Explore the vast range of titles available.

Brain atlases and templates are at the heart of neuroimaging analyses, for which they facilitate multimodal registration, enable group comparisons and provide anatomical reference. However, as atlas-based approaches rely on correspondence mapping between images they perform poorly in the presence of structural pathology. Whilst several strategies exist to overcome this problem, their performance is often dependent on the type, size and homogeneity of any lesions present. We therefore propose a new solution, referred to as Virtual Brain Grafting (VBG), which is a fully-automated, open-source workflow to reliably parcellate magnetic resonance imaging (MRI) datasets in the presence of a broad spectrum of focal brain pathologies, including large, bilateral, intra- and extra-axial, heterogeneous lesions with and without mass effect. The core of the VBG approach is the generation of a lesion-free T1-weighted image, which enables further image processing operations that would otherwise fail. Here we validated our solution based on Freesurfer recon-all parcellation in a group of 10 patients with heterogeneous gliomatous lesions, and a realistic synthetic cohort of glioma patients (n = 100) derived from healthy control data and patient data. We demonstrate that VBG outperforms a non-VBG approach assessed qualitatively by expert neuroradiologists and Mann-Whitney U tests to compare corresponding parcellations (real patients U(6,6) = 33, z = 2.738, P < .010, synthetic-patients U(48,48) = 2076, z = 7.336, P < .001). Results were also quantitatively evaluated by comparing mean dice scores from the synthetic-patients using one-way ANOVA (unilateral VBG = 0.894, bilateral VBG = 0.903, and non-VBG = 0.617, P < .001). Additionally, we used linear regression to show the influence of lesion volume, lesion overlap with, and distance from the Freesurfer volumes of interest, on labeling accuracy. VBG may benefit the neuroimaging community by enabling automated state-of-the-art MRI analyses in clinical populations using methods such as FreeSurfer, CAT12, SPM, Connectome Workbench, as well as structural and functional connectomics. To fully maximize its availability, VBG is provided as open software under a Mozilla 2.0 license (https://github.com/KUL-Radneuron/KUL_VBG). [Display omitted] (A) shows T1 images from two patients with gliomatous lesions. VBG is a lesion replacement/filling workflow with one approach for unilateral lesions (uVBG) and one for bilateral lesion (bVBG). (B) shows the lesion filling and recon-all combination selected, (C) & (D) show the output, tissue segmentations (C) and whole brain parcellations (D). If VBG is not used (non-VBG) recon-all may quit without generating a parcellation (hard failure) shown on the lower left, or finish with some errors (soft failures) in the parcellations shown on the lower right. However, using either VBG method allows recon-all to complete where it had previously failed and also improves parcellation quality. (PAT = patient, VBG = virtual brain grafting, uVBG = unilateral VBG, bVBG = bilateral VBG)

Objectives To investigate the cost-effectiveness of supplemental short-protocol brain MRI after negative non-contrast CT for the detection of minor strokes in emergency patients with mild and unspecific neurological symptoms. Methods The economic evaluation was centered around a prospective single-center diagnostic accuracy study validating the use of short-protocol brain MRI in the emergency setting. A decision-analytic Markov model distinguished the strategies “no additional imaging” and “additional short-protocol MRI” for evaluation. Minor stroke was assumed to be missed in the initial evaluation in 40% of patients without short-protocol MRI. Specialized post-stroke care with immediate secondary prophylaxis was assumed for patients with detected minor stroke. Utilities and quality-of-life measures were estimated as quality-adjusted life years (QALYs). Input parameters were obtained from the literature. The Markov model simulated a follow-up period of up to 30 years. Willingness to pay was set to $100,000 per QALY. Cost-effectiveness was calculated and deterministic and probabilistic sensitivity analysis was performed. Results Additional short-protocol MRI was the dominant strategy with overall costs of $26,304 (CT only: $27,109). Cumulative calculated effectiveness in the CT-only group was 14.25 QALYs (short-protocol MRI group: 14.31 QALYs). In the deterministic sensitivity analysis, additional short-protocol MRI remained the dominant strategy in all investigated ranges. Probabilistic sensitivity analysis results from the base case analysis were confirmed, and additional short-protocol MRI resulted in lower costs and higher effectiveness. Conclusion Additional short-protocol MRI in emergency patients with mild and unspecific neurological symptoms enables timely secondary prophylaxis through detection of minor strokes, resulting in lower costs and higher cumulative QALYs. Key Points • Short-protocol brain MRI after negative head CT in selected emergency patients with mild and unspecific neurological symptoms allows for timely detection of minor strokes. • This strategy supports clinical decision-making with regard to immediate initiation of secondary prophylactic treatment, potentially preventing subsequent major strokes with associated high costs and reduced QALY. • According to the Markov model, additional short-protocol MRI remained the dominant strategy over wide variations of input parameters, even when assuming disproportionally high costs of the supplemental MRI scan.

Background: Young children requiring neurosurgical care frequently undergo repeated neuroimaging. Whereas CT involves exposure to ionizing radiation, conventional MRI is time-consuming and often necessitates sedation in non-cooperative children. To address these limitations, ultrafast cranial MRI (UF-MRI) based on T2-HASTE sequences was implemented at our institution in 2019 for selected indications. The aim of this study was to evaluate the real-world implementation of UF-MRI in children younger than six years of age. Methods: We retrospectively analyzed cranial MRI examinations consisting exclusively of ultrafast sequences performed between July 2019 and December 2024 in children younger than six years. Clinical settings, diagnostic adequacy, immediate consequences for patient management, and the impact on MRI and CT utilization were systematically assessed. Results: A total of 404 UF-MRI examinations were performed in 198 inpatients and outpatients (mean age: 2 years 2 months) without the need for dedicated anesthesia team support solely for imaging. Only one examination (0.2%) required same-day repetition after mild oral sedation. In 20 patients (5.0%), UF-MRI was supplemented by conventional MRI under anesthesia, most commonly for preoperative planning. Immediate clinical consequences included no change in management in 54.5% of examinations, early follow-up in 22.8%, shunt valve adjustment in 11.6%, neurosurgical intervention in 7.7%, and other measures in 5.0%. UF-MRI accounted for 24.5% of all cranial MRI examinations in this age group and was associated with a 41% reduction in CT utilization compared with the corresponding period prior to UF-MRI implementation. Conclusions: In routine clinical practice, UF-MRI provides rapid, clinically sufficient neuroimaging in young children without the need for sedation or exposure to ionizing radiation. Its implementation significantly streamlines imaging workflows, optimizes resources utilization, reduces the need for CT, and supports timely clinical decision-making, underscoring its value as a complementary imaging modality in pediatric neuroimaging.

Cerebral microbleeds (CMBs) are commonly detected on MRI and have recently received an increased interest, because they are associated with vascular disease and dementia. Identification and rating of CMBs on MRI images may be facilitated by semi-automatic detection, particularly on high-resolution images acquired at high field strength. For these images, visual rating is time-consuming and has limited reproducibility. We present the radial symmetry transform (RST) as an efficient method for semi-automated CMB detection on 7.0T MR images, with a high sensitivity and a low number of false positives that have to be censored manually. The RST was computed on both echoes of a dual-echo T2*-weighted gradient echo 7.0T MR sequence in 18 participants from the Second Manifestations of ARTerial disease (SMART) study. Potential CMBs were identified by combining the output of the transform on both echoes. Each potential CMB identified through the RST was visually checked by two raters to identify probable CMBs. The scoring time needed to manually reject false positives was recorded. The sensitivity of 71.2% is higher than that of individual human raters on 7.0T scans and the required human rater time is reduced from 30 to 2minutes per scan on average. The RST outperforms published semi-automated methods in terms of either a higher sensitivity or less false positives, and requires much less human rater time. ► Cerebral microbleeds can be efficiently detected using the radial symmetry transform. ► 7T dual echo MRI allows detection of very small cerebral microbleeds (0.3–2.0mm). ► The sensitivity of the method (71.2%) outperforms individual raters (51.5% and 66.7%). ► Required human rater time is reduced from 30 to 2minutes per participant.

Background Brain CT can be used to evaluate pediatric patients with suspicion of cerebral pathology when anesthetic and MRI resources are scarce. This study aimed to assess if pediatric patients referred for an elective brain CT could endure a diagnostic fast brain MRI without general anesthesia using a one-minute multi-contrast EPI-based sequence (EPIMix) with comparable diagnostic performance. Methods Pediatric patients referred for an elective brain CT between March 2019 and March 2020 were prospectively included and underwent EPIMix without general anesthesia in addition to CT. Three readers (R1–3) independently evaluated EPIMix and CT images on two separate occasions. The two main study outcomes were the tolerance to undergo an EPIMix scan without general anesthesia and its performance to classify a scan as normal or abnormal. Secondary outcomes were assessment of disease category, incidental findings, diagnostic image quality, diagnostic confidence, and image artifacts. Further, a side-by-side evaluation of EPIMix and CT was performed. The signal-to-noise ratio (SNR) was calculated for EPIMix on T1-weighted, T2-weighted, and ADC images. Descriptive statistics, Fisher’s exact test, and Chi-squared test were used to compare the two imaging modalities. Results EPIMix was well tolerated by all included patients ( n  = 15) aged 5–16 (mean 11, SD 3) years old. Thirteen cases on EPIMix and twelve cases on CT were classified as normal by all readers (R1–3), while two cases on EPIMix and three cases on CT were classified as abnormal by one reader (R1), (R1–3, p  = 1.00). There was no evidence of a difference in diagnostic confidence, image quality, or the presence of motion artifacts between EPIMix and CT (R1–3, p  ≥ 0.10). Side-by-side evaluation (R2 + R4 + R5) reviewed all scans as lacking significant pathological findings on EPIMix and CT images. Conclusions Full brain MRI-based EPIMix sequence was well tolerated without general anesthesia with a diagnostic performance comparable to CT in elective pediatric patients. Trial registration This study was approved by the Swedish Ethical Review Authority (ethical approval number/ID Ethical approval 2017/2424-31/1). This study was a clinical trial study, with study protocol published at ClinicalTrials.gov with Trial registration number NCT03847051, date of registration 18/02/2019.

Background Differentiation between the grades of brain gliomas is a crucial step in the management of patients. The gold standard technique for grading is biopsy but MR imaging may play a more substantial role as a non-invasive method by using promising molecular sequences. Our purpose was to assess the added value of the relative amide proton transfer signal [rAPT] to advanced multiparametric MRI protocol. Methods We enrolled a pathologically confirmed 102 patients with low-grade glioma [n = 38] and high-grade glioma [n = 64] who underwent advanced multiparametric MRI protocol on the same scanner. The protocol included anatomic, diffusion, MRS, and perfusion sequences. The newly added sequence was Amide proton transfer. The rAPT values of all lesions were investigated by two neuroradiologists to assess the inter-rater agreement of using interclass correlation coefficient [ICC]. HGGs demonstrated significantly higher mean values of relative cerebral blood volume (rCBV), choline to creatine ratio (Cho/cr), and rAPT with lower Apparent diffusion coefficient (ADC) values compared to LGGs. ROC analyses revealed medium to high diagnostic performance with an AUC of 0.941 for rAPT, 0.907 for mean ADC, and 0.906 for rCBV. Discriminant function analysis of two models, the first one included mean ADC, rCBV, and Cho/Cr, while in the second Model, we added rAPT to them. Model two demonstrated higher accuracy and a significant difference in the AUC after adding the rAPT. The inter-rater agreement was reasonable (ICC 0.61). Conclusions rAPT adds significant value to multiparametric MRI for distinguishing LGG from HGG.

Brain tumors are among the deadliest diseases in the modern world. This study proposes an optimized machine-learning approach for the detection and identification of the type of brain tumor (glioma, meningioma, or pituitary tumor) in brain images recorded using magnetic resonance imaging (MRI). The Gaussian features of the image are extracted using speed-up robust features (SURF), whereas its non-linear features are obtained using KAZE, owing to their high performance against rotation, scaling, and noise problems. To retrieve local-level information, all brain MRI images are segmented into an 8 × 8 pixel grid. To enhance the accuracy and reduce the computational time, the variance-based k-means clustering and PSO-ReliefF algorithms are employed to eliminate the redundant features of the brain MRI images. Finally, the performance of the proposed hybrid optimized feature vector is evaluated using various machine learning classifiers. An accuracy of 96.30% is obtained with 169 features using a support vector machine (SVM). Furthermore, the computational time is also reduced to 1 min compared to the non-optimized features used for training of the SVM. The findings are also compared with previous research, demonstrating that the suggested approach might assist physicians and doctors in the timely detection of brain tumors.

With the advancement in technology, machine learning can be applied to diagnose the mass/tumor in the brain using magnetic resonance imaging (MRI). This work proposes a novel developed transfer deep-learning model for the early diagnosis of brain tumors into their subclasses, such as pituitary, meningioma, and glioma. First, various layers of isolated convolutional-neural-network (CNN) models are built from scratch to check their performances for brain MRI images. Then, the 22-layer, binary-classification (tumor or no tumor) isolated-CNN model is re-utilized to re-adjust the neurons’ weights for classifying brain MRI images into tumor subclasses using the transfer-learning concept. As a result, the developed transfer-learned model has a high accuracy of 95.75% for the MRI images of the same MRI machine. Furthermore, the developed transfer-learned model has also been tested using the brain MRI images of another machine to validate its adaptability, general capability, and reliability for real-time application in the future. The results showed that the proposed model has a high accuracy of 96.89% for an unseen brain MRI dataset. Thus, the proposed deep-learning framework can help doctors and radiologists diagnose brain tumors early.

Every year, millions of brain MRI scans are acquired in hospitals, which is a figure considerably larger than the size of any research dataset. Therefore, the ability to analyze such scans could transform neuroimaging research. Yet, their potential remains untapped since no automated algorithm is robust enough to cope with the high variability in clinical acquisitions (MR contrasts, resolutions, orientations, artifacts, and subject populations). Here, we present SynthSeg⁺, an AI segmentation suite that enables robust analysis of heterogeneous clinical datasets. In addition to wholebrain segmentation, SynthSeg⁺ also performs cortical parcellation, intracranial volume estimation, and automated detection of faulty segmentations (mainly caused by scans of very low quality). We demonstrate SynthSeg⁺ in seven experiments, including an aging study on 14,000 scans, where it accurately replicates atrophy patterns observed on data of much higher quality. SynthSeg⁺ is publicly released as a ready-to-use tool to unlock the potential of quantitative morphometry.

Background Unsupervised learning can discover various unseen abnormalities, relying on large-scale unannotated medical images of healthy subjects. Towards this, unsupervised methods reconstruct a 2D/3D single medical image to detect outliers either in the learned feature space or from high reconstruction loss. However, without considering continuity between multiple adjacent slices, they cannot directly discriminate diseases composed of the accumulation of subtle anatomical anomalies, such as Alzheimer’s disease (AD). Moreover, no study has shown how unsupervised anomaly detection is associated with either disease stages, various (i.e., more than two types of) diseases, or multi-sequence magnetic resonance imaging (MRI) scans. Results We propose unsupervised medical anomaly detection generative adversarial network (MADGAN), a novel two-step method using GAN-based multiple adjacent brain MRI slice reconstruction to detect brain anomalies at different stages on multi-sequence structural MRI: ( Reconstruction ) Wasserstein loss with Gradient Penalty + 100 ℓ 1 loss—trained on 3 healthy brain axial MRI slices to reconstruct the next 3 ones—reconstructs unseen healthy/abnormal scans; ( Diagnosis ) Average ℓ 2 loss per scan discriminates them, comparing the ground truth/reconstructed slices. For training, we use two different datasets composed of 1133 healthy T1-weighted (T1) and 135 healthy contrast-enhanced T1 (T1c) brain MRI scans for detecting AD and brain metastases/various diseases, respectively. Our self-attention MADGAN can detect AD on T1 scans at a very early stage, mild cognitive impairment (MCI), with area under the curve (AUC) 0.727, and AD at a late stage with AUC 0.894, while detecting brain metastases on T1c scans with AUC 0.921. Conclusions Similar to physicians’ way of performing a diagnosis, using massive healthy training data, our first multiple MRI slice reconstruction approach, MADGAN, can reliably predict the next 3 slices from the previous 3 ones only for unseen healthy images. As the first unsupervised various disease diagnosis, MADGAN can reliably detect the accumulation of subtle anatomical anomalies and hyper-intense enhancing lesions, such as (especially late-stage) AD and brain metastases on multi-sequence MRI scans.