European Journal of Radiology Open最新文献

{"title":"The predictive power of baseline metabolic and volumetric [18F]FDG PET parameters with different thresholds for early therapy failure and mortality risk in DLBCL patients undergoing CAR-T-cell therapy","authors":"Emil Novruzov ,&nbsp;Helena A. Peters ,&nbsp;Kai Jannusch ,&nbsp;Guido Kobbe ,&nbsp;Sascha Dietrich ,&nbsp;Johannes C. Fischer ,&nbsp;Jutta Rox ,&nbsp;Gerald Antoch ,&nbsp;Frederik L. Giesel ,&nbsp;Christina Antke ,&nbsp;Ben-Niklas Baermann ,&nbsp;Eduards Mamlins","doi":"10.1016/j.ejro.2024.100619","DOIUrl":"10.1016/j.ejro.2024.100619","url":null,"abstract":"<div><h3>Objective</h3><div>[¹⁸F]FDG imaging is an integral part of patient management in CAR-T-cell therapy for recurrent or therapy-refractory DLBCL. The calculation methods of predictive power of specific imaging parameters still remains elusive. With this retrospective study, we sought to evaluate the predictive power of the baseline metabolic parameters and tumor burden calculated with automated segmentation via different thresholding methods for early therapy failure and mortality risk in DLBCL patients.</div></div><div><h3>Materials and methods</h3><div>Eighteen adult patients were enrolled, who underwent CAR-T-cell therapy accompanied by at least one pretherapeutic and two posttherapeutic [¹⁸F]FDG PET scans within 30 and 90 days between December 2018 and October 2023. We performed single-click automatic segmentation within VOIs in addition to extracting the SUV parameters to calculate the MTVs and TLGs by applying thresholds based on the concepts of a fixed absolute threshold with an SUV_max &gt; 4.0, a relative absolute threshold with an isocontour of &gt; 40 % of the SUV_max, a background threshold involving the addition of the liver SUV value and its 2 SD values, and only the liver SUV value.</div></div><div><h3>Results</h3><div>For early therapy failure, baseline metabolic parameters such as the SUV_max, SUV_peak and SUV_mean tended to have greater predictive power than did the baseline metabolic burden. However, the baseline metabolic burden was superior in the prediction of mortality risk regardless of the thresholding method used.</div></div><div><h3>Conclusion</h3><div>This study revealed that automated delineation methods of metabolic tumor burden using different thresholds do not differ in outcome substantially. Therefore, the current clinical standard with a fixed absolute threshold value of SUV &gt; 4.0 seems to be a feasible option.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100619"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11719856/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142972521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rare pancreatic cystic neoplasms: A pictorial review","authors":"Francescamaria Donati,&nbsp;Rosa Cervelli,&nbsp;Piero Boraschi","doi":"10.1016/j.ejro.2024.100620","DOIUrl":"10.1016/j.ejro.2024.100620","url":null,"abstract":"<div><div>Since rare pancreatic cystic tumors may differ from common pancreatic cystic neoplasms in terms of treatment plan and prognosis, the differential diagnosis of these diseases is clinically relevant. Various imaging tests play an important role in the differential diagnosis of rare cystic pancreatic tumors, but accurately distinguishing these diseases solely on the basis of imaging findings is challenging. The purpose of this pictorial review is to present CT and in particular MR imaging features of rare pancreatic cystic tumors and discuss potential elements for differential diagnosis.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100620"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11730956/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142985029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[18F]FDG-PET/CT in DLBCL-patients treated with CAR-T cell therapy: potential for defining patient prognosis","authors":"Helena A. Peters ,&nbsp;Ben-Niklas Bärmann ,&nbsp;Emil Novruzov ,&nbsp;Daniel Weiss ,&nbsp;Matthias Boschheidgen ,&nbsp;Vivien Lorena Ivan ,&nbsp;Nora Liebers ,&nbsp;Johannes Fischer ,&nbsp;Eduards Mamlins ,&nbsp;Aleksandar Radujkovic ,&nbsp;Guido Kobbe ,&nbsp;Julian Kirchner ,&nbsp;Peter Minko ,&nbsp;Kathrin Nachtkamp ,&nbsp;Paul Jäger ,&nbsp;Christina Antke ,&nbsp;Frederik L. Giesel ,&nbsp;Sascha Dietrich ,&nbsp;Gerald Antoch ,&nbsp;Kai Jannusch","doi":"10.1016/j.ejro.2025.100663","DOIUrl":"10.1016/j.ejro.2025.100663","url":null,"abstract":"<div><h3>Objectives</h3><div>The aim of this study is to evaluate the potential of [¹⁸F]FDG-PET/CT in terms of prognostic value and treatment monitoring in relapsed / refractory diffuse large B-cell lymphoma (DLBCL)-patients treated with chimeric antigen receptor T-cell (CAR-T) therapy.</div></div><div><h3>Material &amp; methods</h3><div>Forty-eight [¹⁸F]FDG-PET/CT scans, acquired at pre-defined time points (t₀ – t₂) of 18 DLBCL-patients (mean age: 60 ± 12 years) treated with CAR-T cell therapy were retrospectively enrolled. Median time of follow-up was ten months (IQR 6–16) following CAR-T cell infusion. SUV_max, sum of the product of diameters (SPD), Deauville score (DS) and Lugano classification (LC) were evaluated. Clinical parameters (age, sex) were obtained. Survival time analyses for progression-free survival (PFS) and overall survival (OS) were calculated, the latter by using the Kaplan-Meier method and Cox regression including a hazard ratio (HR). <em>P</em> values below 0.05 were defined as statistically significant. 95 %-confidence intervals (CI) were calculated.</div></div><div><h3>Results</h3><div>Patients with a SUV_max&gt; 9.0 at t₀ (median as threshold value) had a significantly shorter PFS (<em>p</em> = 0.04) and OS (<em>p</em> &lt; 0.01). According to LC, a progressive disease (PD) at t₁ (<em>p</em> = 0.02) and t₂ (<em>p</em> &lt; 0.01) was correlated with a reduced OS. SUV_max &gt; 9.0 at t₀ (<em>p</em> = 0.03, HR = 7.0, CI: 1.3–40.5) and DS &gt; 3 at t₁ (<em>p</em> = 0.04, HR = 8.2, CI: 1.1–61.3) were associated with an increased risk of a PD.</div></div><div><h3>Conclusion</h3><div>SUV_max of [¹⁸F]FDG-PET/CT seems to be useful as a prognostic marker in DLBCL-patients undergoing CAR-T cell therapy. Furthermore, scores of clinical established Deauville classification and Lugano response criteria acquired at post-CAR-T [¹⁸F]FDG-PET/CT might be an indicator for early therapy failure.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100663"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144220956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bone lesions of the tibia: Multimodal iconographic review and diagnostic algorithms, Part 2: Metaphyseal and epiphyseal lesions","authors":"Vincent Salmon,&nbsp;Pedro Augusto Gondim Teixeira,&nbsp;Alain Blum","doi":"10.1016/j.ejro.2025.100654","DOIUrl":"10.1016/j.ejro.2025.100654","url":null,"abstract":"<div><div>This article focuses on the analysis of bone lesions of the tibia, addressing the main diagnostic challenges and imaging strategies used to characterize them. It examines the different etiologies of tibial lesions, emphasizing the importance of a systematic approach to distinguishing tumoral from non-tumoral lesions, as well as from bone dysplasia. The article underlines the essential role of imaging, particularly radiography, CT, and MRI, in accurate lesion characterization. It also highlights typical clinical and radiological features that help guide diagnosis and management. The main aim is to provide radiologists with clear guidelines for improving the identification of bony lesions of the tibia. Part 2 of this 2-part article proposes some illustrations of metaphyseal and epiphyseal lesions of the tibia.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100654"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of hemodynamic disturbances and impaired ventricular filling in asymptomatic fontan patients: A 4D flow CMR study","authors":"Li-Wei Hu ,&nbsp;Xiaodan Zhao ,&nbsp;Shuang Leng ,&nbsp;RongZhen Ouyang ,&nbsp;Qian Wang ,&nbsp;Ai-Min Sun ,&nbsp;Yi-Man Liu ,&nbsp;Wei Dong ,&nbsp;Liang Zhong ,&nbsp;Yu-Min Zhong","doi":"10.1016/j.ejro.2024.100631","DOIUrl":"10.1016/j.ejro.2024.100631","url":null,"abstract":"<div><h3>Background</h3><div>The Fontan procedure is a surgical intervention designed for patients with single ventricle physiology, wherein the systemic venous return is redirected into the pulmonary circulation, thereby facilitating passive pulmonary blood flow without the assistance of ventricular propulsion. Consequently, long-term follow-up of individuals who have undergone the asymptomatic Fontan procedure is essential.</div></div><div><h3>Objectives</h3><div>The aims of this investigation were to: 1) examine the impact of flow components and kinetic energy (KE) parameters on hemodynamic disturbances in asymptomatic Fontan patients and control group; 2) Assess left ventricular diastolic dysfunction through the analysis of 4D flow parameters across different Fontan sub-groups; 3) Compare intracardiac flow parameters among Fontan sub-groups based on morphological features of the left ventricle (LV) and right ventricle (RV).</div></div><div><h3>Methods</h3><div>Twenty-five Fontan patients (mean age: 10 ± 3 years, male/female: 15/10) and fourteen control subjects (mean age: 10 ± 2 years, male/female: 8/6) were recruited retrospectively for the study. The Fontan patients were further categorized into three groups based on their ventricular function: left ventricular (LV), right ventricular (RV), and biventricular (BiV). Each participant underwent cardiovascular magnetic resonance (CMR) imaging, including cine and 4D flow sequences on a 3.0 T scanner. Ventricular flow components and KE were assessed using 4D flow. The study utilized cine images to analyze cardiac function and inter-ventricular mechanical dyssynchrony. Echocardiography evaluated functional ventricular diastolic dysfunction.</div></div><div><h3>Results</h3><div>Fontan patients had a higher median functional single ventricle (FSV) residual volume compared to controls (28 % vs. 23 %, P = 0.034), with lower median FSV direct flow (32 % vs. 40 %, P = 0.005) and delayed ejection flow (17 % vs. 24 %, P = 0.024). The parameters of FSV normalized to the ventricular end-diastolic volume (KEi_EDV) were found to be significantly lower in Fontan patients (all P &lt; 0.05). In both left ventricle (LV) and biventricular (BiV) Fontan subgroups, direct flow was identified as an independent predictor of LV diastolic dysfunction (AUC=0.76, Sensitivity=86 %, Specificity=70 %). Furthermore, residual volume and E-wave KEi_EDV were observed to be significantly different between LV and right ventricle (RV) Fontan subgroups.</div></div><div><h3>Conclusions</h3><div>The altered flow pattern and reduced kinetic energy observed in Fontan patients may indicate hemodynamic disturbances and compromised ventricular filling. Reduced direct flow is associated with LV diastolic dysfunction in LV and BiV Fontan subgroups. Systemic LV exhibited a more efficient intracardiac flow pattern compare with systemic RV in Fontan patients.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100631"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11762912/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143048211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A systematic review on deep learning-enabled coronary CT angiography for plaque and stenosis quantification and cardiac risk prediction","authors":"Priyal Shrivastava ,&nbsp;Shivali Kashikar ,&nbsp;P.H. Parihar ,&nbsp;Pachyanti Kasat ,&nbsp;Paritosh Bhangale ,&nbsp;Prakher Shrivastava","doi":"10.1016/j.ejro.2025.100652","DOIUrl":"10.1016/j.ejro.2025.100652","url":null,"abstract":"<div><h3>Background</h3><div>Coronary artery disease (CAD) is a major worldwide health concern, contributing significantly to the global burden of cardiovascular diseases (CVDs). According to the 2023 World Health Organization (WHO) report, CVDs account for approximately 17.9 million deaths annually. This emphasizies the need for advanced diagnostic tools such as coronary computed tomography angiography (CCTA). The incorporation of deep learning (DL) technologies could significantly improve CCTA analysis by automating the quantification of plaque and stenosis, thus enhancing the precision of cardiac risk assessments. A recent meta-analysis highlights the evolving role of CCTA in patient management, showing that CCTA-guided diagnosis and management reduced adverse cardiac events and improved event-free survival in patients with stable and acute coronary syndromes.</div></div><div><h3>Methods</h3><div>An extensive literature search was carried out across various electronic databases, such as MEDLINE, Embase, and the Cochrane Library. This search utilized a specific strategy that included both Medical Subject Headings (MeSH) terms and pertinent keywords. The review adhered to PRISMA guidelines and focused on studies published between 2019 and 2024 that employed deep learning (DL) for coronary computed tomography angiography (CCTA) in patients aged 18 years or older. After implementing specific inclusion and exclusion criteria, a total of 10 articles were selected for systematic evaluation regarding quality and bias.</div></div><div><h3>Results</h3><div>This systematic review included a total of 10 studies, demonstrating the high diagnostic performance and predictive capabilities of various deep learning models compared to different imaging modalities. This analysis highlights the effectiveness of these models in enhancing diagnostic accuracy in imaging techniques. Notably, strong correlations were observed between DL-derived measurements and intravascular ultrasound findings, enhancing clinical decision-making and risk stratification for CAD.</div></div><div><h3>Conclusion</h3><div>Deep learning-enabled CCTA represents a promising advancement in the quantification of coronary plaques and stenosis, facilitating improved cardiac risk prediction and enhancing clinical workflow efficiency. Despite variability in study designs and potential biases, the findings support the integration of DL technologies into routine clinical practice for better patient outcomes in CAD management.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100652"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143898529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Variability of the posteromedial meniscocapsular junction of the knee on MRI: Pitfall to imaging diagnosis of ramp lesions","authors":"Jack Porrino ,&nbsp;Eric Marten ,&nbsp;Michael L. Richardson ,&nbsp;Jay Moran ,&nbsp;Colby Shreve ,&nbsp;Hyojeong Lee ,&nbsp;Andrew Haims","doi":"10.1016/j.ejro.2025.100647","DOIUrl":"10.1016/j.ejro.2025.100647","url":null,"abstract":"<div><h3>Objective</h3><div>A ramp lesion describes injury at the junction of the posterior horn medial meniscus and posteromedial joint capsule occurring with anterior cruciate ligament deficiency. We sought to apply the consensus of the literature’s description of a ramp lesion on MRI (fluid signal interposed between the posterior medial meniscus and adjacent capsule) to a general population to determine how often this “abnormality” is present on routine MRI and help clarify its specificity.</div></div><div><h3>Material and methods</h3><div>100 consecutive MRI knee studies were retrospectively reviewed by 2 radiologists and in binary fashion characterized as either having features of a ramp lesion or normal appearance. If a ramp lesion was present, the lesion was subclassified according to the Thanaut et al. classification. Patient age, laterality, sex, clinical indication, and ancillary findings on MRI were recorded.</div></div><div><h3>Results</h3><div>Thirty-five of 100 (35 %) knees had MRI findings suggesting a ramp lesion with 31/35 (88.6 %) most consistent with a Thanaut et al. type 1. Only 7 of the 35 (20 %) with ramp lesion had ACL insufficiency. Age (p = 0.00044), right laterality (p = 0.019), and female sex (p = 0.029) were statistically associated with this lesion. There was no association with clinical history indicating recent trauma (p = 0.2399).</div></div><div><h3>Conclusion</h3><div>The appearance of the meniscocapsular junction of the posterior horn medial meniscus may be more varied than the literature discussing ramp lesions suggests. Most notably, fluid interposed between the posterior horn medial meniscus and adjacent posteromedial capsule is not uncommon in those undergoing knee MRI and appears to be nonspecific.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100647"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bone lesions of the tibia: Multimodal iconographic review and diagnostic algorithms, Part 1: Diagnostic algorithms, dysplasia and diaphyseal lesions","authors":"Vincent Salmon,&nbsp;Pedro Augusto Gondim Teixeira,&nbsp;Alain Blum","doi":"10.1016/j.ejro.2025.100653","DOIUrl":"10.1016/j.ejro.2025.100653","url":null,"abstract":"<div><div>This article focuses on the analysis of bone lesions of the tibia, addressing the main diagnostic challenges and imaging strategies used to characterize them. It examines the different etiologies of tibial lesions, emphasizing the importance of a systematic approach to distinguishing tumoral from non-tumoral lesions, as well as from bone dysplasia. The article underlines the essential role of imaging, particularly radiography, CT, and MRI, in accurate lesion characterization. It also highlights typical clinical and radiological features that help guide diagnosis and management. The main aim is to provide radiologists with clear guidelines for improving the identification of bony lesions of the tibia. Part 1 of this 2-part article proposes simplified diagnostic algorithms and some illustrations of dysplasia and diaphyseal lesions of the tibia.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100653"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"3.0 T multi-parametric MRI combined with clinical features improve malignancy prediction of BI-RADS 4 lesions and preoperative prediction of Nottingham Prognostic Index","authors":"Han Zhou ,&nbsp;Haofan Huang ,&nbsp;Kaibin Huang ,&nbsp;XiaoYan Chen ,&nbsp;Yao Fu ,&nbsp;ZiJie Fu ,&nbsp;Xiaolei Zhang ,&nbsp;Renhua Wu ,&nbsp;Yi Gao ,&nbsp;Yan Lin","doi":"10.1016/j.ejro.2025.100665","DOIUrl":"10.1016/j.ejro.2025.100665","url":null,"abstract":"<div><h3>Purpose</h3><div>To establish an optimal model to improve the malignancy prediction of BI-RADS 4 lesions and the preoperative prediction of tumor prognosis.</div></div><div><h3>Materials and methods</h3><div>Ninety-six patients with 126 histopathology-confirmed breast lesions were included in the study. Conventional imaging features, radiomic features based on 3.0 T multi-parametric MRI and patient`s clinical characteristics were analyzed and selected as model candidate features. The least absolute shrinkage and selection operator (Lasso) and Random Forest (RF) were used to construct the combined model. Receiver operating characteristic (ROC) and Net Reclassification Improvement Index (NRI) were performed to assess the diagnostic efficiency between the model and BI-RADS category. Relative ratio (RR) was calculated to assess the ability of model to predict the invasiveness of breast cancers. Finally, the malignant probability (MP) calculated by the optimal model, MRI-based size and lymph node (LN) stage were used by logistic algorithm to construct a preoperative Nottingham Prognostic Index (NPI) model.</div></div><div><h3>Results</h3><div>The combined model incorporating multi-parametric MRI and clinical characteristics was superior to BI-RADS category in the diagnosis of breast cancer (NRI: 1.71, p &lt; 0.05), and had an accuracy of 94 % to predict the malignancy of BI-RADS 4 lesions<strong>.</strong> In addition, MP calculated by the combined model in association with MRI-based size and LN stage can accurately predict the NPI preoperatively (AUC: 92.1 %).</div></div><div><h3>Conclusions</h3><div>The combined model based on multi-parametric MRI and clinical characteristics improves the malignancy prediction of BI-RADS 4 lesions and the preoperative prediction of NPI, therefore providing comprehensive information on the characteristics and treatment plans for breast cancer.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100665"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144264112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning radiomics analysis for prediction of survival in patients with unresectable gastric cancer receiving immunotherapy","authors":"Miaomiao Gou ,&nbsp;Hongtao Zhang ,&nbsp;Niansong Qian ,&nbsp;Yong Zhang ,&nbsp;Zeyu Sun ,&nbsp;Guang Li ,&nbsp;Zhikuan Wang ,&nbsp;Guanghai Dai","doi":"10.1016/j.ejro.2024.100626","DOIUrl":"10.1016/j.ejro.2024.100626","url":null,"abstract":"<div><h3>Objective</h3><div>Immunotherapy has become an option for the first-line therapy of advanced gastric cancer (GC), with improved survival. Our study aimed to investigate unresectable GC from an imaging perspective combined with clinicopathological variables to identify patients who were most likely to benefit from immunotherapy.</div></div><div><h3>Method</h3><div>Patients with unresectable GC who were consecutively treated with immunotherapy at two different medical centers of Chinese PLA General Hospital were included and divided into the training and validation cohorts, respectively. A deep learning neural network, using a multimodal ensemble approach based on CT imaging data before immunotherapy, was trained in the training cohort to predict survival, and an internal validation cohort was constructed to select the optimal ensemble model. Data from another cohort were used for external validation. The area under the receiver operating characteristic curve was analyzed to evaluate performance in predicting survival. Detailed clinicopathological data and peripheral blood prior to immunotherapy were collected for each patient. Univariate and multivariable logistic regression analysis of imaging models and clinicopathological variables was also applied to identify the independent predictors of survival. A nomogram based on multivariable logistic regression was constructed.</div></div><div><h3>Result</h3><div>A total of 79 GC patients in the training cohort and 97 patients in the external validation cohort were enrolled in this study. A multi-model ensemble approach was applied to train a model to predict the 1-year survival of GC patients. Compared to individual models, the ensemble model showed improvement in performance metrics in both the internal and external validation cohorts. There was a significant difference in overall survival (OS) among patients with different imaging models based on the optimum cutoff score of 0.5 (HR = 0.20, 95 % CI: 0.10–0.37, <em>P</em> &lt; 0.001). Multivariate Cox regression analysis revealed that the imaging models, PD-L1 expression, and lung immune prognostic index were independent prognostic factors for OS. We combined these variables and built a nomogram. The calibration curves showed that the C-index of the nomogram was 0.85 and 0.78 in the training and validation cohorts.</div></div><div><h3>Conclusion</h3><div>The deep learning model in combination with several clinical factors showed predictive value for survival in patients with unresectable GC receiving immunotherapy.</div></div>","PeriodicalId":38076,"journal":{"name":"European Journal of Radiology Open","volume":"14 ","pages":"Article 100626"},"PeriodicalIF":1.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11728962/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142979626","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}