Neuroradiology最新文献

{"title":"Machine learning-based assessment of morphometric abnormalities distinguishes bipolar disorder and major depressive disorder.","authors":"Kewei He, Jingbo Zhang, Yang Huang, Xue Mo, Renqiang Yu, Jing Min, Tong Zhu, Yunfeng Ma, Xiangqian He, Fajin Lv, Jianguang Zeng, Chao Li, Robert K McNamara, Du Lei, Mengqi Liu","doi":"10.1007/s00234-025-03544-x","DOIUrl":"10.1007/s00234-025-03544-x","url":null,"abstract":"<p><strong>Introduction: </strong>Bipolar disorder (BD) and major depressive disorder (MDD) have overlapping clinical presentations which may make it difficult for clinicians to distinguish them potentially resulting in misdiagnosis. This study combined structural MRI and machine learning techniques to determine whether regional morphological differences could distinguish patients with BD and MDD.</p><p><strong>Methods: </strong>A total of 123 participants, including BD (n = 31), MDD (n = 48), and healthy controls (HC, n = 44), underwent high-resolution 3D T1-weighted imaging. Cortical thickness, surface area, and subcortical volumes were measured using FreeSurfer software. Common and classic machine learning models were utilized to identify distinct morphometric alterations between BD and MDD.</p><p><strong>Results: </strong>Significant morphological differences were observed in both common and distinct brain regions between BD, MDD, and HC. Specifically, abnormalities in the amygdala, thalamus, medial orbitofrontal cortex and fusiform were observed in both BD and MDD compared with HC. Relative to HC, unique differences in BD were identified in the lateral occipital and inferior/middle temporal regions, whereas MDD exhibited differences in nucleus accumbens and middle temporal regions. BD exhibited larger surface area in right middle temporal gyrus and greater right nucleus accumbens volume compared to MDD. The integration of two-stage models, including deep neural network (DNN) and support vector machine (SVM), achieved an accuracy rate of 91.2% in discriminating individuals with BD from MDD.</p><p><strong>Conclusion: </strong>These findings demonstrate that structural MRI combined with machine learning techniques can accurately discriminate individuals with BD from MDD, and provide a foundation supporting the potential of this approach to improve diagnostic accuracy.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":"921-930"},"PeriodicalIF":2.4,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143008863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spinal cord cross sign: a potential marker for hereditary spastic paraplegia type 5.","authors":"Fan Zhang, Jianping Hu, Zebin Xiao, Chenlin Lin, Zhuoting Huang, Ning Wang, Ying Liu","doi":"10.1007/s00234-025-03543-y","DOIUrl":"10.1007/s00234-025-03543-y","url":null,"abstract":"<p><strong>Purpose: </strong>Spastic paraplegia type 5 (SPG5) is a rare neurodegenerative disease diagnosed primarily through genetic testing.We identified a specific spinal cord sign on conventional MR imaging to help narrow the scope of genetic screening.</p><p><strong>Methods: </strong>In 25 patients with SPG5 and 21 healthy controls (HCs), the spinal cord cross sign was evaluated on T2*-weighted imaging. The morphological and signal characteristics of the dorsal column (DC), ventral funiculi (VF), dorsal horn (DH), ventral horn (VH), and intermediate zone (IMZ) were assessed. Differences in fractional anisotropy (FA) values within specific regions between HC and SPG5 were tested using Student's t-test. Spearman correlation was used to evaluate associations between cross-sign scores, FA values, and clinical indicators.</p><p><strong>Results: </strong>The cross sign was detected in the cervical spinal cord of all SPG5 patients. The occurrence of T2 hyperintensity in the DC, VF and IMZ was 100%,100% and 88%,respectively. Bilateral VH morphology was normal in 14.4% of cases, blurred in 49.6%, and absent in 36%.Bilateral DH morphology was normal in 13.6%, blurred in 56%, and absent in 30.4%. FA values were reduced in these spinal cord regions. Cross-sign scores were negatively correlated with FA values in both grey (r = -0.70~-0.37) and white matter (r = -0.78~-0.70). Cross-sign scores were positively correlated with Spastic Paraplegia Rating Scale (r = 0.57) and disease duration (r = 0.42).</p><p><strong>Conclusion: </strong>The spinal cord cross sign was a potential imaging marker for SPG5. Cross-sign scores were associated with disease duration and severity in SPG5 patients.</p><p><strong>Trial registration: </strong>A Registered Cohort Study on Spastic Paraplegia,NCT04006418 Registered 1 July 2019, https://clinicaltrials.gov/study/NCT04006418 .</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":"1081-1090"},"PeriodicalIF":2.4,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143033753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine learning radiomics for H3K27M mutation prediction in gliomas: A systematic review and meta-analysis.","authors":"Bardia Hajikarimloo, Salem M Tos, Alireza Kooshki, Mohammadamin Sabbagh Alvani, Mohammad Shahir Eftekhar, Arman Hasanzade, Roozbeh Tavanaei, Mohammadhosein Akhlaghpasand, Rana Hashemi, Mohammadreza Ghaffarzadeh-Esfahani, Ibrahim Mohammadzadeh, Mohammad Amin Habibi","doi":"10.1007/s00234-025-03597-y","DOIUrl":"https://doi.org/10.1007/s00234-025-03597-y","url":null,"abstract":"<p><strong>Purpose: </strong>Noninvasive prediction and identification of the H3K27M mutation play an important role in optimizing therapeutic strategies and improving outcomes in gliomas. In this systematic review and meta-analysis, we aimed to evaluate the performance of machine learning (ML)-based models in predicting H3K27M mutation in gliomas.</p><p><strong>Methods: </strong>Literature records were retrieved on September 16th, 2024, in PubMed, Embase, Scopus, and Web of Science. Records were screened according to the eligibility criteria, and the data from the included studies were extracted. The meta-analysis, sensitivity analysis, and meta-regression were conducted using R software.</p><p><strong>Results: </strong>A total of 15 studies were included in our study. Our meta-analysis demonstrated a pooled AUC, sensitivity, and specificity of 0.87 (95% CI: 0.77-0.97), 92% (95% CI: 83%-96%), and 89% (95% CI: 86%-91%)), respectively. The subgroup meta-analysis revealed that despite the higher sensitivity of the deep learning (DL) models, the sensitivity is not superior to ML (P = 0.6). In contrast, the ML-based pooled specificity was significantly higher (P < 0.01). The meta-analysis revealed a 78.1 (95% CI: 33.3 - 183.5). The SROC curve indicated an AUC of 0.921, and the estimated sensitivity is 0.898 concurrent with the false positive rate of 0.126, which indicates high sensitivity with a low false positive rate.</p><p><strong>Conclusion: </strong>Our systematic review and meta-analysis demonstrated that ML-based magnetic resonance imaging (MRI) radiomics models are associated with promising diagnostic performance in predicting H3K27M mutation in gliomas.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":""},"PeriodicalIF":2.4,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143753652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cangrelor, a newer P2Y12 inhibitor, in neuro-interventional procedures: a systematic review and updated meta-analysis.","authors":"Ocílio Ribeiro Gonçalves, Ana B Santos, Anthony Hong, Leonardo Januário Campos Cardoso, Márcio Yuri Ferreira, Christian Ken Fukunaga, Victor Arthur Ohannesian, Kelson James Almeida, Filipe Virgilio Ribeiro","doi":"10.1007/s00234-025-03573-6","DOIUrl":"https://doi.org/10.1007/s00234-025-03573-6","url":null,"abstract":"<p><strong>Background: </strong>Cangrelor, a reversible P2Y12 receptor inhibitor commonly utilized in cardiovascular interventions, is increasingly being investigated for its potential in cerebrovascular applications. This study presents a comprehensive evaluation of its safety and efficacy in neuro-interventional procedures, based on a systematic review and meta-analysis of recent evidence.</p><p><strong>Methods: </strong>We searched PubMed, Embase, Cochrane, Web of Science, and Scopus for studies on endovascular therapy with cangrelor-based intravenous antiplatelet therapy for cerebrovascular pathologies. Endpoints included 90-day functional outcomes, successful reperfusion (mTICI 2b-3), mortality, symptomatic intracranial hemorrhage (sICH), hemorrhagic transformation, gastrointestinal bleeding, intraprocedural complications (in-stent thrombosis, thromboembolic events), and retroperitoneal hematoma. Single-proportion analysis with 95% CIs under a random-effects model was conducted.</p><p><strong>Results: </strong>Seventeen studies with 646 patients were included. Favorable 90-day functional outcomes occurred in 57.06% (44.37-69.76%), and successful recanalization in 98.74% (96.63-100.00%). Rates of hemorrhagic transformation, sICH, gastrointestinal bleeding, and retroperitoneal hematoma were 24.06%, 4.64%, 0.02%, and 0.07%, respectively. Intraprocedural in-stent thrombosis occurred in 0.28%, thromboembolic events in 0.48%, and 90-day all-cause mortality was 7.94%.</p><p><strong>Conclusion: </strong>The findings suggest that intravenous cangrelor, used as antiplatelet therapy following neuro-interventional procedures, is both safe and effective. This is reflected in high rates of favorable clinical outcomes, successful recanalization, and low complication rates.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":""},"PeriodicalIF":2.4,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143743286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accelerated flat panel computed tomography for pre-operative temporal bone imaging: Image quality and dosimetry comparison to conventional high resolution multislice computed tomography.","authors":"Elie Diamandis, Sebastian Johannes Müller, Eya Khadhraoui, Stefan Klebingat, Eric Einspänner, Martin Durisin, Anne Albrecht, Daniel Behme","doi":"10.1007/s00234-025-03592-3","DOIUrl":"https://doi.org/10.1007/s00234-025-03592-3","url":null,"abstract":"<p><strong>Purpose: </strong>High-resolution multislice CT (HR-MSCT) and cone beam CT (CBCT) are commonly used for preoperative temporal bone imaging, with HR-MSCT often preferred due to its shorter scan duration and lower susceptibility to motion artifacts. However, recent advancements in accelerated flat panel CT (Acc-FPCT) available with the latest generation angiography systems have addressed traditional limitations of CBCT by significantly decreasing scan time. This cadaver-based study evaluates the diagnostic performance and radiation dose of Acc-FPCT compared to HR-MSCT in preoperative temporal bone imaging.</p><p><strong>Methods: </strong>Six different Acc-FPCT protocols were acquired on five whole-head cadaveric specimens (ten temporal bones). Three neuroradiologists experienced in temporal bone imaging assessed the image quality of Acc-FPCT protocols in comparison to that of HR-MSCT for the visualization of 31 landmarks of middle and inner ear using a 5-point Likert scale. We also compared radiation dose parameters (CT dose index and dose length product) among the protocols.</p><p><strong>Results: </strong>Two high-Resolution Acc-FPCT protocols were found to be superior to HR-MSCT by all raters (p < 0.001). There were no significant differences between the two HR-FPCT protocols (p = 0.25). The remaining Acc-FPCT protocols were rated significantly inferior to HR-MSCT. The inter-rater reliability was excellent (ICC (2,k) = 0.925; CI [0.92-0.93]). The dose length product was significantly lower in all Acc-FPCT protocols compared to HR-MSCT.</p><p><strong>Conclusion: </strong>The results of our cadaver-based study highlight the utility of certain Acc-FPCT protocols as a viable alternative to HR-MSCT in preoperative temporal bone imaging, improving the visualization of critical anatomical landmarks without increasing radiation exposure.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":""},"PeriodicalIF":2.4,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143710665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving differentiation of hemorrhagic brain metastases from non-neoplastic hematomas using radiomics and clinical feature fusion.","authors":"Linyang Cui, Luyue Yu, Sai Shao, Liping Zuo, Hongjun Hou, Jie Liu, Wenjun Zhang, Ju Liu, Qiang Wu, Dexin Yu","doi":"10.1007/s00234-025-03590-5","DOIUrl":"https://doi.org/10.1007/s00234-025-03590-5","url":null,"abstract":"<p><strong>Objectives: </strong>This study aimed to develop and validate a fusion model combining multi-sequence MRI radiomics and clinico-radiological features to distinguish hemorrhagic brain metastasis covered by hematoma (HBM.cbh) from non-neoplastic intracranial hematomas (nn-ICH).</p><p><strong>Methods: </strong>The data of 146 patients with pathologically or clinically proven HBM.cbh (n = 55) and nn-ICH (n = 91) were collected from two clinical institutions. Radiomics features were extracted from various regions (hemorrhage and/or edema) based on T2-weighted, T1-weighted, fluid-attenuated inversion-recovery, and T1 contrast-enhanced imaging. Synthetic minority over-sampling technique (SMOTE) was performed to balance the minority group (HBM.cbh). Logistic regression (LR) and k-nearest neighbors (KNN) were utilized to construct the models based on clinico-radiological factors (clinical model), radiomic features from various modalities of MRI (radiomics model), and their combination (fusion model). The area under the curve (AUC) values of different models on the external dataset were compared using DeLong's test.</p><p><strong>Results: </strong>The 4-sequence radiomics model based on the entire region performed the best in all radiomics models, with or without SMOTE, where the AUCs were 0.83 and 0.84, respectively. The AUC of clinical mode was 0.71 with SMOTE, and 0.62 without SMOTE. The fusion model demonstrated excellent predictive value with or without SMOTE (AUC: 0.93 and 0.90, respectively), outperforming both the radiomics and clinical model (0.93 vs. 0.83, 0.71, p < 0.05 and 0.90 vs. 0.84, 0.62, p < 0.05, respectively).</p><p><strong>Conclusions: </strong>The multi-sequence radiomics model is an effective method for differentiating HBM.cbh from nn-ICH. It can yield the best diagnostic performance prediction model when combined with clinico-radiological features.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":""},"PeriodicalIF":2.4,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143710680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Neurodegeneration correlates of iron-related lesions and leptomeningeal inflammation in multiple sclerosis clinical subtypes.","authors":"Aigli G Vakrakou, Ioannis Papadopoulos, Maria-Evgenia Brinia, Dimitrios Karathanasis, Dimitrios Panaretos, Panos Stathopoulos, Anastasia Alexaki, Varvara Pantoleon, Efstratios Karavasilis, Georgios Velonakis, Leonidas Stefanis, Maria-Eleftheria Evangelopoulos, Constantinos Kilidireas","doi":"10.1007/s00234-025-03595-0","DOIUrl":"https://doi.org/10.1007/s00234-025-03595-0","url":null,"abstract":"<p><strong>Purpose: </strong>The aim of this study was to investigate the significant implications of different types of lesions as assessed by QSM (quantitative-susceptibility-mapping) as well as leptomeningeal contrast-enhancement in a cohort of Relapsing-Remitting (RR) and Primary Progressive (PP) MS patients and to assess their association with clinical disability and MRI-measures of brain structural damage.</p><p><strong>Methods: </strong>Different types of white-matter lesions were identified and quantified using QSM in 24 RRMS and 15 PPMS (11 patients with follow-up MRI). Leptomeningeal contrast-enhancement (LMCE; foci) was assessed on 3D-FLAIR post-gadolinium.</p><p><strong>Results: </strong>Both RRMS and PPMS presented PRL (paramagnetic-rim lesions) and LMCE, with PPMS showing a trend towards more LMCE (RRMS 37%, PPMS 53%). In QSM RRMS patients showed more hyperintense white-matter lesions with greater lesion volume. In RRMS PRL correlated with disease duration and lesion burden especially the volume of juxtacortical Flair-hyperintense lesions. Besides, the presence of PRL lesions in PPMS was associated with subcortical atrophy mainly thalamus and pallidum volumetry. In all MS-cohort, patients with more than 3-PRLs exhibited reduced regional cortical thickness in specific temporal areas and post/para central gyrus. Forest-analysis selected age, increased NAWM (normal appearing white-matter) QSM intensity, total lesion volume and the presence of LMCE as informative predictors of cortical thickness. After anti-CD20 treatment, no significant change was observed regarding the number of PRL and LMCE, but the percentage of PRL lesions over the total lesion types and the QSM rim intensity increased.</p><p><strong>Conclusion: </strong>Our findings suggest that QSM-lesion types and leptomeningeal inflammation capture different aspects of progressive disease biology in both RRMS and PPMS.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":""},"PeriodicalIF":2.4,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143710683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiomics based on preoperative magnetic resonance imaging predict the cell lineages of nonfunctioning pituitary neuroendocrine tumors.","authors":"Xuening Zhao, Xu Fu, Xiaochen Wang, Sihui Wang, Lingxu Chen, Mengyuan Yuan, Jiangang Liu, Shengjun Sun","doi":"10.1007/s00234-025-03593-2","DOIUrl":"https://doi.org/10.1007/s00234-025-03593-2","url":null,"abstract":"<p><strong>Objective: </strong>Accurate preoperative predict the cell lineages of non-functioning pituitary neuroendocrine tumors (NFPitNETs) can help neurosurgeons develop treatment strategies. This study aimed to predict the three cell lineages of NFPitNETs using radiomics based on MRI.</p><p><strong>Methods: </strong>NFPitNETs patients from January 2019 and January 2023 were retrospectively enrolled, with adenoma lineages including SF-1 (n = 239), TPIT (n = 204), and PIT-1 (n = 100). Sagittal T1-weighted images (T1WI), contrast-enhanced (CE) sagittal T1WI, CE-coronal T1WI, and axial T2WI were obtained for tumor segmentation on ITK-SNAP. Pyradiomics was used for features extracted. Variance threshold method, t-test, and LASSO were used for feature selection. Support vector machine (SVM) and random forest (RF) were used to predict the three-lineages adenomas based on their radiomics and semantic features. Receiver operating characteristic curve-area under the curve (ROC-AUC) analysis was used to assess the model's performance.</p><p><strong>Results: </strong>A total of 543 patients with NFPitNETs (mean age, 49.46 ± 12.39) were included. Patients with SF-1 adenomas had a higher mean age than those with TPIT and PIT-1 adenomas (52.84 ± 11.56 vs 49.94 ± 10.54 vs 40.42 ± 13.41, p < 0.001). Female patients are more common in TPIT and PIT-1 adenomas than SF-1 ones (96.57% vs 69% vs 41%, p < 0.001). The SVM model incorporating semantic and radiomics features based on CE-coronal T1WI performed the best, with a macro-average AUC of 0.899. CE-coronal T1WI were the best among all the MR sequences for predicting the cell lineages of NFPitNETs.</p><p><strong>Conclusion: </strong>Radiomics based on preoperative MRI can help predict the cell lineages of NFPitNETs, which prove useful to neurosurgeons to develop treatment strategies.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":""},"PeriodicalIF":2.4,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143674328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning-assisted detection of intracranial hemorrhage: validation and impact on reader performance.","authors":"Dong-Wan Kang, Museong Kim, Gi-Hun Park, Yong Soo Kim, Moon-Ku Han, Myungjae Lee, Dongmin Kim, Wi-Sun Ryu, Han-Gil Jeong","doi":"10.1007/s00234-025-03560-x","DOIUrl":"https://doi.org/10.1007/s00234-025-03560-x","url":null,"abstract":"<p><strong>Purpose: </strong>Intracranial hemorrhage (ICH) requires urgent treatment, and accurate and timely diagnosis is essential for improving outcomes. This pivotal clinical trial aimed to validate a deep learning algorithm for ICH detection and assess its clinical utility through a reader performance test.</p><p><strong>Methods: </strong>Retrospective CT scans from patients with and without ICH were collected from a tertiary hospital. Two experts evaluated all scans, with a third expert reviewing disagreements for the final diagnosis. We analyzed the performance of the deep learning algorithm, JLK-ICH, for all cases and ICH subtypes. Additional external validation was performed using a multi-ethnic U.S.</p><p><strong>Dataset: </strong>A reader performance study included six non-expert readers who evaluated 800 CT scans, with and without JLK-ICH assistance, following a washout period. ICH presence and five-point scale confidence level for decisions were rated.</p><p><strong>Results: </strong>A total of 1,370 CT scans were evaluated. The deep learning model showed 98.7% sensitivity (95% confidence interval [CI] 97.8-99.3%), 88.5% specificity (95% CI, 83.6-92.3%), and an area under the receiver operating characteristic curve (AUROC) of 0.936 (95% CI, 0.915-0.957). The model maintained high accuracy across all ICH subtypes, and additional external validation confirmed these results. In the reader performance study, AUROC with JLK-ICH assistance (0.967 [0.953-0.981]) surpassed that without assistance (0.953 [0.938-0.957]; P = 0.009). JLK-ICH particularly improved performance when readers were highly uncertain.</p><p><strong>Conclusion: </strong>The JLK-ICH algorithm demonstrated high accuracy in detecting all ICH subtypes. Non-expert readers significantly improved diagnostic accuracy for brain CT scans with deep learning assistance.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":""},"PeriodicalIF":2.4,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143674299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Use of amide proton transfer (APT) imaging in the differentiation of pediatric low-grade brain tumors from tumor-like brain lesions.","authors":"Antonia Ramaglia, Costanza Parodi, Claudia Milanaccio, Antonio Verrico, Marta Molteni, Maria Luisa Garrè, Mattia Pacetti, Stephanie Vella, Martina Resaz, Domenico Tortora, Mariasavina Severino, Andrea Rossi","doi":"10.1007/s00234-025-03582-5","DOIUrl":"https://doi.org/10.1007/s00234-025-03582-5","url":null,"abstract":"<p><strong>Background and purpose: </strong>Low grade tumors (LGT) are the most frequent central nervous system lesions observed in children. Despite the high-throughput research, differentiating LGT from tumor- like lesions (TLL) and providing an accurate differential diagnosis based on conventional MRI remains a challenge. For this reason, advanced MR sequences are routinely investigated and applied in clinical practice. The aim of this study is to explore the potential of the amide proton transfer (APTw) sequence as a tool for discriminating LGT from TLL.</p><p><strong>Materials and methods: </strong>In this single-center retrospective study, we recruited 35 patients (20 with a histologically confirmed LGT, and 15 with a TLL) with both conventional and APT MRI images obtained on a 3T clinical scanner at onset or prior to treatment/surgery. Two volumes of interest (VOI), namely the whole lesion and the normal appearing white matter (NAWM), were defined using the semi-automatic segmentation tool from Philips Intellispace portal for Windows (v. 8). The mean APTw (mAPTw) and difference between the mAPTw lesion and the NAWM (dAPTw) were measured and compared between the two groups.</p><p><strong>Results: </strong>Lower values were found in the TLL group compared to the LGT group for both the mAPTw (1.51 ± 0.64% vs. 2.87 ± 0.96%) and dAPTw (0.24 ± 0.72% vs. 1.53 ± 1.08%) (p-value < 0.001). Based on ROC curve analysis, optimal cut-offs value for mAPTw and dATPw were 1.79 and 0.53, respectively.</p><p><strong>Conclusion: </strong>APT imaging may prove useful to discriminate between LGT and TLL in pediatric patients.</p>","PeriodicalId":19422,"journal":{"name":"Neuroradiology","volume":" ","pages":""},"PeriodicalIF":2.4,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143674257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}