Experienced Neuroradiologist Research Articles

Can the second phase of contrast-enhanced MRA of the neck provide additional information in the acute stroke setting?

Double-concentration magnetic resonance imaging (MRI) contrast agents are frequently used in contrast-enhanced MR angiography (CE-MRA) of the head and neck. To avoid mistiming the peak concentration of intraluminal contrast (due to shorter duration of peak), a second acquisition is sometimes performed. To evaluate additional information from the second acquisition of CE-MRA and compare the collateral scoring to the hypoperfusion index obtained on MR perfusion, and to investigate presence of pseudo-occlusion using the second phase of CE-MRA. A retrospective study was conducted. CE-MRA of the brain/neck, dynamic susceptibility contrast (DSC) MR perfusion scan (in majority) and subsequent digital subtraction angiography (DSA) were evaluated in patients with previous acute internal carotid artery (ICA)/middle cerebral artery (MCA) occlusion. Evaluation of CE-MRA/MR perfusion and DSA was performed by three experienced neuroradiologists and one neurointerventionist, respectively. The site of ICA occlusion was seen to be distal to the site noted on early arterial phase (pseudo-occlusion of ICA) in 28.5% of patients. A significant negative correlation was seen between a higher HIR and collateral score. Evaluation of second phase CE-MRA can provide valuable information that may be otherwise lost if only the early arterial phase is evaluated.

Contribution of automated hippocampal volume measurements to radiological diagnosis in cases of mesial temporal sclerosis

Aims: Epilepsy, particularly drug-resistant epilepsy in adults, is often caused by Mesial Temporal Sclerosis (MTS), which can develop after brain injury from febrile illnesses or trauma. Magnetic Resonance Imaging (MRI) is crucial for diagnosing MTS, although challenges such as patient movement and mild or bilateral hippocampal atrophy can complicate the diagnosis. T1-weighted hippocampal volume measurements are effective in detecting MTS, with recent software advancements enabling automatic hippocampal segmentation. This study compared the hippocampal volumes and indices between adults with MTS and a control group of similar age and sex in the Turkish population. This study aimed to highlight structural differences in the hippocampus associated with MTS. Methods: This study involved a retrospective review of cranial MRIs scans from patients with MTS, confirmed through histopathological examination after epilepsy surgery. To ensure unbiased comparisons, a control group was selected using propensity score matching by age and gender. Two experienced neuroradiologists independently assessed the MRIs findings for MTS without knowledge of hippocampal volumetric data. Hippocampal volumes were measured using FreeSurfer software and standardized using the hippocampal volume index (HVI) and hippocampal asymmetry index (HAI). Results: In our study of 38 patients, MTS was found in 55.2% of patients on the right side and in 44.8% of patients on the left side, with no bilateral cases. Visual MRI analysis identified MTS in 84.2% of patients, with an area under of curve (AUC) of 0.921. Automatic volumetry detected MTS in 23 patients with an AUC of 0.791. Combining both methods, MTS was diagnosed in 33 patients, with an AUC of 0.922. Conclusion: Automated volumetric analyses have been shown to enhance the detection of hippocampal volume loss in patients with MTS.

Aneurysm growth evaluation and detection: a computer-assisted follow-up MRA analysis

Growing intracranial aneurysms pose a high risk of rupture, making the detection and quantification of the growth crucial for timely treatment strategy adoption. In this paper we propose a computer-assisted approach based on the extraction of IA shapes from associated baseline and follow-up angiographic scans and non-rigid morphing of the two shapes. From the obtained shape deformations we computed four novel features, including differential volume (dV), surface area (dSA), aneurysm-size normalized median deformation path length (dMPL), and integral of cumulative deformation distances (dICDD). An experienced neuroradiologist manually extracted the IA shape models from the baseline and follow-up MRAs and, by utilizing size change and visual assessments, classified each aneurysm into stable with morphology changes, stable or growing. We investigated the classification performance and found that three of the novel and one cross-sectional feature exhibited significantly different mean values (p-value <0.05\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$< 0.05$$\\end{document}; Tukey’s HSD test) between the stable and growing IA groups, while the mean dICDD was significantly different between all the three groups. The cross-sectional features has sensitivity to growing IAs in range 0.05–0.86, while novel features had generally higher sensitivity in range 0.81–0.90, making them promising candidates as surrogate follow-up imaging-based biomarkers for IA growth detection. These findings may offer valuable information for clinical management of patients with IAs based on follow-up imaging.

MDB-91. MEDULLOBLASTOMA – MOLECULAR CHARACTERISTICS AND RADIOLOGICAL ASPECTS OF A BRAZILIAN COHORT FROM A SINGLE CENTER

Abstract BACKGROUND Four molecular subgroups of medulloblastoma are distinguished by their genomic, epidemiological, and prognostic characteristics: Wingless (WNT), Sonic hedgehog (SHH), Group 3 and Group 4. Genotypic classification permits individualized therapeutic planning based on risk stratification, thereby improving patient outcome. This study aims to characterize the epidemiological and molecular data of a cohort treated at a single Brazilian center, as well as to correlate the heterogeneous aspects of magnetic resonance imaging (MRI) with the molecular characteristics of this tumor. METHODS The study analyzed 47 medulloblastoma cases, treated at the children’s oncology reference institution of a middle-income country between 2011 and 2021. The medical records, radiological findings and the histological and molecular classification are reviewed. The images were evaluated by an experienced neuroradiologist, blinded to clinical and molecular information. RESULTS Thirty-two out 47 patients met the eligibility criteria and were analyzed. The frequency of molecular groups found was 6.25% Non-SHH/ Non-WNT, 12.50% WNT, 34.38% SHH, 3% Group 3 and 44 % Group 4. The analysis revealed tumor location, presence of hemorrhage/calcification, limits and enhancement patterns are significant for molecular group prediction. SHH shows a significant prevalence in the cerebellar hemisphere, lower presence of hemorrhage/calcification and an ill-defined tumor margins when compared with other subgroups. Group 4 showed significantly less enhancement. CONCLUSIONS In accordance with the frequency observed in studies on childhood medulloblastoma, our series found a significant prevalence of Group 4 and a lack of representativeness of Group 3. Despite the limitations, it was possible to identify radiological tumor characteristics that may predict the molecular subgroup and contribute to a more effective therapeutic approach, thereby reducing the incidence of post-surgical complications.

Spot sign score is associated with hematoma expansion and longer hospital stay but not functional outcomes in primary intracerebral hemorrhage survivors.

The computed tomography angiography (CTA) spot sign is a validated predictor of 30-day mortality in intracerebral hemorrhage (ICH). However, its role in predicting unfavorable functional outcomes remains unclear. This study explores the frequency of the spot sign and its association with functional outcomes, hematoma expansion, and length of hospital stay among survivors of ICH. This was a retrospective analysis of consecutive patients with primary ICH who received CTA within 24h of admission to two medical centers between January 2007 and August 2022. Patients who died before discharge and those referred from other hospitals were excluded. Spot signs were assessed by an experienced neuroradiologist. Functional outcomes were determined by modified Rankin Scale (mRS) scores and the Barthel Index (BI). In total, 98 patients were included; 14 (13.64%) had a spot sign. No significant differences were observed in the baseline characteristics between the patients with and without a spot sign. Higher spot sign scores were associated with higher odds of experiencing hematoma expansion (p = 0.013, 95% CI = 1.16-3.55), undergoing surgery (p = 0.012, 95% CI = 0.19-1.55), and having longer hospital stay (p = 0.02, 95% CI = 1.22-13.92). However, higher spot sign scores were not associated with unfavorable functional outcomes (p = 0.918 for BI, and p = 0.782 for mRS). Spot signs are common findings among patients with ICH, and higher spot sign scores were associated with subsequent hematoma expansion and longer hospital stays but not unfavorable functional outcomes.

MR Imaging Features of Critical Spinal Demyelinating Lesions Associated with Progressive Motor Impairment.

Progressive MS is typically heralded by a myelopathic pattern of asymmetric progressive motor weakness. Focal individual "critical" demyelinating spinal cord lesions anatomically associated with progressive motor impairment may be a compelling explanation for this clinical presentation as described in progressive solitary sclerosis (single CNS demyelinating lesion), progressive demyelination with highly restricted MR imaging lesion burden (2-5 total CNS demyelinating lesions; progressive paucisclerotic MS), and progressive, exclusively unilateral hemi- or monoparetic MS (>5 CNS demyelinating progressive unilateral hemi- or monoparetic MS [PUHMS] lesions). Critical demyelinating lesions appear strikingly similar across these cohorts, and we describe their specific spinal cord MR imaging characteristics. We performed a retrospective, observational MR imaging study comparing spinal cord critical demyelinating lesions anatomically associated with progressive motor impairment with any additional "noncritical" (not anatomically associated with progressive motor impairment) spinal cord demyelinating lesions. All spinal cord MR images (302 cervical and 91 thoracic) were reviewed by an experienced neuroradiologist with final radiologic assessment on the most recent MR imaging. Anatomic association with clinical progressive motor impairment was confirmed independently by MS subspecialists. Ninety-one individuals (PUHMS, 37 [41%], progressive paucisclerosis 35 [38%], progressive solitary sclerosis 19 [21%]) with 91 critical and 98 noncritical spinal cord MR imaging demyelinating lesions were evaluated. MR imaging characteristics that favored critical spinal cord demyelinating lesions over noncritical lesions included moderate-to-severe, focal, lesion-associated spinal cord atrophy: 41/91 (45%) versus 0/98 (0%) (OR, 161.91; 9.43 to >999.9); lateral column axial location (OR, 10.43; 3.88-28.07); central region (OR, 3.23; 1.78-5.88); ventral column (OR, 2.98; 1.55-5.72); and larger lesion size of the axial width (OR, 2.01;1.49-2.72), transverse axial size (OR, 1.66; 1.36-2.01), or lesion area (OR, 1.14; 1.08-1.2). Multiple regression analysis revealed focal atrophy and lateral axial location as having the strongest association with critical demyelinating lesions. Focal, lesion-associated atrophy, lateral column axial location, and larger lesion size are spinal cord MR imaging characteristics of critical demyelinating lesions. The presence of critical demyelinating lesions should be sought as these features may be associated with the development of progressive motor impairment in MS.

Speeding Up and Improving Image Quality in Glioblastoma MRI Protocol by Deep Learning Image Reconstruction.

A fully diagnostic MRI glioma protocol is key to monitoring therapy assessment but is time-consuming and especially challenging in critically ill and uncooperative patients. Artificial intelligence demonstrated promise in reducing scan time and improving image quality simultaneously. The purpose of this study was to investigate the diagnostic performance, the impact on acquisition acceleration, and the image quality of a deep learning optimized glioma protocol of the brain. Thirty-three patients with histologically confirmed glioblastoma underwent standardized brain tumor imaging according to the glioma consensus recommendations on a 3-Tesla MRI scanner. Conventional and deep learning-reconstructed (DLR) fluid-attenuated inversion recovery, and T2- and T1-weighted contrast-enhanced Turbo spin echo images with an improved in-plane resolution, i.e., super-resolution, were acquired. Two experienced neuroradiologists independently evaluated the image datasets for subjective image quality, diagnostic confidence, tumor conspicuity, noise levels, artifacts, and sharpness. In addition, the tumor volume was measured in the image datasets according to Response Assessment in Neuro-Oncology (RANO) 2.0, as well as compared between both imaging techniques, and various clinical-pathological parameters were determined. The average time saving of DLR sequences was 30% per MRI sequence. Simultaneously, DLR sequences showed superior overall image quality (all p < 0.001), improved tumor conspicuity and image sharpness (all p < 0.001, respectively), and less image noise (all p < 0.001), while maintaining diagnostic confidence (all p > 0.05), compared to conventional images. Regarding RANO 2.0, the volume of non-enhancing non-target lesions (p = 0.963), enhancing target lesions (p = 0.993), and enhancing non-target lesions (p = 0.951) did not differ between reconstruction types. The feasibility of the deep learning-optimized glioma protocol was demonstrated with a 30% reduction in acquisition time on average and an increased in-plane resolution. The evaluated DLR sequences improved subjective image quality and maintained diagnostic accuracy in tumor detection and tumor classification according to RANO 2.0.

Dual-Energy Computed Tomography Virtual Noncalcium Imaging of Intracranial Arteries in Acute Ischemic Stroke: Differentiation Between Acute Thrombus and Calcification.

Hyperdense artery sign (HAS) on noncontrast brain computed tomography (CT) indicates an acute thrombus within the cerebral artery. It is a valuable imaging biomarker for diagnosing large-vessel occlusion; however, its identification may be challenging with the presence of vascular calcification. Dual-energy CT virtual noncalcium (VNCa) imaging using a 3-material decomposition algorithm is helpful for differentiating between calcification and hemorrhage. This study aimed to clarify the potential of VNCa imaging for differentiating HAS from vascular calcification. Patients with acute ischemic stroke and large-vessel occlusion identified on MR angiography, who also underwent noncontrast dual-energy CT, were included. The 80 kV/Sn 140 kV mixed images, with a weighting factor of 0.4, were considered 120 kVp-equivalent images. Postprocessing using a 3-material decomposition algorithm to differentiate between calcium (Ca), cerebrospinal fluid, and hemorrhage was performed via a commercially available 3-dimensional workstation. A mixed image, VNCa image, color-coded Ca image, and color-coded Ca image with VNCa image overlay (color-coded Ca-overlay image) were obtained, and axial reconstruction with a 1-mm slice thickness was performed for each image type. Two experienced neuroradiologists conducted imaging evaluations in consensus. Thirty-four patients (mean age, 76.0 years; 21 male and 13 female patients) were included. The mixed and VNCa images revealed an HAS (indicating an acute clot) corresponding to the large-vessel occlusion site in 30 patients. Among them, the VNCa and color-coded Ca-overlay images enabled clear differentiation between the acute thrombus and adjacent vessel wall calcification in 5 patients. Among the other 4 patients, the VNCa, Ca-overlay, and Ca images identified calcified cerebral emboli in the M1 segment in 1 patient. For the other 3 patients, no high attenuation corresponding to magnetic resonance angiography findings was observed in any of the mixed, VNCa, Ca-overlay, or Ca images. VNCa and color-coded Ca-overlay images obtained via dual-energy brain CT enabled differentiation of acute thrombus from vessel wall calcification and calcified cerebral emboli in patients with acute ischemic stroke.

Amide proton transfer weighted and diffusion weighted imaging based radiomics classification algorithm for predicting 1p/19q co-deletion status in low grade gliomas

Background1p/19q co-deletion in low-grade gliomas (LGG, World Health Organization grade II and III) is of great significance in clinical decision making. We aim to use radiomics analysis to predict 1p/19q co-deletion in LGG based on amide proton transfer weighted (APTw), diffusion weighted imaging (DWI), and conventional MRI.MethodsThis retrospective study included 90 patients histopathologically diagnosed with LGG. We performed a radiomics analysis by extracting 8454 MRI-based features form APTw, DWI and conventional MR images and applied a least absolute shrinkage and selection operator (LASSO) algorithm to select radiomics signature. A radiomics score (Rad-score) was generated using a linear combination of the values of the selected features weighted for each of the patients. Three neuroradiologists, including one experienced neuroradiologist and two resident physicians, independently evaluated the MR features of LGG and provided predictions on whether the tumor had 1p/19q co-deletion or 1p/19q intact status. A clinical model was then constructed based on the significant variables identified in this analysis. A combined model incorporating both the Rad-score and clinical factors was also constructed. The predictive performance was validated by receiver operating characteristic curve analysis, DeLong analysis and decision curve analysis. P < 0.05 was statistically significant.ResultsThe radiomics model and the combined model both exhibited excellent performance on both the training and test sets, achieving areas under the curve (AUCs) of 0.948 and 0.966, as well as 0.909 and 0.896, respectively. These results surpassed the performance of the clinical model, which achieved AUCs of 0.760 and 0.766 on the training and test sets, respectively. After performing Delong analysis, the clinical model did not significantly differ in predictive performance from three neuroradiologists. In the training set, both the radiomic and combined models performed better than all neuroradiologists. In the test set, the models exhibited higher AUCs than the neuroradiologists, with the radiomics model significantly outperforming resident physicians B and C, but not differing significantly from experienced neuroradiologist.ConclusionsOur results suggest that our algorithm can noninvasively predict the 1p/19q co-deletion status of LGG. The predictive performance of radiomics model was comparable to that of experienced neuroradiologist, significantly outperforming the diagnostic accuracy of resident physicians, thereby offering the potential to facilitate non-invasive 1p/19q co-deletion prediction of LGG.

18F]FET PET/MRI: An Accurate Technique for Detection of Small Functional Pituitary Tumors.

Small functional pituitary tumors can cause severely disabling symptoms and early death. The gold standard diagnostic approach includes laboratory tests and MRI, with or without inferior petrosal sinus sampling (IPSS). In up to 40% of patients, however, the source of excess hormone production remains unidentified or uncertain. This excludes patients from surgical, Gamma Knife, and CyberKnife therapy and adversely affects overall cure rates. We here assess the diagnostic yield of O-(2-[18F]fluoroethyl)-l-tyrosine ([18F]FET) PET/MRI for detection of small functional pituitary tumors in these patients. Methods: This retrospective analysis included patients with Cushing disease (CD) but prior negative or inconclusive MRI results who underwent [18F]FET PET/MRI between February 1, 2021, and December 1, 2022. PET/MR images and MR images alone were evaluated by experienced nuclear radiologists, neuroradiologists, or radiologists. Postoperative tissue analysis (when performed) was used as a reference standard to assess diagnostic metrics (i.e., sensitivity and positive predictive value). Results were also compared with previously obtained MR images, preceding IPSS, and clinical or biochemical follow-up. Results: Twenty-two patients (68% female; mean age ± SD, 48 ± 15 y; range, 24-68 y) were scanned. All patients showed a clear metabolic focus on [18F]FET PET, whereas reading of the MRI alone yielded a suspected lesion in only 50%. Fifteen patients underwent surgery directed at the [18F]FET-positive focus. Tissue analysis confirmed a pituitary adenoma/pituitary neuroendocrine tumor of the corticotroph cell type (TPIT lineage) in 10 of 15 and a pituicytoma in 1 of 15, rendering a sensitivity of 100% and a positive predictive value of 73%. Lateralization was more accurate with [18F]FET PET/MRI than with IPSS in 33%. Twelve of 16 (75%) patients who received surgical, Gamma Knife, or CyberKnife therapy after [18F]FET PET/MRI reached short-term remission. Conclusion: [18F]FET PET/MRI shows a high diagnostic yield for localizing small functional pituitary tumors. This multimodal imaging technique provides a welcome improvement for diagnosis, planning of surgery, and clinical outcome in patients with Cushing disease, particularly those with repeated negative or inconclusive MRI results with or without IPSS.

Multiparametric MRI-Based Interpretable Radiomics Machine Learning Model Differentiates Medulloblastoma and Ependymoma in Children: A Two-Center Study

Rationale and ObjectivesMedulloblastoma (MB) and Ependymoma (EM) in children, share similarities in age group, tumor location, and clinical presentation. Distinguishing between them through clinical diagnosis is challenging. This study aims to explore the effectiveness of using radiomics and machine learning on multiparametric magnetic resonance imaging (MRI) to differentiate between MB and EM and validate its diagnostic ability with an external set. Materials and methodsAxial T2 weighted image (T2WI) and contrast-enhanced T1weighted image (CE-T1WI) MRI sequences of 135 patients from two centers were collected as train/test sets. Volume of interest (VOI) was manually delineated by an experienced neuroradiologist, supervised by a senior. Feature selection analysis and the least absolute shrinkage and selection operator (LASSO) algorithm identified valuable features, and Shapley additive explanations (SHAP) evaluated their significance. Five machine-learning classifiers- extreme gradient boosting (XGBoost), Bernoulli naive Bayes (Bernoulli NB), Logistic Regression (LR), support vector machine (SVM), linear support vector machine (Linear SVC) classifiers were built based on T2WI (T2 model), CE-T1WI (T1 model), and T1+T2WI (T1+T2 model). A human expert diagnosis was developed and corrected by senior radiologists. External validation was performed at Sun Yat-Sen University Cancer Center. Results31 valuable features were extracted from T2WI and CE-T1WI. XGBoost demonstrated the highest performance with an area under the curve (AUC) of 0.92 on the test set and maintained an AUC of 0.80 during external validation. For the T1 model, XGBoost achieved the highest AUC of 0.85 on the test set and the highest accuracy of 0.71 on the external validation set. In the T2 model, XGBoost achieved the highest AUC of 0.86 on the test set and the highest accuracy of 0.82 on the external validation set. The human expert diagnosis had an AUC of 0.66 on the test set and 0.69 on the external validation set. The integrated T1+T2 model achieved an AUC of 0.92 on the test set, 0.80 on the external validation set, achieved the best performance. Overall, XGBoost consistently outperformed in different classification models. ConclusionThe combination of radiomics and machine learning on multiparametric MRI effectively distinguishes between Medulloblastoma and Ependymoma in childhood, surpassing human expert diagnosis in training and testing sets.

A Retrospective Analysis of Temporal Lobe Gliosis after Middle Fossa Resection of Small Vestibular Schwannomas.

The middle cranial fossa (MCF) approach is a well-established procedure in surgery of the internal auditory canal, as well as with the retrosigmoid and translabyrinthine approaches. It is commonly used in the hearing-preserving microsurgery of small vestibular schwannomas (VS). The debate about the "best" approach for the microsurgery of small VS without contact to the brainstem is controversial. It has been stated that the MCF approach leads to irreversible damage to the temporal lobe, which may be evident in follow-up magnet resonance imaging (MRI) as gliosis in up to 70% of patients. This study represents a retrospective chart analysis conducted at a tertiary university hospital. Here, 76 postoperative MRIs were re-evaluated by an experienced neuroradiologist and compared with the preoperative images. Temporal lobe gliosis was classified on an ordinal scale as absent, slight, moderate or severe. Occurrence of gliosis was matched to the clinical predictors (tumor stage, tumor volume, sex, age, and side). No case of severe or moderate gliosis was found in the patient group. Slight gliosis of the temporal lobe was rare and was only detected in four patients (5%). There was no relation between clinical predictors and the incidence of gliosis. In our cohort, postoperative MR imaging did not reveal relevant damage to the temporal lobe parenchyma. This confirms the safe concept of microsurgery of small tumors via the middle fossa approach. The severe glioses described in other studies may be caused by a forced insertion of the retractor or by more extended approaches. However, further prospective neurocognitive studies seem to be necessary in order to assess functional changes in the temporal lobe.

Structural brain abnormalities in Pallister-Killian syndrome: a neuroimaging study of 31 children

BackgroundPallister-Killian syndrome (PKS) is a rare genetic disorder caused by mosaic tetrasomy of 12p with wide neurological involvement. Intellectual disability, developmental delay, behavioral problems, epilepsy, sleep disturbances, and brain malformations have been described in most individuals, with a broad phenotypic spectrum. This observational study, conducted through brain MRI scan analysis on a cohort of patients with genetically confirmed PKS, aims to systematically investigate the neuroradiological features of this syndrome and identify the possible existence of a typical pattern. Moreover, a literature review differentiating the different types of neuroimaging data was conducted for comparison with our population.ResultsThirty-one individuals were enrolled (17 females/14 males; age range 0.1–17.5 years old at first MRI). An experienced pediatric neuroradiologist reviewed brain MRIs, blindly to clinical data. Brain abnormalities were observed in all but one individual (compared to the 34% frequency found in the literature review). Corpus callosum abnormalities were found in 20/30 (67%) patients: 6 had callosal hypoplasia; 8 had global hypoplasia with hypoplastic splenium; 4 had only hypoplastic splenium; and 2 had a thin corpus callosum. Cerebral hypoplasia/atrophy was found in 23/31 (74%) and ventriculomegaly in 20/31 (65%). Other frequent features were the enlargement of the cisterna magna in 15/30 (50%) and polymicrogyria in 14/29 (48%). Conversely, the frequency of the latter was found to be 4% from the literature review. Notably, in our population, polymicrogyria was in the perisylvian area in all 14 cases, and it was bilateral in 10/14.ConclusionsBrain abnormalities are very common in PKS and occur much more frequently than previously reported. Bilateral perisylvian polymicrogyria was a main aspect of our population. Our findings provide an additional tool for early diagnosis.Further studies to investigate the possible correlations with both genotype and phenotype may help to define the etiopathogenesis of the neurologic phenotype of this syndrome.

Application of a computer vision algorithm to identify foci of demyelination in multiple sclerosis on MRI images

The research was aimed at analyzing modern algorithms for diagnosing lesions in multiple sclerosis on MRI images. Multiple sclerosis is a severe disease of the central nervous system and ranks first among the causes of disability in patients of young working age. In connection with the development of computer vision and machine learning technologies, the relevance of using these technologies for medical diagnostics is growing. Such approaches are necessary for the effective development and implementation of diagnostic systems using artificial intelligence. Modern algorithms and models for lesion segmentation were selected and implemented. Material and methods. The paper presents CV features of diagnosing multiple sclerosis on MRI images, existing data sets: ISBI-2015, MSSEG-2016, MSSEG-2021; existing algorithms and models for lesion segmentation: U-Net, nnU-Net, TransUnet, TransBTS, UNETR, Swin UNETR. Results and discussion. The architectures and models of nnU-Net, UNETR, Swin UNETR were trained and compared at ISBI2015 with various parameters and loss functions. Four MRI sequences were used: T2-WI, T2-FLAIR, PD, MPRAGE. Lesion segmentation was approved by certified experienced neuroradiologists. Conclusions. The approaches described in the paper including data processing, model training, and results analysis, focused on the selection and development of high-quality computer vision algorithms for identifying multiple sclerosis lesions in MRI images. Identification and segmentation of demyelination foci is a necessary step for diagnosing the disease, as well as for calculating and interpreting more meaningful indicators of disease severity and progression.

De novo formation of remote dural arteriovenous fistula following treated cavernous sinus dural arteriovenous fistula

BackgroundThe development of new dural arteriovenous fistulas (DAVFs) at another location following endovascular treatment of cavernous sinus DAVFs (CSDAVFs) are extremely rare. Our aim is to review cases of de Novo DAVFs that occurred after treatment of CSDAVFs at our institution and those reported in the literature. MethodsWe reviewed all cases of CSDAVFs evaluated by 2 experienced neuroradiologists. A literature search was performed using the PRISMA (preferred reporting items for systematic reviews and meta-analyses) guidelines focusing on De Novo DAVFs following the endovascular treatment of cerebrovascular malformations. Addition articles were searched through the reference lists of the included articles. ResultsFrom June 2004 and September 2019., we identified 3 (2.5%) cases of De Novo DAVFs occurred after endovascular treatment or spontaneous obliteration of CSDAVFs from 119 treated CSDAVFs at our institute. Our review yielded 9 articles involving 12 patients with 15 de novo DAVFs, including our 3 patients. The mean age was 55.08 ± 12.9 years (range 43–69), 83.3% were females (n = 10). The new remote DAVFs occurred after endovascular treatment of CSDAVFs in 10 (83.3%) patients. The de novo DAVFs occurred following spontaneous complete regression in 2 (16.7%) patients. All de novo DAVFs developed after complete obliteration of treated CSDAVFs. ConclusionSinus thrombosis and elevated venous pressure may play an important role in the pathogenesis of a de novo DAVF formation. In addition, thrombophilic abnormalities and the use of contraceptives may contribute to sinus thrombosis, leading to the development of the second remote DAVF after treatment of CSDAVFs.

Random expert sampling for deep learning segmentation of acute ischemic stroke on non-contrast CT.

Outlining acutely infarcted tissue on non-contrast CT is a challenging task for which human inter-reader agreement is limited. We explored two different methods for training a supervised deep learning algorithm: one that used a segmentation defined by majority vote among experts and another that trained randomly on separate individual expert segmentations. The data set consisted of 260 non-contrast CT studies in 233 patients with acute ischemic stroke recruited from the multicenter DEFUSE 3 (Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3) trial. Additional external validation was performed using 33 patients with matched stroke onset times from the University Hospital Lausanne. A benchmark U-Net was trained on the reference annotations of three experienced neuroradiologists to segment ischemic brain tissue using majority vote and random expert sampling training schemes. The median of volume, overlap, and distance segmentation metrics were determined for agreement in lesion segmentations between (1) three experts, (2) the majority model and each expert, and (3) the random model and each expert. The two sided Wilcoxon signed rank test was used to compare performances (1) to 2) and (1) to (3). We further compared volumes with the 24 hour follow-up diffusion weighted imaging (DWI, final infarct core) and correlations with clinical outcome (modified Rankin Scale (mRS) at 90 days) with the Spearman method. The random model outperformed the inter-expert agreement ((1) to (2)) and the majority model ((1) to (3)) (dice 0.51±0.04 vs 0.36±0.05 (P<0.0001) vs 0.45±0.05 (P<0.0001)). The random model predicted volume correlated with clinical outcome (0.19, P<0.05), whereas the median expert volume and majority model volume did not. There was no significant difference when comparing the volume correlations between random model, median expert volume, and majority model to 24 hour follow-up DWI volume (P>0.05, n=51). The random model for ischemic injury delineation on non-contrast CT surpassed the inter-expert agreement ((1) to (2)) and the performance of the majority model ((1) to (3)). We showed that the random model volumetric measures of the model were consistent with 24 hour follow-up DWI.

Correlation between neuroimaging, neurological phenotype, and functional outcomes in Wilson's disease.

Wilson's disease (WD) is associated with a variety of movement disorders and progressive neurological dysfunction. The aim of this study was to correlate baseline brain magnetic resonance imaging (MRI) features with clinical phenotype and long-term outcomes in chronically treated WD patients. Patients were retrospectively selected from an institutional database. Two experienced neuroradiologists reviewed baseline brain MRI. Functional assessment was performed using the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) scale, and disease severity was classified using the Global Assessment Scale for Wilson's Disease (GASWD). Of 27 patients selected, 14 were female (51.9%), with a mean (standard deviation [SD]) age at onset of 19.5 (7.1) years. Neurological symptoms developed in 22 patients (81.5%), with hyperkinetic symptoms being the most common (70.4%). Baseline brain MRI showed abnormal findings in 18 cases (66.7%), including T2 hyperintensities in 59.3% and atrophy in 29.6%. After a mean (SD) follow-up of 20.9 (11.0) years, WD patients had a mean score of 19.2 (10.2) on WHODAS 2.0 and 6.4 (5.7) on GASWD. The presence of hyperkinetic symptoms correlated with putaminal T2 hyperintensities (p = 0.003), putaminal T2 hypointensities (p = 0.009), and mesencephalic T2 hyperintensities (p = 0.009). Increased functional disability was associated with brain atrophy (p = 0.007), diffusion abnormalities (p = 0.013), and burden of T2 hyperintensities (p = 0.002). A stepwise regression model identified atrophy as a predictor of increased WHODAS 2.0 (p = 0.023) and GASWD (p = 0.007) scores. Atrophy and, to a lesser extent, deep T2 hyperintensity are associated with functional disability and disease severity in long-term follow-up of WD patients.

Prevalence of Possible Idiopathic Normal Pressure Hydrocephalus in Sweden: A Population-Based MRI Study in 791 70-Year-Old Participants

Very divergent prevalence rates for idiopathic normal pressure hydrocephalus (iNPH) are reported, probably due to differences in study sample selection and diagnostic criteria. This MRI-based study aimed to determine the prevalence of iNPH and iNPH-specific radiologic changes and their association with clinical symptoms in a large, 70-year-old population-based cohort (Gothenburg H70). In this cross-sectional study, disturbances in gait and balance, cognition, and urinary continence were assessed using clinical examination and self-report. MRI was evaluated for iNPH-specific imaging markers. iNPH was diagnosed according to International Guidelines (I.G.). Based on radiologic findings, participants were allocated to 1 of 4 groups: (A) Evans index (EI) ≤0.3 (reference), (B) EI >0.3 without other iNPH-typical radiologic findings, (C) radiologically probable iNPH according to I.G., and (D) radiologically holistically probable (h-probable) iNPH fulfilling radiologic criteria according to I.G. plus highly iNPH-specific changes according to an experienced neuroradiologist. The Gothenburg H70 Studies include 791 individuals (377 men, 414 women) born in 1944 who underwent brain MRI. The prevalence of iNPH was 1.5% (2.1% for men, 0.96% for women) according to I.G. Ninety participants (11%) had EI >0.3 without other iNPH-typical radiologic findings, 29 (3.7%) fulfilled the I.G. radiologic probable iNPH criteria alone, and 11 (1.4%) were classified as radiologically h-probable iNPH. Forty participants (5.1%) had I.G. radiologic features of iNPH (70% men vs 30% women, p = 0.005). Gait disturbances were more common in participants with EI >0.3 without other radiologic iNPH features (B) (33%) compared with the reference group (A) (19%) (p = 0.006). All clinical symptoms were more common in participants with I.G. radiologic features of iNPH (C + D) than they were in the reference group (A) (p < 0.03). The iNPH prevalence of 1.5% among 70-year-olds, which is considerably higher than earlier reported in this age group, suggests that iNPH may be more common than previously assumed. This is supported by the 5.1% total prevalence of imaging signs of iNPH. Ventriculomegaly without other iNPH-typical radiologic findings may be an early sign of developing iNPH in some patients.

A Convolutional Neural Network Model for Distinguishing Hemangioblastomas From Other Cerebellar-and-Brainstem Tumors Using Contrast-Enhanced MRI.

Hemangioblastoma (HB) is a highly vascularized tumor most commonly occurring in the posterior cranial fossa, requiring accurate preoperative diagnosis to avoid accidental intraoperative hemorrhage and even death. To accurately distinguish HBs from other cerebellar-and-brainstem tumors using a convolutional neural network model based on a contrast-enhanced brain MRI dataset. Retrospective. Four hundred five patients (182 = HBs; 223 = other cerebellar-and brainstem tumors): 305 cases for model training, and 100 for evaluation. 3 T/contrast-enhanced T1-weighted imaging (T1WI + C). A CNN-based 2D classification network was trained by using sliced data along the z-axis. To improve the performance of the network, we introduced demographic information, various data-augmentation methods and an auxiliary task to segment tumor region. Then, this method was compared with the evaluations performed by experienced and intermediate-level neuroradiologists, and the heatmap of deep feature, which indicates the contribution of each pixel to model prediction, was visualized by Grad-CAM for analyzing the misclassified cases. The Pearson chi-square test and an independent t-test were used to test for distribution difference in age and sex. And the independent t-test was exploited to evaluate the performance between experts and our proposed method. P value <0.05 was considered significant. The trained network showed a higher accuracy for identifying HBs (accuracy = 0.902 ± 0.031, F1 = 0.891 ± 0.035, AUC = 0.926 ± 0.040) than experienced (accuracy = 0.887 ± 0.013, F1 = 0.868 ± 0.011, AUC = 0.881 ± 0.008) and intermediate-level (accuracy = 0.827 ± 0.037, F1 = 0.768 ± 0.068, AUC = 0.810 ± 0.047) neuroradiologists. The recall values were 0.910 ± 0.050, 0.659 ± 0.084, and 0.828 ± 0.019 for the trained network, intermediate and experienced neuroradiologists, respectively. Additional ablation experiments verified the utility of the introduced demographic information, data augmentation, and the auxiliary-segmentation task. Our proposed method can successfully distinguish HBs from other cerebellar-and-brainstem tumors and showed diagnostic efficiency comparable to that of experienced neuroradiologists. 3 TECHNICAL EFFICACY: Stage 2.

Diagnosis of Unruptured Intracranial Aneurysms Using Proton-Density Magnetic Resonance Angiography: A Comparison With High-Resolution Time-of-Flight Magnetic Resonance Angiography.

Differentiating intracranial aneurysms from normal variants using CT angiography (CTA) or MR angiography (MRA) poses significant challenges. This study aimed to evaluate the efficacy of proton-density MRA (PD-MRA) compared to high-resolution time-of-flight MRA (HR-MRA) in diagnosing aneurysms among patients with indeterminate findings on conventional CTA or MRA. In this retrospective analysis, we included patients who underwent both PD-MRA and HR-MRA from August 2020 to July 2022 to assess lesions deemed indeterminate on prior conventional CTA or MRA examinations. Three experienced neuroradiologists independently reviewed the lesions using HR-MRA and PD-MRA with reconstructed voxel sizes of 0.253 mm3 or 0.23 mm3, respectively. A neurointerventionist established the gold standard with digital subtraction angiography. We compared the performance of HR-MRA, PD-MRA (0.253-mm3 voxel), and PD-MRA (0.23-mm3 voxel) in diagnosing aneurysms, both per lesion and per patient. The Fleiss kappa statistic was used to calculate inter-reader agreement. The study involved 109 patients (average age 57.4 ± 11.0 years; male:female ratio, 11:98) with 141 indeterminate lesions. Of these, 78 lesions (55.3%) in 69 patients were confirmed as aneurysms by the reference standard. PD-MRA (0.253-mm3 voxel) exhibited significantly higher per-lesion diagnostic performance compared to HR-MRA across all three readers: sensitivity ranged from 87.2%-91.0% versus 66.7%-70.5%; specificity from 93.7%-96.8% versus 58.7%-68.3%; and accuracy from 90.8%-92.9% versus 63.8%-69.5% (P ≤ 0.003). Furthermore, PD-MRA (0.253-mm3 voxel) demonstrated significantly superior per-patient specificity and accuracy compared to HR-MRA across all evaluators (P ≤ 0.013). The diagnostic accuracy of PD-MRA (0.23-mm3 voxel) surpassed that of HR-MRA and was comparable to PD-MRA (0.253-mm3 voxel). The kappa values for inter-reader agreements were significantly higher in PD-MRA (0.820-0.938) than in HR-MRA (0.447-0.510). PD-MRA outperformed HR-MRA in diagnostic accuracy and demonstrated almost perfect inter-reader consistency in identifying intracranial aneurysms among patients with lesions initially indeterminate on CTA or MRA.