AJNR. American journal of neuroradiology最新文献

{"title":"Endovascular Treatment for Tentorial Dural Arteriovenous Fistulas: A Retrospective Single-Center Study.","authors":"Guanghao Zhang, Miao Pang, Zhe Li, Chenghao Shang, Yuhang Zhang, Qi Zhang, Qinghai Huang, Yi Xu, Guoli Duan, Qiang Li, Jianmin Liu","doi":"10.3174/ajnr.A8676","DOIUrl":"10.3174/ajnr.A8676","url":null,"abstract":"<p><strong>Background and purpose: </strong>Current evidence suggests that tentorial dural arteriovenous fistulas with endovascular treatment offer a high rate of occlusion and reduced procedural risks. We report the clinical and angiographic outcomes in patients with tentorial dural arteriovenous fistulas who underwent endovascular treatment as first-line treatment.</p><p><strong>Materials and methods: </strong>A retrospective analysis was conducted on 83 patients with tentorial dural arteriovenous fistulas treated at our center from April 2009 to November 2023 by using endovascular treatment. Patient demographics, clinical presentations, angiographics, treatment results, and follow-up outcomes were registered. Univariable and multivariable logistic regression were performed to identify the onset of intracranial hemorrhage predictors, procedure-related complications predictors, and predictors of poor functional outcome.</p><p><strong>Results: </strong>Eighty-three patients underwent endovascular treatment as first-line treatment. Most patients (80.7%) were symptomatic, and a total of 25 patients (30.1%) presented with intracranial hemorrhage. The presence of deep venous drainage and midline location were associated with a lower incidence of intracranial hemorrhage. Middle meningeal artery (57.8%, 48/83) and pial artery (15.7%, 13/83) were the most frequently used access routes for embolization. A total of 74 (89.1%) tentorial dural arteriovenous fistulas were completely occluded immediately after the last endovascular treatment session. Complications occurred in 11 patients (13.3%). Six-month angiographic follow-up was performed in 68 patients (81.8%), and 91.2% (62/68) of tentorial dural arteriovenous fistulas were occluded. At clinical follow-up (100%, 83/83), good functional outcome (mRS 0-2) was documented in 74 (89.1%) patients. Logistic regression analysis identified baseline mRS 3-5 as the most significant independent predictor of poor functional outcome. Subgroup analysis showed no statistically significant differences in baseline characteristics, angiography, and clinical results between patients treated with targeted pial artery embolization and nontargeted pial artery embolization.</p><p><strong>Conclusions: </strong>Endovascular treatment is a safe and effective primary technique for managing tentorial dural arteriovenous fistulas, achieving high rates of complete angiographic occlusion and favorable functional outcomes. Transarterial embolization, predominantly via the middle meningeal artery, was the mainstay of treatment. In patients with pial arterial feeders, omitting aggressive embolization did not compromise efficacy or increase complications.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143069613","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":"Outcomes of CT-Guided Targeted Epidural Patching For Lateral Dural Tears In Spontaneous Intracranial Hypotension: A Multicenter Retrospective Cohort Study.","authors":"Andrew L Callen, Daniel Montes, Debayan Bhaumik, Peter Lennarson, Mark D Mamlouk, Niklas Lützen, Jürgen Beck, Horst Urbach, Daniel Scoffings, David Butteriss, Lalani Carlton Jones","doi":"10.3174/ajnr.A8886","DOIUrl":"10.3174/ajnr.A8886","url":null,"abstract":"<p><strong>Background and purpose: </strong>Spontaneous intracranial hypotension (SIH) due to lateral dural tears with spinal longitudinal extradural fluid collections (SLECs) can cause disabling orthostatic headaches. While epidural patching is commonly used as first-line treatment, outcomes specific to lateral dural tears have not been well characterized. We aimed to evaluate clinical and radiologic outcomes following CT-guided patching for lateral dural tears and assess whether anatomic or procedural factors- including presence of a herniated arachnoid pouch, patch volume, material, or approach-influence treatment success.</p><p><strong>Materials and methods: </strong>This was a retrospective multicenter cohort study of patients with lateral dural tears treated with CT-guided epidural patching between December 2013 and March 2025. Demographics, leak characteristics, patching details, and clinical and imaging outcomes were collected. The presence of herniated arachnoid pouches on spine MRI and pre/post-treatment Bern scores were recorded. Associations between clinical resolution, SLEC resolution, and procedural variables were analyzed using univariate methods.</p><p><strong>Results: </strong>Fifty-six patients (mean age 38.7 ± 11.7 years; 80% female) were included. Mean pretreatment Bern score was 6.6 ± 2.3; no patients had superficial siderosis. Clinical resolution occurred in 20/56 (35.7%), and SLEC resolution in 10/40 (25%) on post-patch spine MRI. A herniated arachnoid pouch was present in 69.7% and associated with lower SLEC resolution (30% vs. 80%, p=0.003). Patch type, volume, transforaminal approach, and needle placement into the herniated pouch were not associated with outcomes. Post-patch Bern scores were lower among those with clinical resolution (0.9 ± 1.1 vs. 2.6 ± 2.5, p=0.046). Of 11 patients with complete clinical improvement who had post-patch spine MRI, 5 (45.5%) had persistent SLECs.</p><p><strong>Conclusions: </strong>CT-guided patching led to complete symptom resolution in roughly one-third of patients with lateral dural tears. Herniated arachnoid predicted lower SLEC resolution, while procedural variables were not predictive of outcome. A subset of patients improved clinically despite persistent extradural fluid, emphasizing the need for long-term monitoring.<b>ABBREVIATIONS:</b> SIH＝ spontaneous intracranial hypotension; SLEC = spinal longitudinal extradural fluid collection.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144337254","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":"7 Tesla Time-of-Flight MRA in Adolescents.","authors":"Sergio Valencia, Fedel Machado-Rivas, Maria Camila Cortes-Albornoz, Samuel Cd Cartmell, Harry Griffin, Darren B Orbach, Vanessa Rameh, Alfred Pokmeng See, Simon K Warfield, Onur Afacan, Camilo Jaimes","doi":"10.3174/ajnr.A8908","DOIUrl":"https://doi.org/10.3174/ajnr.A8908","url":null,"abstract":"<p><strong>Background and purpose: </strong>The increased signal-to-noise ratio (SNR) at 7 Tesla (7T) enables higher spatial resolution for neurovascular imaging, yet its application in pediatric magnetic resonance angiography (MRA) remains underexplored. This study systematically evaluates the advantages of 7T time-of-flight (TOF) MRA compared to 3T in pediatric patients, hypothesizing that 7T would provide superior vessel contrast and increased vascular volume, given the use of smaller voxels, as well as higher SNR, despite these smaller voxels.</p><p><strong>Materials and methods: </strong>This HIPAA-compliant, IRB-approved retrospective study included pediatric patients (<19 years) who underwent 7T TOF MRA. Controls consisted of either same-subject 3T MRAs within 6 months (when available) or age-and sex-matched 3T MRA subjects. Imaging parameters were optimized for spatial resolution at 7T to achieve 0.3-0.4 mm isotropic voxels. Quantitative analysis included contrast ratio (CR) and SNR measurements for the ICA, M1 to M4 arterial segments, and lenticulostriate perforating arteries. Vascular volume was assessed using 3D segmentation. Semiquantitative vessel conspicuity ratings and motion artifact scoring were performed by blinded neuroradiologists.</p><p><strong>Results: </strong>Fifteen patients (10 with 7T MRA, 5 with matched 3T controls) and 20 MRAs were analyzed. CR was significantly higher at 7T for perforators, M3, and M4 branches (P < 0.05), with the greatest improvement in M4 branches. Vascular volume was 147% greater at 7T (P = 0.018), reflecting improved small vessel depiction and segmentation. Semiquantitative analysis showed significantly better vessel conspicuity at 7T for M4 branches and lenticulostriate perforators (P < 0.01). Motion artifact scores were similar between field strengths (P = 0.118).</p><p><strong>Conclusions: </strong>7T TOF MRA significantly enhances vascular contrast and improves visualization of small arteries compared to 3T, making it a valuable tool for pediatric cerebrovascular imaging.</p><p><strong>Abbreviations: </strong>CR＝ Contrast Ratio; LSA＝ Lenticulostriate Perforating Arteries.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144546526","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":"Fusion of FDG and FMZ PET Reduces False-Positives in Predicting Epileptogenic Zone.","authors":"Bingyang Cai, Shize Jiang, Hui Huang, Jiwei Li, Siyu Yuan, Ya Cui, Weiqi Bao, Jie Hu, Jie Luo, Liang Chen","doi":"10.3174/ajnr.A8647","DOIUrl":"10.3174/ajnr.A8647","url":null,"abstract":"<p><strong>Background and purpose: </strong>Epilepsy, a globally prevalent neurologic disorder, necessitates precise identification of the epileptogenic zone (EZ) for effective surgical management. While the individual utilities of FDG-PET and flumazenil (FMZ)-PET have been demonstrated, their combined efficacy in localizing the epileptogenic zone remains underexplored. We aim to improve the noninvasive prediction of EZ in temporal lobe epilepsy (TLE) by combining FDG-PET and FMZ-PET with statistical feature extraction and machine learning.</p><p><strong>Materials and methods: </strong>This study included 20 drug-resistant patients with unilateral TLE (14 mesial TLE, 6 lateral TLE) and 2 control groups (<i>n</i> = 29 for FDG, <i>n</i> = 20 for FMZ). EZ of each patient was confirmed by postsurgical pathology and 1-year follow-up, while propagation zone (PZ) and noninvolved zone (NIZ) were derived from the epileptogenicity index based on presurgical stereo-encephalography (SEEG) monitoring. Whole brain PET scans were obtained with dual tracers [¹⁸F]FDG and [¹⁸F]FMZ on separate days, from which standard uptake value ratio (SUVR) was calculated by global mean scaling. Low-order statistical parameters of SUVRs and t-maps derived against control groups were extracted. Additionally, fused FDG and FMZ features were created by using arithmetic operations. Spearman correlation was used to investigate the associations between FDG and FMZ, while multiple linear regression analyses were used to explore the interaction effects of imaging features in predicting epileptogenicity. Crafted imaging features were used to train logistic regression models to predict EZ, whose performance was evaluated by using 10-fold cross-validation at ROI level, and leave-1-patient-out cross-validation at patient level.</p><p><strong>Results: </strong>FDG SUVR significantly decreased in EZ and PZ compared with NIZ, while FMZ SUVR in EZ significantly differed from PZ. Interaction effects were found between FDG and FMZ in their prediction of epileptogenicity. Fusion of FDG and FMZ provided the best prediction model with an area under the curve (AUC) of 0.86 [0.84-0.87] for EZ versus NIZ and an AUC of 0.79 [0.77-0.81] for EZ versus PZ, eliminating 100% false-positives in 50% of patients, and ≥80% FPs in 90% of patients at patient level.</p><p><strong>Conclusions: </strong>Combined FDG and FMZ offer a promising avenue for noninvasive localization of the epileptogenic zone in TLE, potentially refining surgical planning.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":"1493-1500"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142967374","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":"Quantitative Shape Irregularity and Density Heterogeneity of Preoperative Hematoma Can Predict Rebleeding following Minimally Invasive Catheter Evacuation for Intracerebral Hemorrhage.","authors":"Kaijiang Kang, Zeqiang Ji, Yang Du, Guangshuo Li, Jing Yan, Zeyu Ding, Yiming Shi, Yanfang Liu, Jianwei Wu, Xingquan Zhao","doi":"10.3174/ajnr.A8680","DOIUrl":"10.3174/ajnr.A8680","url":null,"abstract":"<p><strong>Background and purpose: </strong>Postoperative rebleeding is a critical factor associated with poor outcomes in patients with intracerebral hemorrhage (ICH) who undergo minimally invasive catheter evacuation (MICE) followed by thrombolysis. This study aimed to explore the association between quantitative shape irregularity and density heterogeneity of preoperative hematoma and rebleeding after MICE.</p><p><strong>Materials and methods: </strong>We analyzed patients with ICH who underwent MICE between February 2021 and January 2024. The surface regularity index (SRI) and density coefficient of variation (DCV) of the hematomas were obtained based on preoperative CT by using 3D Slicer software. Postoperative rebleeding was defined as a hematoma increase of >6 mL or >33% compared with the previous CT. The predictive value of shape irregularity (reflected by SRI) and density heterogeneity (reflected by DCV) for postoperative rebleeding were comprehensively analyzed.</p><p><strong>Results: </strong>In total, 240 patients were included, of whom 45 (18.8%) experienced postoperative rebleeding. Patients with postoperative rebleeding exhibited lower SRI (37.2 versus 51.4, <i>P</i> = .001) and higher DCV (13.8% versus 11.7%, <i>P</i> < .001) after adjusting for preoperative hematoma volume, surface area, standard deviation of hematoma density, intraventricular hemorrhage (IVH), hematoma expansion (HE), time period from onset to surgery, and catheter misplacement. The combination of SRI, DCV, IVH, and HE demonstrated optimal discrimination in predicting postoperative rebleeding, with an area under the curve (AUC) and 95% CI of 0.880 (0.824-0.935).</p><p><strong>Conclusions: </strong>Hematoma shape irregularity and density heterogeneity are risk factors for rebleeding after MICE for ICH. SRI and DCV can be used to identify individuals at high risk of postoperative rebleeding.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":"1360-1366"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144287444","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":"Reentry Technique for Rescue Recanalization of Carotid Near-Total Occlusion after Subintimal Penetration.","authors":"Dang Khoi Tran, Huu-Thao Nguyen, Chih-Wei Huang, Kai-Chen Chung, Yuang Seng Tsuei","doi":"10.3174/ajnr.A8860","DOIUrl":"10.3174/ajnr.A8860","url":null,"abstract":"","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":"1403"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144334624","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 Invasion Volume of the Primary Nasopharyngeal Carcinoma Predicts Treatment Outcome.","authors":"Qi Yong H Ai, Ho Sang Leung, Frankie K F Mo, Lun M Wong, Linfang Lan, Edwin P Hui, Brigette B Y Ma, Ann D King","doi":"10.3174/ajnr.A8678","DOIUrl":"10.3174/ajnr.A8678","url":null,"abstract":"<p><strong>Background and purpose: </strong>Quantification of deep invasion of the primary tumor is a predictor of outcome in oral cancer, but its predictive value in nasopharyngeal carcinoma (NPC) is unknown. This study aimed to investigate deep invasion of the primary NPC by using volumetric measurements on MRI for the prediction of outcome.</p><p><strong>Materials and methods: </strong>Retrospective review was conducted of 822 MRIs from patients with newly diagnosed nonmetastatic NPC with volumetric analysis of the primary tumor to obtain total primary tumor volume (PTV), deep invasion volume (DIV), and ratio of deep to the total primary tumor volume (DIVr). Optimal predictors were identified by the multivariable Cox regression and c-index correlating with disease-free survival (DFS), distant metastases-free survival (DMFS), and overall survival (OS).</p><p><strong>Results: </strong>High DIVr, DIV, and PTV significantly correlated with poor DFS, DMFS, and OS (all <i>P</i> < .01); DIVr being the optimal measurement (hazard ratio = 3.234 for DFS, 3.409 for DMFS, and 3.184 for OS). Compared with the eighth edition American Joint Committee on Cancer (AJCC) T-category, DIVr showed modest improvement in c-indexes for predicting DFS (0.602 versus 0.620, <i>P</i> = .03) and DMFS (0.597 versus 0.626, <i>P</i> < .01), but not OS (<i>P</i> = .15). The use of a DIVr-based T-category had similar survival prognostication to the eighth edition AJCC T-category although there was improved prediction in DMFS.</p><p><strong>Conclusions: </strong>DIVr is a better predictor of outcome in NPC than PTV or DIV, with slightly superior performance to the eighth edition AJCC T-category especially for DMFS.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":"1446-1452"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144509931","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":"MRI Interpretation Errors in Adult Patients with Medically Refractory Epilepsy.","authors":"Aoife M Haughey, Nadav Gasner, Timo Krings","doi":"10.3174/ajnr.A8681","DOIUrl":"10.3174/ajnr.A8681","url":null,"abstract":"<p><strong>Background and purpose: </strong>The aim is to determine the most common MRI interpretation errors in patients with medically refractory epilepsy (MRE) according to our experience in a tertiary academic center with a large volume epilepsy program, to raise awareness of the \"blind spots\" in imaging patients with MRE, and to highlight the importance of combining clinical and electroencephalographic information to obtain the most accurate diagnosis.</p><p><strong>Materials and methods: </strong>This is a retrospective observational study. All patients with MRE who underwent MRI brain (on 3T, with dedicated epilepsy protocol), and who were discussed at the weekly interdisciplinary epilepsy conference in our center between January 2008 and July 2023 were included. The initial MRI interpretation and final MRI interpretation results were reviewed.</p><p><strong>Results: </strong>A total of 886 patients with MRE were included. Three hundred patients were MRI-negative (33.86%), diagnoses were missed in 95 patients (10.7%), a second diagnosis was missed in patients with dual pathologies in 42 patients (4.74%), findings were misinterpreted in 4 patients (0.45%), an overcall was made in 2 patients (0.2%), and in 5 patients the correct diagnosis was discussed but erroneously deemed absent on initial interpretation (0.56%), resulting in a total of 148 discrepant MRI reports. Mesial temporal sclerosis comprised the most common pathology encountered overall, followed by encephalomalacia related to prior insult, enlarged amygdala (EA), malformations of cortical development, cavernoma, and ulegyria. The relative proportion of missed focal cortical dysplasia, encephalocele, and EA accounted for the most common misses/misdiagnoses.</p><p><strong>Conclusions: </strong>Evaluation of patients with MRE requires excellent interdisciplinary care. Input from all members of the interdisciplinary team is essential for accurate interpretation of MRI in patients with MRE for the neuroradiologist. We hope to inform radiologists of commonly overlooked pathologies in MRI brain interpretation for patients with epilepsy. In doing so, we want to maximize the yield of initial MRI interpretation in these patients.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":"1501-1509"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143257502","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":"Neuroimaging Spectrum of Erdheim-Chester Disease: An Image-Based Review.","authors":"Pranjal Rai, Hayden J Swartz, Neetu Soni, John C Benson, Amit Agarwal, Steven A Messina, Paul J Farnsworth, Carrie M Carr, Girish Bathla","doi":"10.3174/ajnr.A8599","DOIUrl":"10.3174/ajnr.A8599","url":null,"abstract":"<p><p>Erdheim-Chester disease (ECD) is a rare, multisystem histiocytic disorder characterized by its variable clinical presentations. CNS involvement is observed in approximately one-half of patients with ECD (up to 76% in some series) and often carries a poorer prognosis. While CNS involvement may remain asymptomatic, others may experience a range of neurologic symptoms, including cognitive decline, neuropsychiatric disturbances, motor deficits, cranial or peripheral neuropathies, and endocrine abnormalities. Neuroimaging findings in CNS-ECD are diverse, including neurodegeneration manifesting as cerebral or cerebellar volume loss; solitary or multifocal variably enhancing intraparenchymal lesions along the neuroaxis; meningeal infiltration; and involvement of the hypothalamo-pituitary axis, perivascular sheathing, or basal ganglia lesions. Other well-documented sites of involvement include the craniofacial region, orbits, and spine. Awareness of these findings is relevant, not only because of the nonspecific nature of these findings, but also because of the high proportion of CNS involvement in ECD and the higher mortality associated with CNS involvement. This review provides an in-depth overview of the various manifestations of CNS involvement in ECD and their imaging features, along with a brief overview of the differential considerations, which include other histiocytic and nonhistiocytic processes.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":"1300-1308"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142694074","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 MRI Models for the Differential Diagnosis of Tumefactive Demyelination versus <i>IDH</i> Wild-Type Glioblastoma.","authors":"Gian Marco Conte, Mana Moassefi, Paul A Decker, Matthew L Kosel, Christina B McCarthy, Jessica A Sagen, Yalda Nikanpour, Mahboubeh Fereidan-Esfahani, Michael W Ruff, Fiorella S Guido, Heather K Pump, Terry C Burns, Robert B Jenkins, Bradley J Erickson, Daniel H Lachance, W Oliver Tobin, Jeanette E Eckel-Passow","doi":"10.3174/ajnr.A8645","DOIUrl":"10.3174/ajnr.A8645","url":null,"abstract":"<p><strong>Background and purpose: </strong>Diagnosis of tumefactive demyelination can be challenging. The diagnosis of indeterminate brain lesions on MRI often requires tissue confirmation via brain biopsy. Noninvasive methods for accurate diagnosis of tumor and nontumor etiologies allows for tailored therapy, optimal tumor control, and a reduced risk of iatrogenic morbidity and mortality. Tumefactive demyelination has imaging features that mimic <i>isocitrate dehydrogenase</i> wild-type glioblastoma (<i>IDH</i>wt GBM). We hypothesized that deep learning applied to postcontrast T1-weighted (T1C) and T2-weighted (T2) MRI can discriminate tumefactive demyelination from <i>IDH</i>wt GBM.</p><p><strong>Materials and methods: </strong>Patients with tumefactive demyelination (<i>n</i> = 144) and <i>IDH</i>wt GBM (<i>n</i> = 455) were identified by clinical registries. A 3D DenseNet121 architecture was used to develop models to differentiate tumefactive demyelination and <i>IDH</i>wt GBM by using both T1C and T2 MRI, as well as only T1C and only T2 images. A 3-stage design was used: 1) model development and internal validation via 5-fold cross validation by using a sex-, age-, and MRI technology-matched set of tumefactive demyelination and <i>IDH</i>wt GBM, 2) validation of model specificity on independent <i>IDH</i>wt GBM, and 3) prospective validation on tumefactive demyelination and <i>IDH</i>wt GBM. Stratified area under the receiver operating curves (AUROCs) were used to evaluate model performance stratified by sex, age at diagnosis, MRI scanner strength, and MRI acquisition.</p><p><strong>Results: </strong>The deep learning model developed by using both T1C and T2 images had a prospective validation AUROC of 88% (95% CI: 0.82-0.95). In the prospective validation stage, a model score threshold of 0.28 resulted in 91% sensitivity of correctly classifying tumefactive demyelination and 80% specificity (correctly classifying <i>IDH</i>wt GBM). Stratified AUROCs demonstrated that model performance may be improved if thresholds were chosen stratified by age and MRI acquisition.</p><p><strong>Conclusions: </strong>MRI can provide the basis for applying deep learning models to aid in the differential diagnosis of brain lesions. Further validation is needed to evaluate how well the model generalizes across institutions, patient populations, and technology, and to evaluate optimal thresholds for classification. Next steps also should incorporate additional tumor etiologies such as CNS lymphoma and brain metastases.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":"1412-1420"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142960164","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}