 years, range: 25-70 years) and 43 females (mean age 45.56 years, range:20-71 years). DWI images with b-values of 0, 2, 10, 20, 60, and 600 s/mm² were acquired at 1.5T. DDVD were calculated with b = 0, b = 2, b = 10, and b = 20 s/mm² images. ADC were calculated with b = 0, b = 2, b = 60 and b = 600 s/mm² images.</p><p><strong>Results: </strong>There was a statistically significant age-related decline of liver DDVD values for females (p = 0.024). A similar trend was observed for males, though statistical significance was not achieved (p = 0.113). Liver DDVD values were all higher in females than in males (p < 0.001). There was a statistically significant age-related decline of liver ADC values both for males (ADC_(b0b600), p = 0.009) and for females (ADC_(b0b600), p = 0.016). Liver ADC values and spleen ADC values were positively correlated (ADC_(b0b600), r = 0.33 for males and 0.31 for females, p < 0.05). When the spleen ADC was used to normalize the liver ADC, then the age-related trend was largely removed, both for males and for females (p > 0.05).</p><p><strong>Conclusion: </strong>Females have a larger liver perfusion volume than males. There is an age-related decrease of DDVD and ADC, both for males and females. Liver ADC values and spleen ADC values are positively correlated. These gender and age-related changes are unlikely mainly caused by the liver T2 relaxation time variations.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"185"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12105168/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149141","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":"Auto-segmentation of cerebral cavernous malformations using a convolutional neural network.","authors":"Chi-Jen Chou, Huai-Che Yang, Cheng-Chia Lee, Zhi-Huan Jiang, Ching-Jen Chen, Hsiu-Mei Wu, Chun-Fu Lin, I-Chun Lai, Syu-Jyun Peng","doi":"10.1186/s12880-025-01738-6","DOIUrl":"10.1186/s12880-025-01738-6","url":null,"abstract":"<p><strong>Background: </strong>This paper presents a deep learning model for the automated segmentation of cerebral cavernous malformations (CCMs).</p><p><strong>Methods: </strong>The model was trained using treatment planning data from 199 Gamma Knife (GK) exams, comprising 171 cases with a single CCM and 28 cases with multiple CCMs. The training data included initial MRI images with target CCM regions manually annotated by neurosurgeons. For the extraction of data related to the brain parenchyma, we employed a mask region-based convolutional neural network (Mask R-CNN). Subsequently, this data was processed using a 3D convolutional neural network known as DeepMedic.</p><p><strong>Results: </strong>The efficacy of the brain parenchyma extraction model was demonstrated via five-fold cross-validation, resulting in an average Dice similarity coefficient of 0.956 ± 0.002. The segmentation models used for CCMs achieved average Dice similarity coefficients of 0.741 ± 0.028 based solely on T2W images. The Dice similarity coefficients for the segmentation of CCMs types were as follows: Zabramski Classification type I (0.743), type II (0.742), and type III (0.740). We also developed a user-friendly graphical user interface to facilitate the use of these models in clinical analysis.</p><p><strong>Conclusions: </strong>This paper presents a deep learning model for the automated segmentation of CCMs, demonstrating sufficient performance across various Zabramski classifications.</p><p><strong>Trial registration: </strong>not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"190"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12107882/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149145","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":"Clinical, radiological, and radiomics feature-based explainable machine learning models for prediction of neurological deterioration and 90-day outcomes in mild intracerebral hemorrhage.","authors":"Weixiong Zeng, Jiaying Chen, Linling Shen, Genghong Xia, Jiahui Xie, Shuqiong Zheng, Zilong He, Limei Deng, Yaya Guo, Jingjing Yang, Yijun Lv, Genggeng Qin, Weiguo Chen, Jia Yin, Qiheng Wu","doi":"10.1186/s12880-025-01717-x","DOIUrl":"10.1186/s12880-025-01717-x","url":null,"abstract":"<p><strong>Background: </strong>The risks and prognosis of mild intracerebral hemorrhage (ICH) patients were easily overlooked by clinicians. Our goal was to use machine learning (ML) methods to predict mild ICH patients' neurological deterioration (ND) and 90-day prognosis.</p><p><strong>Methods: </strong>This prospective study recruited 257 patients with mild ICH for this study. After exclusions, 148 patients were included in the ND study and 144 patients in the 90-day prognosis study. We trained five ML models using filtered data, including clinical, traditional imaging, and radiomics indicators based on non-contrast computed tomography (NCCT). Additionally, we incorporated the Shapley Additive Explanation (SHAP) method to display key features and visualize the decision-making process of the model for each individual.</p><p><strong>Results: </strong>A total of 21 (14.2%) mild ICH patients developed ND, and 35 (24.3%) mild ICH patients had a 90-day poor prognosis. In the validation set, the support vector machine (SVM) models achieved an AUC of 0.846 (95% confidence intervals (CI), 0.627-1.000) and an F1-score of 0.667 for predicting ND, and an AUC of 0.970 (95% CI, 0.928-1.000), and an F1-score of 0.846 for predicting 90-day prognosis. The SHAP analysis results indicated that several clinical features, the island sign, and the radiomics features of the hematoma were of significant value in predicting ND and 90-day prognosis.</p><p><strong>Conclusion: </strong>The ML models, constructed using clinical, traditional imaging, and radiomics indicators, demonstrated good classification performance in predicting ND and 90-day prognosis in patients with mild ICH, and have the potential to serve as an effective tool in clinical practice.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"184"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12105149/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149084","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":"Ultrasonographic differentiation of medullary thyroid carcinoma from papillary thyroid carcinoma: quantitative comparison of morphologic features and different TIRADS risk categorizations.","authors":"Xiaoyu Li, Jiejie Yao, Weiwei Zhan, Wei Zhou","doi":"10.1186/s12880-025-01679-0","DOIUrl":"10.1186/s12880-025-01679-0","url":null,"abstract":"<p><strong>Introduction: </strong>Comparative Analysis of Ultrasonographic Features and Risk Stratification of Kwak-TIRADS, C-TIRADS, and ACR-TIRADS in Medullary versus Papillary Thyroid Carcinomas.</p><p><strong>Objective: </strong>To compare ultrasonographic characteristics and Risk categorization of medullary thyroid carcinomas (MTCs)and papillary thyroid carcinomas (PTCs) with three Thyroid Imaging Reporting and Data Systems (TIRADS), TIRADS proposed by Kwak (Kwak-TIRADS), the Chinese-TIRADS (C-TIRADS) and the 2017 American College of Radiology management guidelines (ACR-TIRADS).</p><p><strong>Methods: </strong>This retrospective study was approved by the Ruijin hospital institutional review board.118 MTC nodules in 96 patients and 511 PTC nodules in 381 patients were included and that all were surgically and pathologically confirmed. Age, size and multiplicity were analyzed by independent sample t test. Sex and sonographic features, including position, composition, echogenicity, shape, border, margin, microcalcification, vascularization distribution and degree were evaluated byχ²orFisher exact test. Each thyroid nodule was categorized by Kwak-TIRADS, C-TIRADS and ACR-TIRADS, and the diagnostic performances was evaluated by receiver operating characteristic (ROC) curves.</p><p><strong>Results: </strong>MTCs had a large size, and most of them were larger than 1 cm (P = 0.000). Female patients were more common in this study(P = 0.035). There was no statistical difference between MTCs and PTCs in age and multiplicity (P > 0.05). The significant statistical differences appeared in various ultrasound features between PTCs and MTCs (P < 0.05).C-TIRADS had the highest diagnostic efficacy (AUC = 0.721), followed by Kwak-TIRADS (AUC = 0.695) and the lowest ACR-TIRADS (AUC = 0.523) (P < 0.0001). Best cut-off point for Kwak-TIRADS, C-TIRADS and ACR-TIRADS were 4c, 4c and TR5. Among the three types of TIRADS, C-TIRADS had the highest sensitivity (66.73%) and negative predictive value (NPV) (32.00%), while KWAK-TIRADS had the highest specificity (72.03%) and positive predictive value (PPV) (90.52%).</p><p><strong>Conclusion: </strong>MTCs exhibited malignant sonographic features similar to PTCs, but also had their own unique characteristics. C-TIRADS was more suitable for distinguishing MTCs from PTCs than the Kwak-TIRADS and ACR-TIRADS, but their diagnostic performance values were not ideal.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"189"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12107721/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149105","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 radiomics of left atrial appendage features for predicting atrial fibrillation recurrence.","authors":"Yanping Yin, Sixiang Jia, Jing Zheng, Wei Wang, Ziwen Wang, Jiangbo Lin, Wenting Lin, Chao Feng, Shudong Xia, Weili Ge","doi":"10.1186/s12880-025-01740-y","DOIUrl":"10.1186/s12880-025-01740-y","url":null,"abstract":"<p><strong>Background: </strong>Structural remodeling of the left atrial appendage (LAA) is characteristic of atrial fibrillation (AF), and LAA morphology impacts radiofrequency catheter ablation (RFCA) outcomes. In this study, we aimed to develop and validate a predictive model for AF ablation outcomes using LAA morphological features, deep learning (DL) radiomics, and clinical variables.</p><p><strong>Methods: </strong>In this multicenter retrospective study, 480 consecutive patients who underwent RFCA for AF at three tertiary hospitals between January 2016 and December 2022 were analyzed, with follow-up through December 2023. Preprocedural CT angiography (CTA) images and laboratory data were systematically collected. LAA segmentation was performed using an nnUNet-based model, followed by radiomic feature extraction. Cox proportional hazard regression analysis assessed the relationship between AF recurrence and LAA volume. The dataset was randomly split into training (70%) and validation (30%) cohorts using stratified sampling. An AF recurrence prediction model integrating LAA DL radiomics with clinical variables was developed.</p><p><strong>Results: </strong>The cohort had a median follow-up of 22 months (IQR 15-32), with 103 patients (21.5%) experiencing AF recurrence. The nnUNet segmentation model achieved a Dice coefficient of 0.89. Multivariate analysis showed that LAA volume was associated with a 5.8% increase in hazard risk per unit increase (aHR 1.058, 95% CI 1.021-1.095; p = 0.002). The model combining LAA DL radiomics with clinical variables demonstrated an AUC of 0.92 (95% CI 0.87-0.96) in the test set, maintaining robust predictive performance across subgroups.</p><p><strong>Conclusion: </strong>LAA morphology and volume are strongly linked to AF RFCA outcomes. We developed an LAA segmentation network and a predictive model that combines DL radiomics and clinical variables to estimate the probability of AF recurrence.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"186"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12105238/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149086","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":"A novel MRI-based deep learning imaging biomarker for comprehensive assessment of the lenticulostriate artery-neural complex.","authors":"Yan Song, Yunlong Jin, Jianguo Wei, Jiajia Wang, Zhong Zheng, Ying Wang, Ru Zeng, Weiping Lu, Bingcang Huang","doi":"10.1186/s12880-025-01676-3","DOIUrl":"10.1186/s12880-025-01676-3","url":null,"abstract":"<p><strong>Objectives: </strong>To develop a deep learning network for extracting features from the blood-supplying regions of the lenticulostriate artery (LSA) and to establish these features as an imaging biomarker for the comprehensive assessment of the lenticulostriate artery-neural complex (LNC).</p><p><strong>Materials and methods: </strong>Automatic segmentation of brain regions on T1-weighted images was performed, followed by the development of the ResNet18 framework to extract and visualize deep learning features from three regions of interest (ROIs). The root mean squared error (RMSE) was then used to assess the correlation between these features and fractional anisotropy (FA) values from diffusion tensor imaging (DTI) and cerebral blood flow (CBF) values from arterial spin labeling (ASL). The correlation of these features with LSA root numbers and three disease categories was further validated using fine-tuning classification (Task1 and Task2).</p><p><strong>Results: </strong>Seventy-nine patients were enrolled and classified into three groups. No significant differences were found in the number of LSA roots between the right and left hemispheres, nor in the FA and CBF values of the ROIs. The RMSE loss, relative to the mean FA and CBF values across different ROI inputs, ranged from 0.154 to 0.213%. The model's accuracy in Task1 and Task2 fine-tuning classification reached 100%.</p><p><strong>Conclusions: </strong>Deep learning features extracted from the basal ganglia nuclei effectively reflect cerebrovascular and neurological functions and reveal the damage status of the LSA. This approach holds promise as a novel imaging biomarker for the comprehensive assessment of the LNC.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"191"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12107879/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149138","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":"Multiparametric radiomics signature for predicting molecular genotypes in adult-type diffuse gliomas utilizing ¹⁸F-FET PET/MRI.","authors":"Jie Bai, Bixiao Cui, Fengqi Li, Xin Han, Hongwei Yang, Jie Lu","doi":"10.1186/s12880-025-01729-7","DOIUrl":"10.1186/s12880-025-01729-7","url":null,"abstract":"<p><strong>Purpose: </strong>This study aimed to investigate the utility of radiomic features derived from multiparametric O-(2-¹⁸ F-fluoroethyl)-L-tyrosine (¹⁸F-FET) positron emission tomography (PET)/ magnetic resonance imaging (MRI) for the prediction of molecular genotypes in adult-type diffuse gliomas.</p><p><strong>Methods: </strong>This retrospective study analyzed 97 adult-type diffuse glioma patients, divided into 70% training and 30% testing cohorts. Each participant underwent hybrid PET/MRI scans, including FLAIR, 3D T1-CE, apparent diffusion coefficient (ADC), and ¹⁸F-FET PET. After the multimodal images were spatially aligned, tumor segmentation was performed on the ¹⁸F-FET PET and then applied to other MRI sequences. A total of 994 radiomic features were extracted from these specified modalities. The Naive Bayesian algorithm with five-fold validation was trained to develop prediction models for the IDH, TERT, and MGMT genotypes and to calculate the radiomics score (Rad-Score). The predictive performance of these models was evaluated via receiver operating characteristic (ROC) curves and decision curve analysis (DCA).</p><p><strong>Results: </strong>The combined model demonstrated superior performance compared to single-modality and MRI (FLAIR + T1-CE + ADC) models in predicting certain genotype statuses in the testing cohort (IDH AUC = 0.97, MGMT AUC = 0.86, TERT AUC = 0.90). The comparisons of the Rad-Score in multimodal models for identifying IDH, TERT, and MGMT showed significant differences (all P < 0.001). Performance of the radiomics signature surpassed that of clinical and conventional radiological factors. DCA indicated that all multimodal models provided good net clinical benefits.</p><p><strong>Conclusions: </strong>Multiparametric ¹⁸F-FET PET/MRI comprehensively analyzes the structural, proliferative, and metabolic information of adult-type diffuse gliomas, enabling precise preoperative diagnosis of molecular genotypes. This has the potential to aid in the development of personalized clinical treatment plans.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"25 1","pages":"187"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12105308/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149092","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}