AJNR. American journal of neuroradiology最新文献

{"title":"Enhanced Neonatal Brachial Plexus MR Neurography: A Comparative Analysis of Compressed SENSE versus SENSE.","authors":"Baiqi Zhu, Yu Guo, Xuehua Peng, Aiguo Zhai, Jian Li, Jianbo Shao","doi":"10.3174/ajnr.A8915","DOIUrl":"https://doi.org/10.3174/ajnr.A8915","url":null,"abstract":"<p><strong>Background and purpose: </strong>Neonatal brachial plexus imaging faces challenges with extended scan times and motion artifacts. This study assessed whether compressed sensitivity encoding acceleration could achieve image quality comparable to conventional Sensitivity Encoding while significantly reducing scanning time, potentially enhancing diagnostic accuracy and success rates in neonatal brachial plexus imaging.</p><p><strong>Materials and methods: </strong>45 neonates (18 males, 27 females; mean age 14.82±9.62 days) with clinical suspicion of brachial plexus nerve injury were examined using both compressed sensitivity encoding and Sensitivity Encoding 3D Nerve VIEW sequences on a 3.0T MRI scanner. The parallel acquisition acceleration factor was 1.3 for Sensitivity Encoding and 6 for compressed sensitivity encoding. Image quality was evaluated quantitatively using signal-to-noise ratio and nerve-to-muscle contrast-to-noise ratio, and qualitatively through a five-point semiquantitative scale assessment by two senior pediatric radiologists.</p><p><strong>Results: </strong>Compressed sensitivity encoding reduced acquisition time by approximately 30% (3:36 vs. 5:08 minutes) compared to sensitivity encoding, without compromising image quality. No significant differences were found in signal-to-noise ratio and nerveto-muscle contrast-to-noise ratio between compressed sensitivity encoding and sensitivity encoding, with equivalence testing confirming comparable image quality (signal-to-noise ratio: t(44) = 3.109, p = 0.002; nerve-to-muscle contrast-to-noise ratio: t(44) = 1.984, p = 0.03). Radiologists' subjective evaluations revealed no significant difference in image quality scores between CS and SENSE, with strong inter-rater agreement for both methods (compressed sensitivity encoding: κ = 0.773; sensitivity encoding: κ = 0.617).</p><p><strong>Conclusions: </strong>Implementation of compressed sensitivity encoding technology in 3D Nerve VIEW sequences for neonatal brachial plexus imaging is feasible and effective, providing image quality comparable to sensitivity encoding while significantly reducing scanning time. This advancement potentially improving patient outcomes through higher success rates in imaging examinations.</p><p><strong>Abbreviations: </strong>MRN = Magnetic resonance neurography; CS = Compressed Sensitivity Encoding; SENSE = Sensitivity Encoding; CNR = contrast-to-noise ratio; TOST = Two one-sided tests; SD = standard deviation.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144585849","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":"Prevalence of Endolymphatic Hydrops Herniation into the Semicircular Canals in Primary Hydropic Ear Disease: Correlation with Clinical and Radiologic Parameters.","authors":"Lisa M H de Pont, Berit M Verbist, Mark A van Buchem, Claire Bommeljé, Henk M Blom, Sebastiaan Hammer","doi":"10.3174/ajnr.A8765","DOIUrl":"https://doi.org/10.3174/ajnr.A8765","url":null,"abstract":"<p><strong>Background and purpose: </strong>Delayed postcontrast FLAIR MRI can be used to visualize endolymphatic hydrops (EH) and their herniation into the semicircular canals (SCCs), which has been linked to impaired caloric function. However, the prevalence and anatomic distribution of EH herniation and its clinical relevance remains unclear. The purpose of this study is to investigate the frequency and localization of EH herniation into the SCCs in patients with primary hydropic ear disease (HED) and to correlate these findings with clinical parameters.</p><p><strong>Materials and methods: </strong>This was a retrospective study evaluating 409 patients with MRI-confirmed primary HED for the presence and anatomic location of EH herniation into the SCCs. Findings were correlated with the severity of cochleovestibular EH, vertiginous symptoms, audiometric data, and caloric test results.</p><p><strong>Results: </strong>EH herniation into the SCCs was identified in 172 (42%) of patients with primary HED. The most frequent site of herniation was the nonampullated limb of the horizontal SCC (hSCC) (<i>n</i>=174), followed by the common limb of the posterior and superior SCCs (pSCC and sSCC, respectively) (<i>n</i>=114). EH herniation was significantly associated with a prolonged disease duration and a higher grade of vestibular and cochlear EH (<i>P</i> < .001 for all 3 analyses). Caloric testing revealed that EH herniation into the hSCC (hSCC-herniation) was associated with a higher incidence of vestibular hypofunction (<i>P</i> = .002) and a reduced maximum slow phase velocity (SPV_max) of the evoked nystagmus, compared with patients without hSCC-herniation (<i>P</i> < .001).</p><p><strong>Conclusions: </strong>EH herniation predominantly occurs in the nonampullated limb of the hSCC and is associated with prolonged disease duration, greater severity of EH, and a more pronounced impairment of caloric function.</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":"144562325","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":"Multiple Synchronous CSF-Venous Fistulas in Spontaneous Intracranial Hypotension: A Multi-Institutional Case Series.","authors":"Ajay A Madhavan, Timothy J Amrhein, Michelle L Kodet, Niklas Lutzen, Michael D Malinzak, Jeremy K Cutsforth-Gregory, Ian T Mark, Ivan Garza, Eike I Piechowiak, Lalani Carlton Jones","doi":"10.3174/ajnr.A8900","DOIUrl":"10.3174/ajnr.A8900","url":null,"abstract":"<p><p>CSF-venous fistulas are a common cause of spontaneous intracranial hypotension. Due to the more routine use of decubitus myelography and advancements in various imaging techniques, recognition of CSF-venous fistulas has increased in recent years. Most commonly, patients harbor only one fistula at the time of myelography (although additional de novo fistulas can arise after treatment). Occasionally, two synchronous CSF-venous fistulas may be seen on a single myelogram. The co-existence of more than two CSF-venous fistulas, however, is quite rare and has only been previously described in two instances. Here, we present a multi-institutional series of sixteen patients with three or more concurrently discovered CSF-venous fistulas, representing the largest cohort of such patients to date. We describe their clinical features, imaging findings, treatment approaches, and outcomes.ABBREVIATIONS: CVF = CSF-venous fistula; CB-CTM = cone beam CT myelogram; DSM = digital subtraction myelography; EID = energy integrating detector; SIH = spontaneous intracranial hypotension; PCD = photon counting detector.</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":"144509932","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":"Proximal Protection Devices for Carotid Artery Stent Placement: A Benchtop Assessment of Flow Reversal Performance.","authors":"Jiahui Li, Esref Alperen Bayraktar, Cem Bilgin, Yang Liu, Yigit Can Senol, Jonathan Cortese, Ramanathan Kadirvel, Waleed Brinjikji, David F Kallmes","doi":"10.3174/ajnr.A8664","DOIUrl":"10.3174/ajnr.A8664","url":null,"abstract":"<p><strong>Background and purpose: </strong>Proximal protection devices, such as TransCarotid Artery Revascularization (TCAR), aim to yield better outcomes in carotid artery stent placement (CAS) than distal protection devices by preventing plaque embolization to the brain. However, transfemoral catheters may not fully reverse flow from the external carotid artery (ECA) to the ICA. We assess a new balloon-sheath device, Femoral Flow Reversal Access for Carotid Artery Stent placement (FFRACAS), for this purpose.</p><p><strong>Materials and methods: </strong>The FFRACAS prototype (inner diameter [ID] = 0.117 inches; L = 80 cm) was compared with TCAR (ID = 0.104 inches, L = 30 cm) and Mo.Ma (ID = 0.083 inches, L = 90 cm) in a pulsatile flow model with blood simulant at 800 mL/min. Mo.Ma was used according to labeled instructions, with both CCA and ECA balloon inflation, without CCA-femoral vein shunt placement, and in an off-label fashion with single balloon occlusion in the CCA and shunt. Flow rates of the ICA, ECA, and shunt, when applicable, were monitored during CAS stages: CCA flow arrest, shunt activation, and stent delivery. Experiments were conducted under 2 ECA inflow conditions (-10 and -20 mL/min). Statistical comparison of ICA flow rates was conducted by using ANOVA and Tukey post hoc tests.</p><p><strong>Results: </strong>The on-label use of Mo.Ma maintained retrograde ICA flow (-0.3 mL/min) throughout CAS. On shunt activation, TCAR and FFRACAS reversed ICA flow similarly under low ECA inflow (ICA = -5.10 mL/min versus -4.83 mL/min; <i>P</i> = .349), but neither achieved ICA flow reversal under high ECA inflow or during stent delivery. Mo.Ma off-label use failed to reverse ICA flow.</p><p><strong>Conclusions: </strong>FFRACAS presents a potential alternative to TCAR, achieving similar degrees of flow reversal from a transfemoral approach to that achieved with the transcarotid approach. The Mo.Ma system reliably prevents anterograde flow in ICA during CAS.</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":"143025887","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":"Implications of Hydrocephalus on FDG-PET Statistical Parametric Mapping Analysis in Neurodegenerative Disease Evaluation.","authors":"Andres Ricaurte-Fajardo, Ana M Franceschi, Debra D'Angelo, Aliah McCalla, Miran Salgado, Moath Hamed, Brielle Intorcia, Carlyn Wisherop, Samantha A Keil, Anna S Nordvig, Joseph R Osborne, Gloria C Chiang, Jana Ivanidze","doi":"10.3174/ajnr.A8698","DOIUrl":"10.3174/ajnr.A8698","url":null,"abstract":"<p><strong>Background and purpose: </strong>FDG-PET is critical in the diagnosis of neurodegenerative disease. Quantitative analysis with statistical parametric mapping (SPM) has been shown to improve the diagnostic accuracy of FDG-PET and has been incorporated in clinical workflows. This study aimed to assess the effects of hydrocephalus on the accuracy of FDG-PET SPM analysis, focusing on the cingulate gyrus regions, which are of particular interest in dementia evaluation and are adjacent to the lateral ventricles.</p><p><strong>Materials and methods: </strong>In this retrospective institutional review board-approved study, patients who underwent brain FDG-PET/CT or PET/MRI were evaluated. Inclusion criteria were a clinical history of cognitive impairment/suspected neurodegenerative disease and MRI evidence of communicating hydrocephalus. Region-specific <i>z</i> scores for the anterior, middle, and posterior cingulate gyri (ACG, MCG, PCG), as well as for the cerebellum were generated using SPM analysis. Blinded expert qualitative assessment was performed for each anatomic region. κ coefficients were computed to evaluate the agreement between quantitative and qualitative results. Paired nonparametric <i>t</i> tests assessed <i>z</i> score differences between the cingulate and cerebellar regions.</p><p><strong>Results: </strong>The study included 48 patients (17 women; mean age, 76 years). SPM analysis found significantly lower cingulate <i>z</i> scores compared with the cerebellum [-4.3 (ACG), -6.9 (MCG), and -3.2 (PCG), -1.2 (cerebellum) <i>P</i> < .0001]. Similar results were observed in the signed-rank tests comparing cingulate regions with the cerebellum [ACG, -3.2 (SD, 2.1); MCG, -5.7 (SD, 3.6); PCG, -1.9 (SD, 2.4), <i>P</i> < .001 for all 3 cingulate regions]. κ coefficients indicated poor agreement between SPM and qualitative assessments (κ = 0.05-0.19, <i>P</i> values = .078-.479).</p><p><strong>Conclusions: </strong>Our study highlights hydrocephalus as an important pitfall of FDG-PET SPM, particularly when analyzing the cingulate regions, integral to the clinical evaluation of dementia. Awareness of this pitfall can improve diagnostic accuracy and thus improve clinical outcomes in this growing patient population.</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":"143416544","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":"Beyond Recanalization: Machine Learning-Based Insights into Post-Thrombectomy Vascular Morphology in Stroke Patients.","authors":"Aditi Deshpande, Kaveh Laksari, Pouya Tahsili-Fahadan, Lawrence L Latour, Marie Luby","doi":"10.3174/ajnr.A8909","DOIUrl":"https://doi.org/10.3174/ajnr.A8909","url":null,"abstract":"<p><p>Many stroke patients have poor outcomes despite successful endovascular therapy (EVT). We hypothesized that machine learning (ML)-based analysis of vascular changes post-EVT could identify macrovascular perfusion deficits such as residual hypoperfusion and distal emboli. Patients with anterior circulation large vessel occlusion (LVO) stroke, pre-and post-EVT MRI, and successful recanalization (mTICI 2b/3) were included. An ML algorithm extracted vascular features from pre-and 24-hour post-EVT MRA. A ≥100% increase in ipsilateral arterial branch length was considered significant. Perfusion deficits were defined using PWI, MTT, or distal clot presence; early neurological improvement (ENI) by a 24-hour NIHSS decrease ≥4 or NIHSS 0-1. Among 44 patients (median age 63), 71% had complete reperfusion. Those with distal clot had smaller arterial length increases (51% vs. 134%, p=0.05). ENI patients showed greater arterial length increases (161% vs. 67%, p=0.023). ML-based vascular analysis post-EVT correlates with perfusion deficits and may guide adjunctive therapy.ABBREVIATIONS: EVT = Endovascular Thrombectomy, LVO = Large Vessel Occlusion, ENI = Early Neurological Improvement, AIS = Acute Ischemic Stroke, mTICI = Modified Thrombolysis in Cerebral Infarction.</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":"144562322","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":"Predicting Ten-Year Clinical Outcomes in Multiple Sclerosis with Radiomics-Based Machine Learning Models.","authors":"Mario Tranfa, Maria Petracca, Renato Cuocolo, Lorenzo Ugga, Vincenzo Brescia Morra, Antonio Carotenuto, Andrea Elefante, Fabrizia Falco, Roberta Lanzillo, Marcello Moccia, Alessandra Scaravilli, Arturo Brunetti, Sirio Cocozza, Mario Quarantelli, Giuseppe Pontillo","doi":"10.3174/ajnr.A8912","DOIUrl":"https://doi.org/10.3174/ajnr.A8912","url":null,"abstract":"<p><strong>Background and purpose: </strong>Identifying patients with multiple sclerosis (pwMS) at higher risk of clinical progression is essential to inform clinical management. We aimed to build prognostic models using machine learning (ML) algorithms predicting long-term clinical outcomes based on a systematic mapping of volumetric, radiomic, and macrostructural disconnection features from routine brain MRI scans of pwMS.</p><p><strong>Materials and methods: </strong>In this longitudinal monocentric study, 3T structural MRI scans of pwMS were retrospectively analyzed. Based on a ten-year clinical follow-up (average duration=9.4±1.1 years), patients were classified according to confirmed disability progression (CDP) and cognitive impairment (CI) as assessed through the Expanded Disability Status Scale (EDSS) and the Brief International Cognitive Assessment of Multiple Sclerosis (BICAMS) battery, respectively. 3D-T1w and FLAIR images were automatically segmented to obtain volumes, disconnection scores (estimated based on lesion masks and normative tractography data), and radiomic features from 116 gray matter regions defined according to the Automated Anatomical Labelling (AAL) atlas. Three ML algorithms (Extra Trees, Logistic Regression, and Support Vector Machine) were used to build models predicting long-term CDP and CI based on MRI-derived features. Feature selection was performed on the training set with a multi-step process, and models were validated with a holdout approach, randomly splitting the patients into training (75%) and test (25%) sets.</p><p><strong>Results: </strong>We studied 177 pwMS (M/F = 51/126; mean±SD age: 35.2±8.7 years). Long-term CDP and CI were observed in 71 and 55 patients, respectively. Regarding the CDP class prediction analysis, the feature selection identified 13-, 12-, and 10-feature subsets obtaining an accuracy on the test set of 0.71, 0.69, and 0.67 for the Extra Trees, Logistic Regression, and Support Vector Machine classifiers, respectively. Similarly, for the CI prediction, subsets of 16, 17, and 19 features were selected, with 0.69, 0.64, and 0.62 accuracy values on the test set, respectively. There were no significant differences in accuracy between ML models for CDP (p=0.65) or CI (p=0.31).</p><p><strong>Conclusions: </strong>Building on quantitative features derived from conventional MRI scans, we obtained long-term prognostic models, potentially informing patients' stratification and clinical decision-making.</p><p><strong>Abbreviations: </strong>MS, multiple sclerosis; pwMS, people with MS; HC, healthy controls; ML, machine learning; DD, disease duration; EDSS, Expanded Disability Status Scale; TLV, total lesion volume; CDP, confirmed disability progression; CI, cognitive impairment; BICAMS, Brief International Cognitive Assessment of Multiple Sclerosis.</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":"144562324","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":"3rd Comprehensive Survey of the Neuroradiology Work Environment in the United States with Reported Trends in Clinical Work, Nonclinical Work, Errors, Burnout and Retirement.","authors":"James Y Chen, Srinivasan Vedantham, Frank J Lexa","doi":"10.3174/ajnr.A8913","DOIUrl":"https://doi.org/10.3174/ajnr.A8913","url":null,"abstract":"<p><strong>Background and purpose: </strong>With the workforce shortage in the United States, neuroadiologists' workloads are increasing with associated increase in burnout and interpretive errors. This article reports on an updated survey deployed to reexamine the United States' neuroradiology work environment, evaluating changes in key results from a prior survey.</p><p><strong>Materials and methods: </strong>A survey was deployed to subscribers of the American Journal of Neuroradiology. Selected measures included work hours, volume, subjectively reported errors and malpractice, burnout symptoms, participation in non-clinical activities, intention to retire early or change careers, preparation for early retirement or career change, availability of artificial intelligence (AI) tools and remote work.</p><p><strong>Results: </strong>Survey respondents (n = 113) included 57.5% with teaching responsibilities. There was a high prevalence of burnout with 79% reporting at least one symptom, despite an increasing percentage of respondents (50.8%) reporting the availability of advanced informatics or AI tools in their practices. More respondents who have AI tools reported anxiety (30/54, 55.6%) compared to those without AI (P=0.04). Being involved in or having a colleague involved in a malpractice suit as a primary defendant was reported by 33% of respondents and was associated with the burnout measure, having difficulty in relaxing after work (P=0.03). Part-time work, remote work hours or percentage, or after-hours remote work were not correlated with burnout (P>0.11). Need to be faster than optimal for interpreting and signing reports, poorly indicated orders, and increases in work hours, workdays, and risk for malpractice suits were correlated with burnout (P<0.05). Intent to retire early was reported by 38.6% of respondents and correlated with all burnout factors (P<0.04) and cutbacks in other non-clinical activities (P<0.003). Among respondents with intent to retire early or make a career change, 27.9% reported making specific preparations.</p><p><strong>Conclusions: </strong>Despite the increasing availability of AI tools, US neuroradiologists report high rates of burnout and high rates of intention and preparation to retire early in the face of increasing clinical workloads and workforce shortage. These results underscore the challenges facing the leaders of radiology practices in balancing the growing demand for radiology services and the available and incoming workforce.</p><p><strong>Abbreviations: </strong>AI＝ artificial intelligence; XYZ＝ definition.</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":"144562321","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":"Nomogram for Predicting 90-Day Outcomes in Patients with Acute Vertebrobasilar Artery Occlusion Undergoing Endovascular Treatment: A Multicenter Cohort Study.","authors":"Bofeng Bai, Shanshan Huang, Pan Liu, Mengxiang Wang, Cong Ning, Yannan Wang, Hong Shi, Jian Cui, Yongbin Li","doi":"10.3174/ajnr.A8648","DOIUrl":"10.3174/ajnr.A8648","url":null,"abstract":"<p><strong>Background and purpose: </strong>Acute vertebrobasilar artery occlusion (VBAO) is associated with high mortality and disability rates, and reliable prediction tools for post-endovascular therapy (EVT) outcomes remain limited. In this study, we aimed to develop and validate a novel Nomogram model for predicting 90-day unfavorable clinical outcomes in patients with acute VBAO after EVT by integrating clinical and MRI features.</p><p><strong>Materials and methods: </strong>This multicenter retrospective study analyzed data from 181 patients with vertebrobasilar artery occlusion eligible for endovascular therapy from 2 Chinese stroke centers. We developed a predictive model for unfavorable clinical outcomes (mRs score >3) by using the data of 125 patients from Stroke Center A (2019-2023). The model was constructed by using univariate and multivariate logistic regression analyses of clinical and MRI characteristics, with continuous variables dichotomized on the basis of receiver operating characteristic curve analysis. Internal validation used smooth bootstrapping, while external validation used 56 cases from Stroke Center B (2019-2023), ensuring model reliability and generalizability across diverse clinical settings.</p><p><strong>Results: </strong>Age, NIHSS baseline score, recanalization, novel posterior circulation scores, and MRA-based posterior circulation collateral scores were independent predictors of 90-day prognosis, which were used to create a nomogram model. Internal validation demonstrated excellent discriminative performance of the model (mean area under the curve, 0.92;95% CI, 0.91-0.93), while external validation further confirmed its robust generalizability (area under the curve, 0.88). The patients were effectively stratified into the low-risk and high-risk groups using the nomogram model.</p><p><strong>Conclusions: </strong>The posterior circulation collateral score was an independent predictor of prognosis. Our novel nomogram model, based on clinical and MRI characteristics, effectively predicts 90-day unfavorable clinical outcomes in patients with vertebrobasilar artery occlusion following endovascular therapy.</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":"142967440","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":"CONSeg: Voxelwise Uncertainty Quantification for Glioma Segmentation Using Conformal Prediction.","authors":"Danial Elyassirad, Benyamin Gheiji, Mahsa Vatanparast, Amir Mahmoud Ahmadzadeh, Shahriar Faghani","doi":"10.3174/ajnr.A8914","DOIUrl":"https://doi.org/10.3174/ajnr.A8914","url":null,"abstract":"<p><strong>Background and purpose: </strong>Accurate glioma segmentation has the potential to enhance clinical decision-making and treatment planning. Uncertainty quantification methods, including conformal prediction (CP), can enhance segmentation models reliability. CP quantifies uncertainty with statistical confidence guarantees. This study aims to use CP in glioma segmentation.</p><p><strong>Materials and methods: </strong>We used the publicly available UCSF and UPenn glioma datasets, with the UCSF dataset (495 cases) split into training (70%), validation (10%), calibration (10%), and test (10%) sets, and the UPenn dataset (147 cases) divided into external calibration (30%) and external test (70%) sets. A UNet model was trained, and its optimal threshold was set to 0.5 using prediction normalization. To apply CP, the conformal threshold was selected based on the internal/external calibration nonconformity score, and CP was subsequently applied to the internal/external test sets, with coverage -the proportion of true labels within prediction sets-reported for all. We defined the uncertainty ratio (UR) and assessed its correlation with the Dice score coefficient (DSC) and 95th percentile Hausdorff distance (HD95). Additionally, we categorized cases into certain and uncertain groups based on UR and compared their DSC and HD95. We also evaluate the correlation between UR and the evaluation metrics (DSC and HD95) of the BraTS fusion model segmentation (BFMS), and compare evaluation metrics in the certain and uncertain subgroups.</p><p><strong>Results: </strong>The base model achieved a DSC of 0.86 and 0.83, and an HD95 of 7.35 and 11.71 on the internal and external test sets, respectively. The CP coverage was 0.9982 for the internal test set and 0.9977 for the external test set. Statistical analysis showed significant correlations between UR and evaluation metrics for test sets (p values <0.001). Additionally, certain cases had significantly better evaluation metrics (higher DSC and lower HD95) than uncertain cases in test sets and the BFMS (p values <0.001).</p><p><strong>Conclusions: </strong>CP effectively quantifies uncertainty in glioma segmentation. Using CONSeg improves the reliability of segmentation models and enhances human-computer interaction. Additionally, CONSeg can identify uncertain cases and suggest them for manual segmentation.</p><p><strong>Abbreviations: </strong>CP = conformal prediction; UR = uncertainty ratio; DSC = Dice score coefficient; BFMS = BraTS fusion model segmentation; DL = deep learning; UQ = uncertainty quantification; BCE = binary cross-entropy; BMOT = base model optimal threshold; NCST = nonconformity score threshold; CONSeg = conformal segmentation; BMPN = base model prediction normalization.</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":"144562323","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}