Radiologia Medica最新文献

{"title":"Shapley-based saliency maps improve interpretability of vertebral compression fractures classification: multicenter study.","authors":"Liang Xia, Jun Zhang, Zhipeng Liang, Jun Tang, Jianguo Xia, Yongkang Liu","doi":"10.1007/s11547-025-01968-2","DOIUrl":"10.1007/s11547-025-01968-2","url":null,"abstract":"<p><strong>Purpose: </strong>Evaluate the classification performance and interpretability of the Vision Transformer (ViT) model on acute and chronic vertebral compression fractures using Shapley significance maps.</p><p><strong>Materials and methods: </strong>This retrospective study utilized medical imaging data from December 2018 to December 2023 from three hospitals in China. The study included 942 patients, with imaging data comprising X-rays, CTs, and MRIs. Patients were divided into training, validation, and test sets with a ratio of 7:2:1. The ViT model variant, SimpleViT, was fine-tuned on the training dataset. Statistical analyses were performed using the PixelMedAI platform, focusing on metrics such as ROC curves, sensitivity, specificity, and AUC values, with statistical significance assessed using the DeLong test.</p><p><strong>Results: </strong>A total of 942 patients (mean age 69.17 ± 10.61 years) were included, with 1076 vertebral fractures analyzed (705 acute, 371 chronic). In the test set, the ViT model demonstrated superior performance over the ResNet18 model, with an accuracy of 0.880 and an AUC of 0.901 compared to 0.843 and 0.833, respectively. The use of ViT Shapley saliency maps significantly enhanced diagnostic sensitivity and specificity, reaching 0.883 (95% CI: 0.800, 0.963) and 0.950 (95% CI: 0.891, 1.00), respectively.</p><p><strong>Conclusion: </strong>In vertebral compression fractures classification, Vision Transformer outperformed Convolutional Neural Network, providing more effective Shapley-based saliency maps that were favored by radiologists over GradCAM.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"412-421"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143483905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive CT study to assess local and systemic involvement in patients with infective endocarditis: experience from a multidisciplinary team of a tertiary referral center.","authors":"Anna Palmisano, Elisa Bruno, Davide Vignale, Ludovica Bognoni, Raffaele Ascione, Giacomo Ingallina, Paolo Scarpellini, Marco Ripa, Silvia Carletti, Andrea Bettinelli, Roberto Mapelli, Elena Busnardo, Ursula Pajoro, Benedetto Del Forno, Cinzia Trumello, Elisabetta La Penna, Francesco Maisano, Michele De Bonis, Eustachio Agricola, Antonio Esposito","doi":"10.1007/s11547-025-01960-w","DOIUrl":"10.1007/s11547-025-01960-w","url":null,"abstract":"<p><strong>Purpose: </strong>To evaluate the value of a computed tomography (CT) protocol, including ECG-gated cardiac angiographic and venous phase, in patients with infective endocarditis (IE).</p><p><strong>Material and methods: </strong>From January 2019 to October 2022, consecutive patients with IE submitted to total-body CT, including ECG-gated cardiac acquisition in angiographic and venous phase, were enrolled. Transesophageal echocardiography was performed in all cases. Rate of local complications including vegetation, pseudoaneurysm, abscess, fistula and valve dehiscence was compared in CT and echocardiography. Systemic embolization was identified through CT scans.</p><p><strong>Results: </strong>Seventy-six adults (median age 69 [IQR 55-77] years old; males 54/76, 71%] were enrolled. Most patients underwent surgery (51/76, 67%), and the in-hospital mortality rate was 8% (6/76). CT showed higher detection rate of valve vegetation compared to echocardiography (67/76, 88% vs 58/76, 76%; p = 0.008), including vegetation smaller than 10 mm (24/76, 36% vs 16/76, 28%; p = 0.013) and higher detection rate of pseudoaneurysm and abscess (p = 0.004 and p = 0.009, respectively). Abscess showed higher contrast-to-noise ratio (CNR) in the venous scan compared to angiographic scan (2.75 [IQR 2.27; 5.17] vs 1.97 [IQR 1.21; 3.32], p = 0.039) and higher density of perivalvular and epicardial fat compared to pseudoaneurysm (35 [IQR 31; 52]HU and - 50 [IQR - 62; - 35]HU versus 52 [IQR - 60; - 18]HU; p = 0.001, and - 91 [IQR - 95; - 81]HU; p = 0.007, respectively), for greater inflammation. CT overestimated valve dehiscence when compared to echocardiography and surgery.</p><p><strong>Conclusion: </strong>A comprehensive CT study enhances the diagnostic assessment of patients with IE, not only by detecting distant sites of embolization, but also increasing sensitivity for valve vegetation and local complications.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"302-314"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143399818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Polyethylene Glycol microspheres for uterine artery embolization for the treatment of symptomatic uterine fibroids.","authors":"Floriana Nardelli, Fernanda Ciferri, Pierluigi Muratore, Federica Fumarola, Riccardo Faletti, Maria Antonella Ruffino, Marco Calandri, Valeria Accortanzo, Paolo Cortese, Andrea Discalzi","doi":"10.1007/s11547-025-01955-7","DOIUrl":"10.1007/s11547-025-01955-7","url":null,"abstract":"<p><strong>Background: </strong>Uterine fibroids are prevalent benign pelvic tumors, often causing debilitating symptoms that impair quality of life. Uterine fibroid embolization (UFE) is a consolidated minimally invasive treatment option. The purpose of this study is to report our experience with polyethylene glycol microspheres (HydroPearl) in UFE for symptomatic patients.</p><p><strong>Methods: </strong>This single-center retrospective study evaluated 37 consecutive patients with symptomatic uterine fibroids referred to our institution since November 2016 to February 2020 for UFE with HydroPearl microspheres. All patients included completed a comprehensive pre-UFE clinical assessment and underwent a magnetic resonance imaging (MRI) pre- and post-procedure. Technical success, procedure-related complications, clinical outcomes, follow-up MRI findings, and patient satisfaction were evaluated.</p><p><strong>Results: </strong>A technical success rate of 97% with complete bilateral uterine artery embolization was achieved. Significant improvements were observed in menorrhagia, bulk-type symptoms, abdominal pain, and urinary dysfunctions after UFE. Post-procedural MRI assessments demonstrated reductions in uterine and dominant fibroid volumes, indicating successful devascularization. No immediate procedural complications were reported. Symptoms interfering with everyday activities significantly improved after the procedure. Patient satisfaction was high, with 89% expressing satisfaction with the treatment and 84% indicating purpose to repeat the procedure if necessary.</p><p><strong>Conclusions: </strong>Uterine artery embolization with HydroPearl is a safe and effective treatment option for symptomatic uterine fibroids. A significant improvement in menorrhagia and bulk-type symptoms after the procedure was observed correlated by a reduction in diameters and volumes of both the uterus and the main fibroid in post-procedural MRI.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"315-326"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143010636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Importance of technıcal standardızatıon on ultrasound-based lıver fat quantıfıcatıon.","authors":"Halit Nahit Şendur, Mahi Nur Cerit, Atiye Cenay Karabörk Kılıç, Ethem Turgay Cerit","doi":"10.1007/s11547-025-01948-6","DOIUrl":"10.1007/s11547-025-01948-6","url":null,"abstract":"","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"284-285"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143010633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Superior performance in classification of breast cancer molecular subtype and histological factors by radiomics based on ultrafast MRI over standard MRI: evidence from a prospective study.","authors":"Juhyun Jeong, Sungwon Ham, Bo Kyoung Seo, Jeong Taek Lee, Shuncong Wang, Min Sun Bae, Kyu Ran Cho, Ok Hee Woo, Sung Eun Song, Hangseok Choi","doi":"10.1007/s11547-025-01956-6","DOIUrl":"10.1007/s11547-025-01956-6","url":null,"abstract":"<p><strong>Purpose: </strong>To compare the performance of ultrafast MRI with standard MRI in classifying histological factors and subtypes of invasive breast cancer among radiologists with varying experience.</p><p><strong>Methods: </strong>From October 2021 to November 2022, this prospective study enrolled 225 participants with 233 breast cancers before treatment (NCT06104189 at clinicaltrials.gov). Tumor segmentation on MRI was performed independently by two readers (R1, dedicated breast radiologist; R2, radiology resident). We extracted 1618 radiomic features and four kinetic features from ultrafast and standard images, respectively. Logistic regression algorithms were adopted for prediction modeling, following feature selection by the least absolute shrinkage and selection operator. The performance of predicting histological factors and subtypes was evaluated using the area under the receiver-operating characteristic curve (AUC). Performance differences between MRI methods and radiologists were assessed using the DeLong test.</p><p><strong>Results: </strong>Ultrafast MRI outperformed standard MRI in predicting HER2 status (AUCs [95% CI] of ultrafast MRI vs standard MRI; 0.87 [0.83-0.91] vs 0.77 [0.64-0.90] for R1 and 0.88 [0.83-0.91] vs 0.77 [0.69-0.84] for R2) (all P < 0.05). Both ultrafast MRI and standard MRI showed comparable performance in predicting hormone receptors. Ultrafast MRI exhibited superior performance to standard MRI in classifying subtypes. The classification of the luminal subtype for both readers, the HER2-overexpressed subtype for R2, and the triple-negative subtype for R1 was significantly better with ultrafast MRI (P < 0.05).</p><p><strong>Conclusion: </strong>Ultrafast MRI-based radiomics holds promise as a noninvasive imaging biomarker for classifying hormone receptors, HER2 status, and molecular subtypes compared to standard MRI, regardless of radiologist experience.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"368-380"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11903601/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143041458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Complexity of intrinsic brain activity in relapsing-remitting multiple sclerosis patients: patterns, association with structural damage, and clinical disability.","authors":"Fuqing Zhou, Zhizheng Zhuo, Lin Wu, Yuxin Li, Ningnannan Zhang, Xuemei Han, Chun Zeng, Lei Wang, Xiaoya Chen, Muhua Huang, Yanyan Zhu, Haiqing Li, Guanmei Cao, Jie Sun, Yongmei Li, Yunyun Duan","doi":"10.1007/s11547-024-01925-5","DOIUrl":"10.1007/s11547-024-01925-5","url":null,"abstract":"<p><p>Functional plasticity has been demonstrated in multiple sclerosis (MS) studies. However, the intrinsic brain activity complexity alterations remain unclear. Here, using a coarse-graining time-series procedure algorithm, we obtained multiscale entropy (MSE) from a retrospective multi-centre dataset (208 relapsing-remitting MS patients and 228 healthy controls). By linear mixed model analysis, we demonstrated (1) increased entropy at scale 1 and decreased entropy at scale 6, indicating that regional brain activity shifted towards randomness in the stable MS subgroups (n = 159), and (2) decreased entropy across scales 1-6, trending towards regularity in the acute MS subgroups (n = 49). The main results of the correlation analysis included the following: (1) Decreased entropy was associated with lesion volume and brain volume specifically on longer time scales (scale 3-5), and (2) increased entropy of scale 3 was associated with clinical disability scores. These findings reflect the critical role of structural disruption in the brain activity complexity of BOLD signals in MS patients.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"286-295"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142953996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated spinopelvic measurements on radiographs with artificial intelligence: a multi-reader study.","authors":"Boj Friedrich Hoppe, Johannes Rueckel, Jan Rudolph, Nicola Fink, Simon Weidert, Wolf Hohlbein, Adrian Cavalcanti-Kußmaul, Lena Trappmann, Basel Munawwar, Jens Ricke, Bastian Oliver Sabel","doi":"10.1007/s11547-025-01957-5","DOIUrl":"10.1007/s11547-025-01957-5","url":null,"abstract":"<p><strong>Purpose: </strong>To develop an artificial intelligence (AI) algorithm for automated measurements of spinopelvic parameters on lateral radiographs and compare its performance to multiple experienced radiologists and surgeons.</p><p><strong>Methods: </strong>On lateral full-spine radiographs of 295 consecutive patients, a two-staged region-based convolutional neural network (R-CNN) was trained to detect anatomical landmarks and calculate thoracic kyphosis (TK), lumbar lordosis (LL), sacral slope (SS), and sagittal vertical axis (SVA). Performance was evaluated on 65 radiographs not used for training, which were measured independently by 6 readers (3 radiologists, 3 surgeons), and the median per measurement was set as the reference standard. Intraclass correlation coefficient (ICC), mean absolute error (MAE), and standard deviation (SD) were used for statistical analysis; while, ANOVA was used to search for significant differences between the AI and human readers.</p><p><strong>Results: </strong>Automatic measurements (AI) showed excellent correlation with the reference standard, with all ICCs within the range of the readers (TK: 0.92 [AI] vs. 0.85-0.96 [readers]; LL: 0.95 vs. 0.87-0.98; SS: 0.93 vs. 0.89-0.98; SVA: 1.00 vs. 0.99-1.00; all p < 0.001). Analysis of the MAE (± SD) revealed comparable results to the six readers (TK: 3.71° (± 4.24) [AI] v.s 1.86-5.88° (± 3.48-6.17) [readers]; LL: 4.53° ± 4.68 vs. 2.21-5.34° (± 2.60-7.38); SS: 4.56° (± 6.10) vs. 2.20-4.76° (± 3.15-7.37); SVA: 2.44 mm (± 3.93) vs. 1.22-2.79 mm (± 2.42-7.11)); while, ANOVA confirmed no significant difference between the errors of the AI and any human reader (all p > 0.05). Human reading time was on average 139 s per case (range: 86-231 s).</p><p><strong>Conclusion: </strong>Our AI algorithm provides spinopelvic measurements accurate within the variability of experienced readers, but with the potential to save time and increase reproducibility.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"359-367"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11903605/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143041456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interobserver and intraobserver variability among different vendors for mitral valve assessment: implications for transcatheter mitral valve repair.","authors":"Apurva B Challa, Monika Radike, Asim Rizvi, Nikkole M Weber, Malgorzata Wamil, Shivaram Poigai Arunachalam, Emily Sheedy, Shuai Leng, Eric E Williamson","doi":"10.1007/s11547-025-01950-y","DOIUrl":"10.1007/s11547-025-01950-y","url":null,"abstract":"<p><strong>Purpose: </strong>Pre-procedural imaging is critical for transcatheter mitral valve repair planning in patients with mitral valve disease. As differences among various measurement techniques for valve evaluation are still poorly understood, we sought to assess the intra- and interobserver agreement of complex measurements derived from a prototype mitral evaluation tool (Siemens) and a commercially available tool (CVI42) using both saddle- and D-shaped mitral annulus techniques.</p><p><strong>Materials and methods: </strong>Multiphasic cardiac computed tomography angiography data were loaded into each software. Three expert readers independently measured the annuli on systolic- and diastolic-phase images using both tools. Measurement agreement between the tools was assessed with t tests, with p ≤ 0.05 considered statistically significant. Intraclass correlation coefficient (ICC) was used for interobserver agreement. Bland-Altman plots were used to assess for systematic differences.</p><p><strong>Results: </strong>Ten patients (mean age: 61.9 ± 9.9 years, 70%males) were included, with either normal mitral valve (n = 5) or with primary mitral regurgitation (n = 5). The intraobserver comparison showed statistically significantly different circumference and planar surface areas only for one reader in each group. However, there was significant difference (p ≤ 0.05) between measurements for intercommissural distance, septolateral distance, and intertrigone distance. The interobserver agreement was good (ICC, 0.60-0.74) to excellent (ICC, 0.75-1.00). There were systematic differences in mitral annulus area parameters for both saddle- and D-shaped annulus.</p><p><strong>Conclusion: </strong>Both tools effectively assess mitral annulus with good-to-excellent interobserver agreement. As there were systematic differences in the annular area and distance, use of the same software should be considered in clinical and research workflows.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"296-301"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143010634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatial-demographic analysis model for brain metastases distribution.","authors":"Lin Zhang, Tongtong Che, Bowen Xin, Shuyu Li, Guanzhong Gong, Xiuying Wang","doi":"10.1007/s11547-025-01965-5","DOIUrl":"10.1007/s11547-025-01965-5","url":null,"abstract":"<p><strong>Purpose: </strong>The distribution analysis of the morphologic characteristics and spatial relations among brain metastases (BMs) to guide screening and early diagnosis.</p><p><strong>Material and methods: </strong>This retrospective study analysed 4314 BMs across 30 brain regions from MRIs of 304 patients. This paper proposed a unified analysis model based on persistent homology (PH) and graph modelling to provide a comprehensive portrait of BMs distribution. Spatial relationships are quantified through dynamic multiple-scale graphs constructed with Rips filtration. The multi-scale centrality importance and clustering coefficients are extracted to decode BMs spatial relations. Morphologic BMs characteristics are further analysed by varying radius and volume values that are considered as clinically influential factors. Finally, two-tailed proportional hypothesis testing is used for BM statistical distribution analysis.</p><p><strong>Results: </strong>For spatial analysis, results have shown a statistical increase in the proportions of high-level centrality BMs at the left cerebellum (p<0.01). BMs rapidly form graphs with high clustering rather than those with high centrality. For demographic analysis, the cerebellum and frontal are the top high-frequency areas of BMs with 0-4 and 5-10 radii. Statistical increases in the proportions of BMs at cerebellum (p<0.01).</p><p><strong>Conclusion: </strong>Results indicate that distributions of both BMs spatial relations and demographics are statistically non-random. This research offers novel insights into the BMs distribution analysis, providing physicians with the BMs demographic to guide screening and early diagnosis.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"397-411"},"PeriodicalIF":9.7,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11903647/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143524277","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of the stability of intracranial aneurysms using a deep learning model based on computed tomography angiography.","authors":"Lu Zeng, Li Wen, Yang Jing, Jing-Xu Xu, Chen-Cui Huang, Dong Zhang, Guang-Xian Wang","doi":"10.1007/s11547-024-01939-z","DOIUrl":"10.1007/s11547-024-01939-z","url":null,"abstract":"<p><strong>Purpose: </strong>Assessment of the stability of intracranial aneurysms is important in the clinic but remains challenging. The aim of this study was to construct a deep learning model (DLM) to identify unstable aneurysms on computed tomography angiography (CTA) images.</p><p><strong>Methods: </strong>The clinical data of 1041 patients with 1227 aneurysms were retrospectively analyzed from August 2011 to May 2021. Patients with aneurysms were divided into unstable (ruptured, evolving and symptomatic aneurysms) and stable (fortuitous, nonevolving and asymptomatic aneurysms) groups and randomly divided into training (833 patients with 991 aneurysms) and internal validation (208 patients with 236 aneurysms) sets. One hundred and ninety-seven patients with 229 aneurysms from another hospital were included in the external validation set. Six models based on a convolutional neural network (CNN) or logistic regression were constructed on the basis of clinical, morphological and deep learning (DL) features. The area under the curve (AUC), accuracy, sensitivity and specificity were calculated to evaluate the discriminating ability of the models.</p><p><strong>Results: </strong>The AUCs of Models A (clinical), B (morphological) and C (DL features from the CTA image) in the external validation set were 0.5706, 0.9665 and 0.8453, respectively. The AUCs of Model D (clinical and DL features), Model E (clinical and morphological features) and Model F (clinical, morphological and DL features) in the external validation set were 0.8395, 0.9597 and 0.9696, respectively.</p><p><strong>Conclusions: </strong>The CNN-based DLM, which integrates clinical, morphological and DL features, outperforms other models in predicting IA stability. The DLM has the potential to assess IA stability and support clinical decision-making.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":"248-257"},"PeriodicalIF":9.7,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11870988/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142814044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}