Automated brain extraction of multisequence MRI using artificial neural networks

Longitudinal evidence for a mutually reinforcing relationship between white matter hyperintensities and cortical thickness in cognitively unimpaired older adults

BACKGROUND

For over three decades, the concomitance of cortical neurodegeneration and white matter hyperintensities (WMH) has sparked discussions about their coupled temporal dynamics. Longitudinal studies supporting this hypothesis nonetheless remain scarce.

METHODS

We applied global and regional bivariate latent growth curve modelling to determine the extent to which WMH and cortical thickness were interrelated over a four-year period. For this purpose, we leveraged longitudinal MRI data from 451 cognitively unimpaired participants (DELCODE; median age 69.71 [IQR 65.51, 75.50] years; 52.32% female). Participants underwent MRI sessions annually over a four-year period (1815 sessions in total, with roughly four MRI sessions per participant). We adjusted all models for demographics and cardiovascular risk.

RESULTS

Our findings were three-fold. First, larger WMH volumes were linked to lower cortical thickness (σ = -0.165, SE = 0.047, Z = -3.515, P < 0.001). Second, individuals with higher WMH volumes experienced more rapid cortical thinning (σ = -0.226, SE = 0.093, Z = -2.443, P = 0.007), particularly in temporal, cingulate, and insular regions. Similarly, those with lower initial cortical thickness had faster WMH progression (σ = -0.141, SE = 0.060, Z = -2.336, P = 0.009), with this effect being most pronounced in temporal, cingulate, and insular cortices. Third, faster WMH progression was associated with accelerated cortical thinning (σ = -0.239, SE = 0.139, Z = -1.710, P = 0.044), particularly in frontal, occipital, and insular cortical regions.

CONCLUSIONS

Our study suggests that cortical thinning and WMH progression could be mutually reinforcing rather than parallel, unrelated processes, which become entangled before cognitive deficits are detectable.

TRIAL REGISTRATION

German Clinical Trials Register (DRKS00007966, 04/05/2015).

SDS-Net: A Synchronized Dual-Stage Network for Predicting Patients Within 4.5-h Thrombolytic Treatment Window Using MRI

Timely and precise identification of acute ischemic stroke (AIS) within 4.5 h is imperative for effective treatment decision-making. This study aims to construct a novel network that utilizes limited datasets to recognize AIS patients within this critical window. We conducted a retrospective analysis of 265 AIS patients who underwent both fluid attenuation inversion recovery (FLAIR) and diffusion-weighted imaging (DWI) within 24 h of symptom onset. Patients were categorized based on the time since stroke onset (TSS) into two groups: TSS ≤ 4.5 h and TSS > 4.5 h. The TSS was calculated as the time from stroke onset to MRI completion. We proposed a synchronized dual-stage network (SDS-Net) and a sequential dual-stage network (Dual-stage Net), which were comprised of infarct voxel identification and TSS classification stages. The models were trained on 181 patients and validated on an independent external cohort of 84 patients using metrics of area under the curve (AUC), sensitivity, specificity, and accuracy. A DeLong test was used to statistically compare the performance of the two models. SDS-Net achieved an accuracy of 0.844 with an AUC of 0.914 in the validation dataset, outperforming the Dual-stage Net, which had an accuracy of 0.822 and an AUC of 0.846. In the external test dataset, SDS-Net further demonstrated superior performance with an accuracy of 0.800 and an AUC of 0.879, compared to the accuracy of 0.694 and AUC of 0.744 of Dual-stage Net (P = 0.049). SDS-Net is a robust and reliable tool for identifying AIS patients within a 4.5-h treatment window using MRI. This model can assist clinicians in making timely treatment decisions, potentially improving patient outcomes.

Impaired maturation of resting-state connectivity in anorexia nervosa from adolescence to adulthood: differential mechanisms of consummatory vs. anticipatory responses through a symptom provocation paradigm

This functional magnetic resonance imaging (fMRI) study examined resting-state (RS) connectivity in adolescent and adult patients with anorexia nervosa (AN) using symptom provocation paradigms. Differential food reward mechanisms were investigated through separate assessments of responses to food images and low-caloric/high-caloric food consumption. Thirteen young (≤ 21 years) and seventeen adult (> 21 years) patients with AN and age-matched controls underwent two stimulus-driven fMRI sessions involving RS scans before and after the presentation of food-related stimuli and food consumption. Graph theory and machine learning were used for analyzing the fMRI and clinical data. Healthy controls (HCs) showed widespread developmental changes, while young participants with AN exhibited cerebellum differences for high-calorie food. Young individuals with AN displayed increased connectivity during the consumption of potato chips compared to zucchini, with no differences in adults with AN. Multiparametric machine learning accurately distinguished young individuals with AN from healthy controls based on RS connectivity following food visual stimulation ("anticipatory") and consumption ("consummatory"). This study highlights the differential food reward mechanisms and minimal developmental changes in RS connectivity from youth to adulthood in individuals with AN compared to healthy controls. Young individuals with AN demonstrated heightened reactivity to high-caloric foods, while adults showed decreased responsiveness, potentially due to desensitization. These findings shed light on aberrant eating behaviors in individuals with AN and contribute to our understanding of the chronicity of the disease.

Perceptual super-resolution in multiple sclerosis MRI

Introduction

Magnetic resonance imaging (MRI) is crucial for diagnosing and monitoring of multiple sclerosis (MS) as it is used to assess lesions in the brain and spinal cord. However, in real-world clinical settings, MRI scans are often acquired with thick slices, limiting their utility for automated quantitative analyses. This work presents a single-image super-resolution (SR) reconstruction framework that leverages SR convolutional neural networks (CNN) to enhance the through-plane resolution of structural MRI in people with MS (PwMS).

Methods

Our strategy involves the supervised fine-tuning of CNN architectures, guided by a content loss function that promotes perceptual quality, as well as reconstruction accuracy, to recover high-level image features.

Results

Extensive evaluation with MRI data of PwMS shows that our SR strategy leads to more accurate MRI reconstructions than competing methods. Furthermore, it improves lesion segmentation on low-resolution MRI, approaching the performance achievable with high-resolution images.

Discussion

Results demonstrate the potential of our SR framework to facilitate the use of low-resolution retrospective MRI from real-world clinical settings to investigate quantitative image-based biomarkers of MS.

White matter alterations associated with chronic cannabis use disorder: a structural network and fixel-based analysis

Cannabis use disorder (CUD) is associated with adverse mental health effects, as well as social and cognitive impairment. Given prevalence rates of CUD are increasing, there is considerable efforts, and need, to identify prognostic markers which may aid in minimising any harm associated with this condition. Previous neuroimaging studies have revealed changes in white matter (WM) organization in people with CUD, though, the findings are mixed. In this study, we applied MRI-based analysis techniques that offer complimentary mechanistic insights, i.e., a connectome approach and fixel-based analysis (FBA) to investigate properties of individual WM fibre populations and their microstructure across the entire brain, providing a highly sensitive approach to detect subtle changes and overcome limitations of previous diffusion models. We compared 56 individuals with CUD (median age 25 years) to a sample of 38 healthy individuals (median age 31.5 years). Compared to controls, those with CUD had significantly increased structural connectivity strength (FDR corrected) across 9 edges between the right parietal cortex and several cortical and subcortical regions, including left orbitofrontal, left temporal pole, and left hippocampus and putamen. Utilizing FBA, WM density was significantly higher in those with CUD (FWE-corrected) across the splenium of the corpus callosum, and lower in the bilateral cingulum and right cerebellum. We observed significant correlation between cannabis use over the past month and connectivity strength of the frontoparietal edge, and between age of regular use and WM density of the bilateral cingulum and right cerebellum. Our findings enhance the understanding of WM architecture alterations associated with CUD.

Analysis and visualization of the effect of multiple sclerosis on biological brain age

Introduction

The rate of neurodegeneration in multiple sclerosis (MS) is an important biomarker for disease progression but can be challenging to quantify. The brain age gap, which quantifies the difference between a patient's chronological and their estimated biological brain age, might be a valuable biomarker of neurodegeneration in patients with MS. Thus, the aim of this study was to investigate the value of an image-based prediction of the brain age gap using a deep learning model and compare brain age gap values between healthy individuals and patients with MS.

Methods

A multi-center dataset consisting of 5,294 T1-weighted magnetic resonance images of the brain from healthy individuals aged between 19 and 89 years was used to train a convolutional neural network (CNN) for biological brain age prediction. The trained model was then used to calculate the brain age gap in 195 patients with relapsing remitting MS (20-60 years). Additionally, saliency maps were generated for healthy subjects and patients with MS to identify brain regions that were deemed important for the brain age prediction task by the CNN.

Results

Overall, the application of the CNN revealed accelerated brain aging with a larger brain age gap for patients with MS with a mean of 6.98 ± 7.18 years in comparison to healthy test set subjects (0.23 ± 4.64 years). The brain age gap for MS patients was weakly to moderately correlated with age at disease onset (ρ = -0.299, p < 0.0001), EDSS score (ρ = 0.206, p = 0.004), disease duration (ρ = 0.162, p = 0.024), lesion volume (ρ = 0.630, p < 0.0001), and brain parenchymal fraction (ρ = -0.718, p < 0.0001). The saliency maps indicated significant differences in the lateral ventricle (p < 0.0001), insula (p < 0.0001), third ventricle (p < 0.0001), and fourth ventricle (p = 0.0001) in the right hemisphere. In the left hemisphere, the inferior lateral ventricle (p < 0.0001) and the third ventricle (p < 0.0001) showed significant differences. Furthermore, the Dice similarity coefficient showed the highest overlap of salient regions between the MS patients and the oldest healthy subjects, indicating that neurodegeneration is accelerated in this patient cohort.

Discussion

In conclusion, the results of this study show that the brain age gap is a valuable surrogate biomarker to measure disease progression in patients with multiple sclerosis.

Leptomeningeal metastases in isocitrate dehydrogenase-wildtype glioblastomas revisited: Comprehensive analysis of incidence, risk factors, and prognosis based on post-contrast fluid-attenuated inversion recovery

BACKGROUND

The incidence of leptomeningeal metastases (LM) has been reported diversely. This study aimed to investigate the incidence, risk factors, and prognosis of LM in patients with isocitrate dehydrogenase (IDH)-wildtype glioblastoma.

METHODS

A total of 828 patients with IDH-wildtype glioblastoma were enrolled between 2005 and 2022. Baseline preoperative MRI including post-contrast fluid-attenuated inversion recovery (FLAIR) was used for LM diagnosis. Qualitative and quantitative features, including distance between tumor and subventricular zone (SVZ) and tumor volume by automatic segmentation of the lateral ventricles and tumor, were assessed. Logistic analysis of LM development was performed using clinical, molecular, and imaging data. Survival analysis was performed.

RESULTS

The incidence of LM was 11.4%. MGMTp unmethylation (odds ratio [OR] = 1.92, P = .014), shorter distance between tumor and SVZ (OR = 0.94, P = .010), and larger contrast-enhancing tumor volume (OR = 1.02, P < .001) were significantly associated with LM. The overall survival (OS) was significantly shorter in patients with LM than in those without (log-rank test; P < .001), with median OS of 12.2 and 18.5 months, respectively. The presence of LM remained an independent prognostic factor for OS in IDH-wildtype glioblastoma (hazard ratio = 1.42, P = .011), along with other clinical, molecular, imaging, and surgical prognostic factors.

CONCLUSIONS

The incidence of LM is high in patients with IDH-wildtype glioblastoma, and aggressive molecular and imaging factors are correlated with LM development. The prognostic significance of LM based on post-contrast FLAIR imaging suggests the acknowledgment of post-contrast FLAIR as a reliable diagnostic tool for clinicians.

Digital "flipbooks" for enhanced visual assessment of simple and complex brain tumors

Typical longitudinal radiographic assessment of brain tumors relies on side-by-side qualitative visualization of serial magnetic resonance images (MRIs) aided by quantitative measurements of tumor size. However, when assessing slowly growing tumors and/or complex tumors, side-by-side visualization and quantification may be difficult or unreliable. Whole-brain, patient-specific "digital flipbooks" of longitudinal scans are a potential method to augment radiographic side-by-side reads in clinical settings by enhancing the visual perception of changes in tumor size, mass effect, and infiltration across multiple slices over time. In this approach, co-registered, consecutive MRI scans are displayed in a slide deck, where one slide displays multiple brain slices of a single timepoint in an array (eg, 3 × 5 "mosaic" view of slices). The flipbooks are viewed similarly to an animated flipbook of cartoons/photos so that subtle radiographic changes are visualized via perceived motion when scrolling through the slides. Importantly, flipbooks can be created easily with free, open-source software. This article describes the step-by-step methodology for creating flipbooks and discusses clinical scenarios for which flipbooks are particularly useful. Example flipbooks are provided in Supplementary Material.

Scanner-specific optimisation of automated lesion segmentation in MS

BACKGROUND & OBJECTIVE

Automatic lesion segmentation techniques on MRI scans of people with multiple sclerosis (pwMS) could support lesion detection and segmentation in trials and clinical practice. However, knowledge on their reliability across scanners is limited, hampering clinical implementation. The aim of this study was to investigate the within-scanner repeatability and between-scanner reproducibility of lesion segmentation tools in pwMS across three different scanners and examine their accuracy compared to manual segmentations with and without optimization.

METHODS

30 pwMS underwent a scan and rescan on three MRI scanners. GE Discovery MR750 (3.0 T), Siemens Sola (1.5 T) and Siemens Vida (3.0 T)). 3D-FLuid Attenuated Inversion Recovery (3D-FLAIR) and 3D T1-weighted scans were acquired on each scanner. Lesion segmentation involved preprocessing and automatic segmentation using the Lesion Segmentation Toolbox (LST) and nicMSlesions (nicMS) as well as manual segmentation. Both automated segmentation techniques were used with default settings, and with settings optimized to match manual segmentations for each scanner specifically and combined for the three scanners. LST settings were optimized by adjusting the threshold to improve the Dice similarity coefficient (DSC) for each scanner separately and a combined threshold for all scanners. For nicMS the last layers were retrained, once with the multi-scanner data to represent a combined optimization and once separately for each scanner for scanner specific optimization. Volumes and counts were extracted. DSC was calculated for accuracy, and reliability was assessed using intra-class correlation coefficients (ICC). Differences in DSC between software was tested with a repeated measures ANOVA and when appropriate post-hoc paired t-tests using Bonferroni correction.

RESULTS

Scanner-specific optimization significantly improved DSC for LST compared to default and combined settings, except for the GE scanner. NicMS showed significantly higher DSC for both the scanner-specific and combined optimization than default. Within-scanner repeatability was excellent (ICC>0.9) for volume and counts. Between-scanner ICC for volume between Vida and Sola was higher (0.94-0.99) than between GE MR750 and Vida or Sola (0.18-0.93), with improved ICCs for nicMS scanner-specific (0.87-0.93) compared to others (0.18-0.79). This was not present for Sola vs. Vida where all ICCs were excellent (>0.94).

CONCLUSION

Scanner-specific optimization strategies proved effective in mitigating inter-scanner variability, addressing the issue of insufficient reproducibility and accuracy found with default settings.

Iron Deposition and Functional Connectivity Differences in Females With Migraine Without Aura: A Comparative Study of Headache Sides

BACKGROUND

The pathophysiological mechanisms underlying migraine without aura (MwoA) in females remain incompletely elucidated. Currently, the association between headache laterality and iron deposition (ID), and functional connectivity (FC) in female MwoA patients has not been fully studied.

METHODS

We prospectively recruited 63 female patients with MwoA and 31 matched healthy controls (HC) from the hospital. ID and FC among the four groups were analyzed using two-sample t-tests (with cluster-wise family-wise error [FWE] correction). Pearson correlation analysis was used to evaluate the relationships between clinical variables and both ID and FC values. Significance level: p < 0.05.

RESULTS

Compared to HC, left-sided MwoA exhibited differences in ID in various brain regions, including the cerebellum, left orbital inferior frontal gyrus, left calcarine gyrus, right putamen, and left caudate nucleus, as well as exhibited enhanced FC between the left lobule III of the cerebellum and the right superior temporal gyrus. Compared to bilateral MwoA, left-sided MwoA showed significantly enhanced in FC values in the left calcarine gyrus, the right precentral gyrus, the right postcentral gyrus, and the right lingual gyrus. Additionally, significant differences were observed in the Pearson correlations between clinical variables and both ID and FC in the female MwoA subgroups.

CONCLUSION

Our study provided preliminary evidence indicating significant differences in ID, FC, and correlations among subgroups of female MwoA. This provides neuroimaging references for further subclassifying MwoA patients. This offers valuable insights into potential pathophysiological mechanisms linked to the brain functional impairment in female MwoA.

Resolving spatial response heterogeneity in glioblastoma

PURPOSE

Spatial intratumoral heterogeneity poses a significant challenge for accurate response assessment in glioblastoma. Multimodal imaging coupled with advanced image analysis has the potential to unravel this response heterogeneity.

METHODS

Based on automated tumor segmentation and longitudinal registration with follow-up imaging, we categorized contrast-enhancing voxels of 61 patients with suspected recurrence of glioblastoma into either true tumor progression (TP) or pseudoprogression (PsP). To allow the unbiased analysis of semantically related image regions, adjacent voxels with similar values of cerebral blood volume (CBV), FET-PET, and contrast-enhanced T1w were automatically grouped into supervoxels. We then extracted first-order statistics as well as texture features from each supervoxel. With these features, a Random Forest classifier was trained and validated employing a 10-fold cross-validation scheme. For model evaluation, the area under the receiver operating curve, as well as classification performance metrics were calculated.

RESULTS

Our image analysis pipeline enabled reliable spatial assessment of tumor response. The predictive model reached an accuracy of 80.0% and a macro-weighted AUC of 0.875, which takes class imbalance into account, in the hold-out samples from cross-validation on supervoxel level. Analysis of feature importances confirmed the significant role of FET-PET-derived features. Accordingly, TP- and PsP-labeled supervoxels differed significantly in their 10th and 90th percentile, as well as the median of tumor-to-background normalized FET-PET. However, CBV- and T1c-related features also relevantly contributed to the model's performance.

CONCLUSION

Disentangling the intratumoral heterogeneity in glioblastoma holds immense promise for advancing precise local response evaluation and thereby also informing more personalized and localized treatment strategies in the future.

Altered Iron and Microstructure in Huntington's Disease Subcortical Nuclei: Insight From 7T MRI

BACKGROUND

Pathophysiological changes of Huntington's disease (HD) can precede symptom onset by decades. Robust imaging biomarkers are needed to monitor HD progression, especially before the clinical onset.

PURPOSE

To investigate iron dysregulation and microstructure alterations in subcortical regions as HD imaging biomarkers, and to associate such alterations with motor and cognitive impairments.

STUDY TYPE

Prospective.

POPULATION

Fourteen individuals with premanifest HD (38.0 ± 11.0 years, 9 females; far-from-onset N = 6, near-onset N = 8), 21 manifest HD patients (49.1 ± 12.1 years, 11 females), and 33 age-matched healthy controls (43.9 ± 12.2 years, 17 females).

FIELD STRENGTH/SEQUENCE

7 T, T1-weighted imaging, quantitative susceptibility mapping, and diffusion tensor imaging.

ASSESSMENT

Volume, susceptibility, fractional anisotropy (FA), and mean diffusivity (MD) within subcortical brain structures were compared across groups, used to establish HD classification models, and correlated to clinical measures and cognitive assessments.

STATISTICAL TESTS

Generalized linear model, multivariate logistic regression, receiver operating characteristics with the area under the curve (AUC), and likelihood ratio test comparing a volumetric model to one that also includes susceptibility and diffusion metrics, Wilcoxon paired signed-rank test, and Pearson's correlation. A P-value <0.05 after Benjamini-Hochberg correction was considered statistically significant.

RESULTS

Significantly higher striatal susceptibility and FA were found in premanifest and manifest HD preceding atrophy, even in far-from-onset premanifest HD compared to controls (putamen susceptibility: 0.027 ± 0.022 vs. 0.018 ± 0.013 ppm; FA: 0.358 ± 0.048 vs. 0.313 ± 0.039). The model with additional susceptibility, FA, and MD features showed higher AUC compared to volume features alone when differentiating premanifest HD from HC (0.83 vs. 0.66), and manifest from premanifest HD (0.94 vs. 0.83). Higher striatal susceptibility significantly correlated with cognitive deterioration in HD (executive function: r = -0.600; socioemotional function: r = -0.486).

DATA CONCLUSION

7 T MRI revealed iron dysregulation and microstructure alterations with HD progression, which could precede volume loss, provide added value to HD differentiation, and might be associated with cognitive changes.

EVIDENCE LEVEL

2 TECHNICAL EFFICACY: Stage 2.

How fiber bundle alterations differ in presumed LATE and amnestic Alzheimer's disease

INTRODUCTION

Typical Alzheimer's disease (AD) and limbic-predominant age-related TAR DNA-binding protein 43 (TDP-43) encephalopathy (LATE) are two neurodegenerative diseases that present with a similar initial amnestic clinical phenotype but are associated with distinct proteinopathies.

METHODS

We investigated white matter (WM) fiber bundle alterations, using fixel-based analysis, a state-of-the-art diffusion magnetic resonance imaging model, in early AD, presumed LATE, and controls. We also investigated regional cortical atrophy.

RESULTS

Both amnestic AD and presumed LATE patients exhibited WM alterations in tracts of the temporal and limbic lobes and in callosal fibers connecting superior frontal gyri. In addition, presumed LATE patients showed alterations in callosal fibers connecting the middle frontal gyri and in the cerebello-thalamo-cortical tract. Cortical thickness was reduced in regions connected by the most altered tracts.

DISCUSSION

These findings, the first to describe WM fiber bundle alterations in presumed LATE, are consistent with results on cortical atrophy and with the staging system of tau or TDP-43 accumulation.

HIGHLIGHTS

Fixel-based analysis revealed white matter (WM) fiber bundle alterations in presumed limbic-predominant age-related TAR DNA-binding protein 43 encephalopathy (LATE) patients identified by isolated episodic/limbic amnesia, the absence of positive Alzheimer's disease (AD) biomarkers, and no other neurological diagnosis after 2 years of follow-up. Presumed LATE and amnestic AD shared similar patterns of WM alterations in fiber bundles of the limbic and temporal lobes, in congruence with their similar limbic cognitive phenotype. Presumed LATE differed from AD by the alteration of the callosal fibers connecting the middle frontal gyri and of the cerebello-thalamo-cortical tract. WM fiber bundle alterations were consistent with results on regional cortical atrophy. The different anatomical patterns of WM degeneration could provide information on the propagation pathways of distinct proteinopathies.

Medical image foundation models in assisting diagnosis of brain tumors: a pilot study

OBJECTIVES

To build self-supervised foundation models for multicontrast MRI of the whole brain and evaluate their efficacy in assisting diagnosis of brain tumors.

METHODS

In this retrospective study, foundation models were developed using 57,621 enhanced head MRI scans through self-supervised learning with a pretext task of cross-contrast context restoration with two different content dropout schemes. Downstream classifiers were constructed based on the pretrained foundation models and fine-tuned for brain tumor detection, discrimination, and molecular status prediction. Metrics including accuracy, sensitivity, specificity, and area under the ROC curve (AUC) were used to evaluate the performance. Convolutional neural networks trained exclusively on downstream task data were employed for comparative analysis.

RESULTS

The pretrained foundation models demonstrated their ability to extract effective representations from multicontrast whole-brain volumes. The best classifiers, endowed with pretrained weights, showed remarkable performance with accuracies of 94.9, 92.3, and 80.4%, and corresponding AUC values of 0.981, 0.972, and 0.852 on independent test datasets in brain tumor detection, discrimination, and molecular status prediction, respectively. The classifiers with pretrained weights outperformed the convolutional classifiers trained from scratch by approximately 10% in terms of accuracy and AUC across all tasks. The saliency regions in the correctly predicted cases are mainly clustered around the tumors. Classifiers derived from the two dropout schemes differed significantly only in the detection of brain tumors.

CONCLUSIONS

Foundation models obtained from self-supervised learning have demonstrated encouraging potential for scalability and interpretability in downstream brain tumor-related tasks and hold promise for extension to neurological diseases with diffusely distributed lesions.

CLINICAL RELEVANCE STATEMENT

The application of our proposed method to the prediction of key molecular status in gliomas is expected to improve treatment planning and patient outcomes. Additionally, the foundation model we developed could serve as a cornerstone for advancing AI applications in the diagnosis of brain-related diseases.

Reproducible and Interpretable Machine Learning-Based Radiomic Analysis for Overall Survival Prediction in Glioblastoma Multiforme

PURPOSE

To develop and validate an MRI-based radiomic model for predicting overall survival (OS) in patients diagnosed with glioblastoma multiforme (GBM), utilizing a retrospective dataset from multiple institutions.

MATERIALS AND METHODS

Pre-treatment MRI images of 289 GBM patients were collected. From each patient's tumor volume, 660 radiomic features (RFs) were extracted and subjected to robustness analysis. The initial prognostic model with minimum RFs was subsequently enhanced by including clinical variables. The final clinical-radiomic model was derived through repeated three-fold cross-validation on the training dataset. Performance evaluation included assessment of concordance index (C-Index), integrated area under curve (iAUC) alongside patient stratification into low and high-risk groups for overall survival (OS).

RESULTS

The final prognostic model, which has the highest level of interpretability, utilized primary gross tumor volume (GTV) and one MRI modality (T2-FLAIR) as a predictor and integrated the age variable with two independent, robust RFs, achieving moderately good discriminatory performance (C-Index [95% confidence interval]: 0.69 [0.62-0.75]) with significant patient stratification (p = 7 × 10-5) on the validation cohort. Furthermore, the trained model exhibited the highest iAUC at 11 months (0.81) in the literature.

CONCLUSION

We identified and validated a clinical-radiomic model for stratification of patients into low and high-risk groups based on OS in patients with GBM using a multicenter retrospective dataset. Future work will focus on the use of deep learning-based features, with recently standardized convolutional filters on OS tasks.

Gray matter based spatial statistics framework in the 1-month brain: insights into gray matter microstructure in infancy

The neurodevelopmental epoch from fetal stages to early life embodies a critical window of peak growth and plasticity in which differences believed to be associated with many neurodevelopmental and psychiatric disorders first emerge. Obtaining a detailed understanding of the developmental trajectories of the cortical gray matter microstructure is necessary to characterize differential patterns of neurodevelopment that may subserve future intellectual, behavioral, and psychiatric challenges. The neurite orientation dispersion density imaging (NODDI) Gray-Matter Based Spatial Statistics (GBSS) framework leverages information from the NODDI model to enable sensitive characterization of the gray matter microstructure while limiting partial volume contamination and misregistration errors between images collected in different spaces. However, limited contrast of the underdeveloped brain poses challenges for implementing this framework with infant diffusion MRI (dMRI) data. In this work, we aim to examine the development of cortical microstructure in infants. We utilize the NODDI GBSS framework and propose refinements to the original framework that aim to improve the delineation and characterization of gray matter in the infant brain. Taking this approach, we cross-sectionally investigate age relationships in the developing gray matter microstructural organization in infants within the first month of life and reveal widespread relationships with the gray matter architecture.

Brainchop: Providing an Edge Ecosystem for Deployment of Neuroimaging Artificial Intelligence Models

Deep learning has proven highly effective in various medical imaging scenarios, yet the lack of an efficient distribution platform hinders developers from sharing models with end-users. Here, we describe brainchop, a fully functional web application that allows users to apply deep learning models developed with Python to local neuroimaging data from within their browser. While training artificial intelligence models is computationally expensive, applying existing models to neuroimaging data can be very fast; brainchop harnesses the end user's graphics card such that brain extraction, tissue segmentation, and regional parcellation require only seconds and avoids privacy issues that impact cloud-based solutions. The integrated visualization allows users to validate the inferences, and includes tools to annotate and edit the resulting segmentations. Our pure JavaScript implementation includes optimized helper functions for conforming volumes and filtering connected components with minimal dependencies. Brainchop provides a simple mechanism for distributing models for additional image processing tasks, including registration and identification of abnormal tissue, including tumors, lesions and hyperintensities. We discuss considerations for other AI model developers to leverage this open-source resource.

Radiomics-based prediction of local control in patients with brain metastases following postoperative stereotactic radiotherapy

BACKGROUND

Surgical resection is the standard of care for patients with large or symptomatic brain metastases (BMs). Despite improved local control after adjuvant stereotactic radiotherapy, the risk of local failure (LF) persists. Therefore, we aimed to develop and externally validate a pre-therapeutic radiomics-based prediction tool to identify patients at high LF risk.

METHODS

Data were collected from A Multicenter Analysis of Stereotactic Radiotherapy to the Resection Cavity of BMs (AURORA) retrospective study (training cohort: 253 patients from 2 centers; external test cohort: 99 patients from 5 centers). Radiomic features were extracted from the contrast-enhancing BM (T1-CE MRI sequence) and the surrounding edema (T2-FLAIR sequence). Different combinations of radiomic and clinical features were compared. The final models were trained on the entire training cohort with the best parameter set previously determined by internal 5-fold cross-validation and tested on the external test set.

RESULTS

The best performance in the external test was achieved by an elastic net regression model trained with a combination of radiomic and clinical features with a concordance index (CI) of 0.77, outperforming any clinical model (best CI: 0.70). The model effectively stratified patients by LF risk in a Kaplan-Meier analysis (P < .001) and demonstrated an incremental net clinical benefit. At 24 months, we found LF in 9% and 74% of the low and high-risk groups, respectively.

CONCLUSIONS

A combination of clinical and radiomic features predicted freedom from LF better than any clinical feature set alone. Patients at high risk for LF may benefit from stricter follow-up routines or intensified therapy.

Transient alteration of Awareness triggered by direct electrical stimulation of the brain

BACKGROUND

Awareness is a state of consciousness that enables a subject to interact with the environment. Transient alteration of awareness (AA) is a disabling sign of many types of epileptic seizures. The brain mechanisms of awareness and its alteration are not well known.

OBJECTIVE/HYPOTHESIS

Transient and isolated AA induced by electrical brain stimulation during a stereoelectroencephalography (SEEG) recording represents an ideal model for studying the associated modifications of functional connectivity and locating the hubs of awareness networks.

METHODS

We investigated the SEEG signals-based brain functional connectivity (FC) changes vs background occurring during AA triggered by three thalamic and two insular stimulations in three patients explored by SEEG in the frame of presurgical evaluation for focal drug-resistant epilepsy. The results were compared to the stimulations of the same sites that did not induce clinical changes (negative stimulations).

RESULTS

We observed decreased node strength in the pulvinar, insula, and parietal associative cortices during the thalamic and insular stimulations that induced AA. The link strengths characterizing functional coupling between the thalamus and the insular, prefrontal, temporal, or parietal associative cortices were also decreased. In contrast, there was an increased synchronization between the precuneus and the temporal lateral cortex. These FC changes were absent during the negative stimulations.

CONCLUSION

Our study highlights the role of the pulvinar, insular, and parietal hubs in maintaining the awareness networks and paves the way for invasive or non-invasive neuromodulation protocols to reduce AA manifestations during epileptic seizures.

Improved Dementia Prediction in Cerebral Small Vessel Disease Using Deep Learning-Derived Diffusion Scalar Maps From T1

BACKGROUND

Cerebral small vessel disease is the most common pathology underlying vascular dementia. In small vessel disease, diffusion tensor imaging is more sensitive to white matter damage and better predicts dementia risk than conventional magnetic resonance imaging sequences, such as T1 and fluid attenuation inversion recovery, but diffusion tensor imaging takes longer to acquire and is not routinely available in clinical practice. As diffusion tensor imaging-derived scalar maps-fractional anisotropy (FA) and mean diffusivity (MD)-are frequently used in clinical settings, one solution is to synthesize FA/MD from T1 images.

METHODS

We developed a deep learning model to synthesize FA/MD from T1. The training data set consisted of 4998 participants with the highest white matter hyperintensity volumes in the UK Biobank. Four external validations data sets with small vessel disease were included: SCANS (St George's Cognition and Neuroimaging in Stroke; n=120), RUN DMC (Radboud University Nijmegen Diffusion Tensor and Magnetic Resonance Imaging Cohort; n=502), PRESERVE (Blood Pressure in Established Cerebral Small Vessel Disease; n=105), and NETWORKS (n=26), along with 1000 normal controls from the UK Biobank.

RESULTS

The synthetic maps resembled ground-truth maps (structural similarity index >0.89 for MD maps and >0.80 for FA maps across all external validation data sets except for SCANS). The prediction accuracy of dementia using whole-brain median MD from the synthetic maps is comparable to the ground truth (SCANS ground-truth c-index, 0.822 and synthetic, 0.821; RUN DMC ground truth, 0.816 and synthetic, 0.812) and better than white matter hyperintensity volume (SCANS, 0.534; RUN DMC, 0.710).

CONCLUSIONS

We have developed a fast and generalizable method to synthesize FA/MD maps from T1 to improve the prediction accuracy of dementia in small vessel disease when diffusion tensor imaging data have not been acquired.

deepbet: Fast brain extraction of T1-weighted MRI using Convolutional Neural Networks

BACKGROUND

Brain extraction in magnetic resonance imaging (MRI) data is an important segmentation step in many neuroimaging preprocessing pipelines. Image segmentation is one of the research fields in which deep learning had the biggest impact in recent years. Consequently, traditional brain extraction methods are now being replaced by deep learning-based methods.

METHOD

Here, we used a unique dataset compilation comprising 7837 T1-weighted (T1w) MR images from 191 different OpenNeuro datasets in combination with advanced deep learning methods to build a fast, high-precision brain extraction tool called deepbet.

RESULTS

deepbet sets a novel state-of-the-art performance during cross-dataset validation with a median Dice score (DSC) of 99.0 on unseen datasets, outperforming the current best performing deep learning (DSC=97.9) and classic (DSC=96.5) methods. While current methods are more sensitive to outliers, deepbet achieves a Dice score of >97.4 across all 7837 images from 191 different datasets. This robustness was additionally tested in 5 external datasets, which included challenging clinical MR images. During visual exploration of each method's output which resulted in the lowest Dice score, major errors could be found for all of the tested tools except deepbet. Finally, deepbet uses a compute efficient variant of the UNet architecture, which accelerates brain extraction by a factor of ≈10 compared to current methods, enabling the processing of one image in ≈2 s on low level hardware.

CONCLUSIONS

In conclusion, deepbet demonstrates superior performance and reliability in brain extraction across a wide range of T1w MR images of adults, outperforming existing top tools. Its high minimal Dice score and minimal objective errors, even in challenging conditions, validate deepbet as a highly dependable tool for accurate brain extraction. deepbet can be conveniently installed via "pip install deepbet" and is publicly accessible at https://github.com/wwu-mmll/deepbet.

Repeatability of Lac+ measurements in healthy human brain at 3 T

PURPOSE

In vivo quantification of lactate has numerous applications in studying the pathology of both cerebral and musculoskeletal systems. Due to its low concentration (~0.5-1 mM), and overlap with lipid signals, traditional 1H MR spectra acquired in vivo using a small voxel and short echo time often result in an inadequate signal to detect and resolve the lactate peak, especially in healthy human volunteers.

METHODS

In this study, using a semi-LASER acquisition with long echo time (TE = 288 ms) and large voxel size (80 × 70 × 20 mm3), we clearly visualize the combined signal of lactate and threonine. Therefore, we call the signal at 1.33 ppm Lac+ and quantify Lac+ concentration from water suppressed spectra in healthy human brains in vivo. Four participants (22-37 years old; mean age = 28 ± 5.4; three male, one female) were scanned on four separate days, and on each day four measurements were taken. Intra-day values are calculated for each participant by comparing the four measurements on a single day. Inter-day values were calculated using the mean intra-day measurements.

RESULTS

The mean intra-participant Lac+ concentration, standard deviation (SD), and coefficient of variation (CV) ranged from 0.49 to 0.61 mM, 0.02 to 0.07 mM, and 4% to 13%, respectively, across four volunteers. The inter-participant Lac+ concentration, SD, and CV was 0.53 mM, ±0.06 mM, and 11%.

CONCLUSION

Repeatability is shown in Lac+ measurement in healthy human brain using a long echo time semi-LASER sequence with a large voxel in about 3.5 min at 3 T.

nnU-Net-based Segmentation of Tumor Subcompartments in Pediatric Medulloblastoma Using Multiparametric MRI: A Multi-institutional Study

Purpose To evaluate nnU-Net-based segmentation models for automated delineation of medulloblastoma tumors on multi-institutional MRI scans. Materials and Methods This retrospective study included 78 pediatric patients (52 male, 26 female), with ages ranging from 2 to 18 years, with medulloblastomas, from three different sites (28 from hospital A, 18 from hospital B, and 32 from hospital C), who had data available from three clinical MRI protocols (gadolinium-enhanced T1-weighted, T2-weighted, and fluid-attenuated inversion recovery). The scans were retrospectively collected from the year 2000 until May 2019. Reference standard annotations of the tumor habitat, including enhancing tumor, edema, and cystic core plus nonenhancing tumor subcompartments, were performed by two experienced neuroradiologists. Preprocessing included registration to age-appropriate atlases, skull stripping, bias correction, and intensity matching. The two models were trained as follows: (a) the transfer learning nnU-Net model was pretrained on an adult glioma cohort (n = 484) and fine-tuned on medulloblastoma studies using Models Genesis and (b) the direct deep learning nnU-Net model was trained directly on the medulloblastoma datasets, across fivefold cross-validation. Model robustness was evaluated on the three datasets when using different combinations of training and test sets, with data from two sites at a time used for training and data from the third site used for testing. Results Analysis on the three test sites yielded Dice scores of 0.81, 0.86, and 0.86 and 0.80, 0.86, and 0.85 for tumor habitat; 0.68, 0.84, and 0.77 and 0.67, 0.83, and 0.76 for enhancing tumor; 0.56, 0.71, and 0.69 and 0.56, 0.71, and 0.70 for edema; and 0.32, 0.48, and 0.43 and 0.29, 0.44, and 0.41 for cystic core plus nonenhancing tumor for the transfer learning and direct nnU-Net models, respectively. The models were largely robust to site-specific variations. Conclusion nnU-Net segmentation models hold promise for accurate, robust automated delineation of medulloblastoma tumor subcompartments, potentially leading to more effective radiation therapy planning in pediatric medulloblastoma. Keywords: Pediatrics, MR Imaging, Segmentation, Transfer Learning, Medulloblastoma, nnU-Net, MRI Supplemental material is available for this article. © RSNA, 2024 See also the commentary by Rudie and Correia de Verdier in this issue.

Deep learning ensembles for detecting brain metastases in longitudinal multi-modal MRI studies

Metastatic brain cancer is a condition characterized by the migration of cancer cells to the brain from extracranial sites. Notably, metastatic brain tumors surpass primary brain tumors in prevalence by a significant factor, they exhibit an aggressive growth potential and have the capacity to spread across diverse cerebral locations simultaneously. Magnetic resonance imaging (MRI) scans of individuals afflicted with metastatic brain tumors unveil a wide spectrum of characteristics. These lesions vary in size and quantity, spanning from tiny nodules to substantial masses captured within MRI. Patients may present with a limited number of lesions or an extensive burden of hundreds of them. Moreover, longitudinal studies may depict surgical resection cavities, as well as areas of necrosis or edema. Thus, the manual analysis of such MRI scans is difficult, user-dependent and cost-inefficient, and - importantly - it lacks reproducibility. We address these challenges and propose a pipeline for detecting and analyzing brain metastases in longitudinal studies, which benefits from an ensemble of various deep learning architectures originally designed for different downstream tasks (detection and segmentation). The experiments, performed over 275 multi-modal MRI scans of 87 patients acquired in 53 sites, coupled with rigorously validated manual annotations, revealed that our pipeline, built upon open-source tools to ensure its reproducibility, offers high-quality detection, and allows for precisely tracking the disease progression. To objectively quantify the generalizability of models, we introduce a new data stratification approach that accommodates the heterogeneity of the dataset and is used to elaborate training-test splits in a data-robust manner, alongside a new set of quality metrics to objectively assess algorithms. Our system provides a fully automatic and quantitative approach that may support physicians in a laborious process of disease progression tracking and evaluation of treatment efficacy.

Functional network modules overlap and are linked to interindividual connectome differences during human brain development

The modular structure of functional connectomes in the human brain undergoes substantial reorganization during development. However, previous studies have implicitly assumed that each region participates in one single module, ignoring the potential spatial overlap between modules. How the overlapping functional modules develop and whether this development is related to gray and white matter features remain unknown. Using longitudinal multimodal structural, functional, and diffusion MRI data from 305 children (aged 6 to 14 years), we investigated the maturation of overlapping modules of functional networks and further revealed their structural associations. An edge-centric network model was used to identify the overlapping modules, and the nodal overlap in module affiliations was quantified using the entropy measure. We showed a regionally heterogeneous spatial topography of the overlapping extent of brain nodes in module affiliations in children, with higher entropy (i.e., more module involvement) in the ventral attention, somatomotor, and subcortical regions and lower entropy (i.e., less module involvement) in the visual and default-mode regions. The overlapping modules developed in a linear, spatially dissociable manner, with decreased entropy (i.e., decreased module involvement) in the dorsomedial prefrontal cortex, ventral prefrontal cortex, and putamen and increased entropy (i.e., increased module involvement) in the parietal lobules and lateral prefrontal cortex. The overlapping modular patterns captured individual brain maturity as characterized by chronological age and were predicted by integrating gray matter morphology and white matter microstructural properties. Our findings highlight the maturation of overlapping functional modules and their structural substrates, thereby advancing our understanding of the principles of connectome development.

Repurposing the Public BraTS Dataset for Postoperative Brain Tumour Treatment Response Monitoring

The Brain Tumor Segmentation (BraTS) Challenge has been a main driver of the development of deep learning (DL) algorithms and provides by far the largest publicly available expert-annotated brain tumour dataset but contains solely preoperative examinations. The aim of our study was to facilitate the use of the BraTS dataset for training DL brain tumour segmentation algorithms for a postoperative setting. To this end, we introduced an automatic conversion of the three-label BraTS annotation protocol to a two-label annotation protocol suitable for postoperative brain tumour segmentation. To assess the viability of the label conversion, we trained a DL algorithm using both the three-label and the two-label annotation protocols. We assessed the models pre- and postoperatively and compared the performance with a state-of-the-art DL method. The DL algorithm trained using the BraTS three-label annotation misclassified parts of 10 out of 41 fluid-filled resection cavities in 72 postoperative glioblastoma MRIs, whereas the two-label model showed no such inaccuracies. The tumour segmentation performance of the two-label model both pre- and postoperatively was comparable to that of a state-of-the-art algorithm for tumour volumes larger than 1 cm3. Our study enables using the BraTS dataset as a basis for the training of DL algorithms for postoperative tumour segmentation.

HarmonyTM: multi-center data harmonization applied to distributed learning for Parkinson's disease classification

Purpose

Distributed learning is widely used to comply with data-sharing regulations and access diverse datasets for training machine learning (ML) models. The traveling model (TM) is a distributed learning approach that sequentially trains with data from one center at a time, which is especially advantageous when dealing with limited local datasets. However, a critical concern emerges when centers utilize different scanners for data acquisition, which could potentially lead models to exploit these differences as shortcuts. Although data harmonization can mitigate this issue, current methods typically rely on large or paired datasets, which can be impractical to obtain in distributed setups.

Approach

We introduced HarmonyTM, a data harmonization method tailored for the TM. HarmonyTM effectively mitigates bias in the model's feature representation while retaining crucial disease-related information, all without requiring extensive datasets. Specifically, we employed adversarial training to "unlearn" bias from the features used in the model for classifying Parkinson's disease (PD). We evaluated HarmonyTM using multi-center three-dimensional (3D) neuroimaging datasets from 83 centers using 23 different scanners.

Results

Our results show that HarmonyTM improved PD classification accuracy from 72% to 76% and reduced (unwanted) scanner classification accuracy from 53% to 30% in the TM setup.

Conclusion

HarmonyTM is a method tailored for harmonizing 3D neuroimaging data within the TM approach, aiming to minimize shortcut learning in distributed setups. This prevents the disease classifier from leveraging scanner-specific details to classify patients with or without PD-a key aspect for deploying ML models for clinical applications.

Next move in movement disorders: neuroimaging protocols for hyperkinetic movement disorders

Introduction

The Next Move in Movement Disorders (NEMO) study is an initiative aimed at advancing our understanding and the classification of hyperkinetic movement disorders, including tremor, myoclonus, dystonia, and myoclonus-dystonia. The study has two main objectives: (a) to develop a computer-aided tool for precise and consistent classification of these movement disorder phenotypes, and (b) to deepen our understanding of brain pathophysiology through advanced neuroimaging techniques. This protocol review details the neuroimaging data acquisition and preprocessing procedures employed by the NEMO team to achieve these goals.

Methods and analysis

To meet the study's objectives, NEMO utilizes multiple imaging techniques, including T1-weighted structural MRI, resting-state fMRI, motor task fMRI, and 18F-FDG PET scans. We will outline our efforts over the past 4 years to enhance the quality of our collected data, and address challenges such as head movements during image acquisition, choosing acquisition parameters and constructing data preprocessing pipelines. This study is the first to employ these neuroimaging modalities in a standardized approach contributing to more uniformity in the analyses of future studies comparing these patient groups. The data collected will contribute to the development of a machine learning-based classification tool and improve our understanding of disorder-specific neurobiological factors.

Ethics and dissemination

Ethical approval has been obtained from the relevant local ethics committee. The NEMO study is designed to pioneer the application of machine learning of movement disorders. We expect to publish articles in multiple related fields of research and patients will be informed of important results via patient associations and press releases.

Brain growth until adolescence after a neonatal focal injury: sex related differences beyond lesion effect

Introduction

Early focal brain injuries lead to long-term disabilities with frequent cognitive impairments, suggesting global dysfunction beyond the lesion. While plasticity of the immature brain promotes better learning, outcome variability across individuals is multifactorial. Males are more vulnerable to early injuries and neurodevelopmental disorders than females, but long-term sex differences in brain growth after an early focal lesion have not been described yet. With this MRI longitudinal morphometry study of brain development after a Neonatal Arterial Ischemic Stroke (NAIS), we searched for differences between males and females in the trajectories of ipsi- and contralesional gray matter growth in childhood and adolescence, while accounting for lesion characteristics.

Methods

We relied on a longitudinal cohort (AVCnn) of patients with unilateral NAIS who underwent clinical and MRI assessments at ages 7 and 16 were compared to age-matched controls. Non-lesioned volumes of gray matter (hemispheres, lobes, regions, deep structures, cerebellum) were extracted from segmented T1 MRI images at 7 (Patients: 23 M, 16 F; Controls: 17 M, 18 F) and 16 (Patients: 18 M, 11 F; Controls: 16 M, 15 F). These volumes were analyzed using a Linear Mixed Model accounting for age, sex, and lesion characteristics.

Results

Whole hemisphere volumes were reduced at both ages in patients compared to controls (gray matter volume: -16% in males, -10% in females). In ipsilesional hemisphere, cortical gray matter and thalamic volume losses (average -13%) mostly depended on lesion severity, suggesting diaschisis, with minimal effect of patient sex. In the contralesional hemisphere however, we consistently found sex differences in gray matter volumes, as only male volumes were smaller than in male controls (average -7.5%), mostly in territories mirroring the contralateral lesion. Females did not significantly deviate from the typical trajectories of female controls. Similar sex differences were found in both cerebellar hemispheres.

Discussion

These results suggest sex-dependent growth trajectories after an early brain lesion with a contralesional growth deficit in males only. The similarity of patterns at ages 7 and 16 suggests that puberty has little effect on these trajectories, and that most of the deviation in males occurs in early childhood, in line with the well-described perinatal vulnerability of the male brain, and with no compensation thereafter.

Contrast-Enhancing Lesion Segmentation in Multiple Sclerosis: A Deep Learning Approach Validated in a Multicentric Cohort

The detection of contrast-enhancing lesions (CELs) is fundamental for the diagnosis and monitoring of patients with multiple sclerosis (MS). This task is time-consuming and suffers from high intra- and inter-rater variability in clinical practice. However, only a few studies proposed automatic approaches for CEL detection. This study aimed to develop a deep learning model that automatically detects and segments CELs in clinical Magnetic Resonance Imaging (MRI) scans. A 3D UNet-based network was trained with clinical MRI from the Swiss Multiple Sclerosis Cohort. The dataset comprised 372 scans from 280 MS patients: 162 showed at least one CEL, while 118 showed no CELs. The input dataset consisted of T1-weighted before and after gadolinium injection, and FLuid Attenuated Inversion Recovery images. The sampling strategy was based on a white matter lesion mask to confirm the existence of real contrast-enhancing lesions. To overcome the dataset imbalance, a weighted loss function was implemented. The Dice Score Coefficient and True Positive and False Positive Rates were 0.76, 0.93, and 0.02, respectively. Based on these results, the model developed in this study might well be considered for clinical decision support.

Handling Missing MRI Data in Brain Tumors Classification Tasks: Usage of Synthetic Images vs. Duplicate Images and Empty Images

BACKGROUND

Deep-learning is widely used for lesion classification. However, in the clinic patient data often has missing images.

PURPOSE

To evaluate the use of generated, duplicate and empty(black) images for replacing missing MRI data in AI brain tumor classification tasks.

STUDY TYPE

Retrospective.

POPULATION

224 patients (local-dataset; low-grade-glioma (LGG) = 37, high-grade-glioma (HGG) = 187) and 335 patients (public-dataset (BraTS); LGG = 76, HGG = 259). The local-dataset was divided into training (64), validation (16), and internal-test-data (20), while the public-dataset was an independent test-set.

FIELD STRENGTH/SEQUENCE

T1WI, T1WI+C, T2WI, and FLAIR images (1.5T/3.0T-MR), obtained from different suppliers.

ASSESSMENT

Three image-to-image translation generative-adversarial-network (Pix2Pix-GAN) models were trained on the local-dataset, to generate T1WI, T2WI, and FLAIR images. The rating-and-preference-judgment assessment was performed by three human-readers (radiologist (MD) and two MRI-technicians). Resnet152 was used for classification, and inference was performed on both datasets, with baseline input, and with missing data replaced by 1) generated images; 2) duplication of existing images; and 3) black images.

STATISTICAL TESTS

The similarity between the generated and the original images was evaluated using the peak-signal-to-noise-ratio (PSNR) and the structural-similarity-index-measure (SSIM). Classification results were evaluated using accuracy, F1-score and the Kolmogorov-Smirnov test and distance.

RESULTS

For baseline-state, the classification model reached to accuracy = 0.93,0.82 on the local and public-datasets. For the missing-data methods, high similarity was obtained between the generated and the original images with mean PSNR = 35.65,32.94 and SSIM = 0.87,0.91 on the local and public-datasets; 39% of the generated-images were labeled as real images by the human-readers. The classification model using generated-images to replace missing images produced the highest results with mean accuracy = 0.91,0.82 compared to 0.85,0.79 for duplicated and 0.77,0.68 for use of black images; DATA CONCLUSION: The feasibility for inference classification model on an MRI dataset with missing images using the Pix2pix-GAN generated images, was shown. The stability and generalization ability of the model was demonstrated by producing consistent results on two independent datasets.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 5.

Synthesizing PET images from high-field and ultra-high-field MR images using joint diffusion attention model

BACKGROUND

Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) stand as pivotal diagnostic tools for brain disorders, offering the potential for mutually enriching disease diagnostic perspectives. However, the costs associated with PET scans and the inherent radioactivity have limited the widespread application of PET. Furthermore, it is noteworthy to highlight the promising potential of high-field and ultra-high-field neuroimaging in cognitive neuroscience research and clinical practice. With the enhancement of MRI resolution, a related question arises: can high-resolution MRI improve the quality of PET images?

PURPOSE

This study aims to enhance the quality of synthesized PET images by leveraging the superior resolution capabilities provided by high-field and ultra-high-field MRI.

METHODS

From a statistical perspective, the joint probability distribution is considered the most direct and fundamental approach for representing the correlation between PET and MRI. In this study, we proposed a novel model, the joint diffusion attention model, namely, the joint diffusion attention model (JDAM), which primarily focuses on learning information about the joint probability distribution. JDAM consists of two primary processes: the diffusion process and the sampling process. During the diffusion process, PET gradually transforms into a Gaussian noise distribution by adding Gaussian noise, while MRI remains fixed. The central objective of the diffusion process is to learn the gradient of the logarithm of the joint probability distribution between MRI and noise PET. The sampling process operates as a predictor-corrector. The predictor initiates a reverse diffusion process, and the corrector applies Langevin dynamics.

RESULTS

Experimental results from the publicly available Alzheimer's Disease Neuroimaging Initiative dataset highlight the effectiveness of the proposed model compared to state-of-the-art (SOTA) models such as Pix2pix and CycleGAN. Significantly, synthetic PET images guided by ultra-high-field MRI exhibit marked improvements in signal-to-noise characteristics when contrasted with those generated from high-field MRI data. These results have been endorsed by medical experts, who consider the PET images synthesized through JDAM to possess scientific merit. This endorsement is based on their symmetrical features and precise representation of regions displaying hypometabolism, a hallmark of Alzheimer's disease.

CONCLUSIONS

This study establishes the feasibility of generating PET images from MRI. Synthesis of PET by JDAM significantly enhances image quality compared to SOTA models.

Deep Learning-based Brain Age Prediction in Patients With Schizophrenia Spectrum Disorders

BACKGROUND AND HYPOTHESIS

The brain-predicted age difference (brain-PAD) may serve as a biomarker for neurodegeneration. We investigated the brain-PAD in patients with schizophrenia (SCZ), first-episode schizophrenia spectrum disorders (FE-SSDs), and treatment-resistant schizophrenia (TRS) using structural magnetic resonance imaging (sMRI).

STUDY DESIGN

We employed a convolutional network-based regression (SFCNR), and compared its performance with models based on three machine learning (ML) algorithms. We pretrained the SFCNR with sMRI data of 7590 healthy controls (HCs) selected from the UK Biobank. The parameters of the pretrained model were transferred to the next training phase with a new set of HCs (n = 541). The brain-PAD was analyzed in independent HCs (n = 209) and patients (n = 233). Correlations between the brain-PAD and clinical measures were investigated.

STUDY RESULTS

The SFCNR model outperformed three commonly used ML models. Advanced brain aging was observed in patients with SCZ, FE-SSDs, and TRS compared to HCs. A significant difference in brain-PAD was observed between FE-SSDs and TRS with ridge regression but not with the SFCNR model. Chlorpromazine equivalent dose and cognitive function were correlated with the brain-PAD in SCZ and FE-SSDs.

CONCLUSIONS

Our findings indicate that there is advanced brain aging in patients with SCZ and higher brain-PAD in SCZ can be used as a surrogate marker for cognitive dysfunction. These findings warrant further investigations on the causes of advanced brain age in SCZ. In addition, possible psychosocial and pharmacological interventions targeting brain health should be considered in early-stage SCZ patients with advanced brain age.

Methodological challenges of measuring brain volumes and cortical thickness in idiopathic normal pressure hydrocephalus with a surface-based approach

Identifying disease-specific imaging features of idiopathic Normal Pressure Hydrocephalus (iNPH) is crucial to develop accurate diagnoses, although the abnormal brain anatomy of patients with iNPH creates challenges in neuroimaging analysis. We quantified cortical thickness and volume using FreeSurfer 7.3.2 in 19 patients with iNPH, 28 patients with Alzheimer's disease (AD), and 30 healthy controls (HC). We noted the frequent need for manual correction of the automated segmentation in iNPH and examined the effect of correction on the results. We identified statistically significant higher proportion of volume changes associated with manual edits in individuals with iNPH compared to both HC and patients with AD. Changes in cortical thickness and volume related to manual correction were also partly correlated with the severity of radiological features of iNPH. We highlight the challenges posed by the abnormal anatomy in iNPH when conducting neuroimaging analysis and emphasise the importance of quality checking and correction in this clinical population.

A Multi-Center, Multi-Parametric MRI Dataset of Primary and Secondary Brain Tumors

Brain metastases (BMs) and high-grade gliomas (HGGs) are the most common and aggressive types of malignant brain tumors in adults, with often poor prognosis and short survival. As their clinical symptoms and image appearances on conventional magnetic resonance imaging (MRI) can be astonishingly similar, their accurate differentiation based solely on clinical and radiological information can be very challenging, particularly for "cancer of unknown primary", where no systemic malignancy is known or found. Non-invasive multiparametric MRI and radiomics offer the potential to identify these distinct biological properties, aiding in the characterization and differentiation of HGGs and BMs. However, there is a scarcity of publicly available multi-origin brain tumor imaging data for tumor characterization. In this paper, we introduce a multi-center, multi-origin brain tumor MRI (MOTUM) imaging dataset obtained from 67 patients: 29 with high-grade gliomas, 20 with lung metastases, 10 with breast metastases, 2 with gastric metastasis, 4 with ovarian metastasis, and 2 with melanoma metastasis. This dataset includes anonymized DICOM files alongside processed FLAIR, T1-weighted, contrast-enhanced T1-weighted, T2-weighted sequences images, segmentation masks of two tumor regions, and clinical data. Our data-sharing initiative is to support the benchmarking of automated tumor segmentation, multi-modal machine learning, and disease differentiation of multi-origin brain tumors in a multi-center setting.

Harmonizing three-dimensional MRI using pseudo-warping field guided GAN

In pursuit of cultivating automated models for magnetic resonance imaging (MRI) to aid in diagnostics, an escalating demand for extensive, multisite, and heterogeneous brain imaging datasets has emerged. This potentially introduces biased outcomes when directly applied for subsequent analysis. Researchers have endeavored to address this issue by pursuing the harmonization of MRIs. However, most existing image-based harmonization methods for MRI are tailored for 2D slices, which may introduce inter-slice variations when they are combined into a 3D volume. In this study, we aim to resolve inconsistencies between slices by introducing a pseudo-warping field. This field is created randomly and utilized to transform a slice into an artificially warped subsequent slice. The objective of this pseudo-warping field is to ensure that generators can consistently harmonize adjacent slices to another domain, without being affected by the varying content present in different slices. Furthermore, we construct unsupervised spatial and recycle loss to enhance the spatial accuracy and slice-wise consistency across the 3D images. The results demonstrate that our model effectively mitigates inter-slice variations and successfully preserves the anatomical details of the images during the harmonization process. Compared to generative harmonization models that employ 3D operators, our model exhibits greater computational efficiency and flexibility.

nBEST: Deep-learning-based non-human primates Brain Extraction and Segmentation Toolbox across ages, sites and species

Accurate processing and analysis of non-human primate (NHP) brain magnetic resonance imaging (MRI) serves an indispensable role in understanding brain evolution, development, aging, and diseases. Despite the accumulation of diverse NHP brain MRI datasets at various developmental stages and from various imaging sites/scanners, existing computational tools designed for human MRI typically perform poor on NHP data, due to huge differences in brain sizes, morphologies, and imaging appearances across species, sites, and ages, highlighting the imperative for NHP-specialized MRI processing tools. To address this issue, in this paper, we present a robust, generic, and fully automated computational pipeline, called non-human primates Brain Extraction and Segmentation Toolbox (nBEST), whose main functionality includes brain extraction, non-cerebrum removal, and tissue segmentation. Building on cutting-edge deep learning techniques by employing lifelong learning to flexibly integrate data from diverse NHP populations and innovatively constructing 3D U-NeXt architecture, nBEST can well handle structural NHP brain MR images from multi-species, multi-site, and multi-developmental-stage (from neonates to the elderly). We extensively validated nBEST based on, to our knowledge, the largest assemblage dataset in NHP brain studies, encompassing 1,469 scans with 11 species (e.g., rhesus macaques, cynomolgus macaques, chimpanzees, marmosets, squirrel monkeys, etc.) from 23 independent datasets. Compared to alternative tools, nBEST outperforms in precision, applicability, robustness, comprehensiveness, and generalizability, greatly benefiting downstream longitudinal, cross-sectional, and cross-species quantitative analyses. We have made nBEST an open-source toolbox (https://github.com/TaoZhong11/nBEST) and we are committed to its continual refinement through lifelong learning with incoming data to greatly contribute to the research field.

Cerebrospinal Fluid cfDNA Sequencing for Classification of Central Nervous System Glioma

PURPOSE

Primary central nervous system (CNS) gliomas can be classified by characteristic genetic alterations. In addition to solid tissue obtained via surgery or biopsy, cell-free DNA (cfDNA) from cerebrospinal fluid (CSF) is an alternative source of material for genomic analyses.

EXPERIMENTAL DESIGN

We performed targeted next-generation sequencing of CSF cfDNA in a representative cohort of 85 patients presenting at two neurooncological centers with suspicion of primary or recurrent glioma. Copy-number variation (CNV) profiles, single-nucleotide variants (SNV), and small insertions/deletions (indel) were combined into a molecular-guided tumor classification. Comparison with the solid tumor was performed for 38 cases with matching solid tissue available.

RESULTS

Cases were stratified into four groups: glioblastoma (n = 32), other glioma (n = 19), nonmalignant (n = 17), and nondiagnostic (n = 17). We introduced a molecular-guided tumor classification, which enabled identification of tumor entities and/or cancer-specific alterations in 75.0% (n = 24) of glioblastoma and 52.6% (n = 10) of other glioma cases. The overlap between CSF and matching solid tissue was highest for CNVs (26%-48%) and SNVs at predefined gene loci (44%), followed by SNVs/indels identified via uninformed variant calling (8%-14%). A molecular-guided tumor classification was possible for 23.5% (n = 4) of nondiagnostic cases.

CONCLUSIONS

We developed a targeted sequencing workflow for CSF cfDNA as well as a strategy for interpretation and reporting of sequencing results based on a molecular-guided tumor classification in glioma. See related commentary by Abdullah, p. 2860.

Unsupervised Anomaly Detection by Learning Elastic Transformations Within an Autoencoder Approach

Machine learning has approached magnetic resonance image (MRI) analysis using multiple techniques, including deep supervised learning methodologies, where lesions such as tumors or features associated with defined pathologies have been identified satisfactorily. However, many of these models require a certain amount of labeled data which access is usually restricted, due to the protection of health information. Furthermore, those methodologies are focused only on the pathologies considered in the training process, causing the model to ignore other kinds of brain lesions. In this paper, an unsupervised anomaly detection methodology is developed to model healthy structures and detect anomaly regions in MRI. Random elastic transform patches are applied over the whole volume to quantify and optimize the reconstruction performance during training, leading to effective anomaly region detection. The experimental results show competitive results with common state-of-the-art unsupervised approaches.

Global and Regional Deep Learning Models for Multiple Sclerosis Stratification From MRI

BACKGROUND

The combination of anatomical MRI and deep learning-based methods such as convolutional neural networks (CNNs) is a promising strategy to build predictive models of multiple sclerosis (MS) prognosis. However, studies assessing the effect of different input strategies on model's performance are lacking.

PURPOSE

To compare whole-brain input sampling strategies and regional/specific-tissue strategies, which focus on a priori known relevant areas for disability accrual, to stratify MS patients based on their disability level.

STUDY TYPE

Retrospective.

SUBJECTS

Three hundred nineteen MS patients (382 brain MRI scans) with clinical assessment of disability level performed within the following 6 months (~70% training/~15% validation/~15% inference in-house dataset) and 440 MS patients from multiple centers (independent external validation cohort).

FIELD STRENGTH/SEQUENCE

Single vendor 1.5 T or 3.0 T. Magnetization-Prepared Rapid Gradient-Echo and Fluid-Attenuated Inversion Recovery sequences.

ASSESSMENT

A 7-fold patient cross validation strategy was used to train a 3D-CNN to classify patients into two groups, Expanded Disability Status Scale score (EDSS) ≥ 3.0 or EDSS < 3.0. Two strategies were investigated: 1) a global approach, taking the whole brain volume as input and 2) regional approaches using five different regions-of-interest: white matter, gray matter, subcortical gray matter, ventricles, and brainstem structures. The performance of the models was assessed in the in-house and the independent external cohorts.

STATISTICAL TESTS

Balanced accuracy, sensitivity, specificity, area under receiver operating characteristic (ROC) curve (AUC).

RESULTS

With the in-house dataset, the gray matter regional model showed the highest stratification accuracy (81%), followed by the global approach (79%). In the external dataset, without any further retraining, an accuracy of 72% was achieved for the white matter model and 71% for the global approach.

DATA CONCLUSION

The global approach offered the best trade-off between internal performance and external validation to stratify MS patients based on accumulated disability.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 2.

Anomaly Detection in Pediatric and Adults Brain MRI with Generative Model

Magnetic Resonance Imaging (MRI) serves as a valuable tool for detecting abnormalities in brain structures. However, a notable 5-10% of pathologies remain unnoticed in MRI scans. To address this challenge and reduce the burden on radiologists, machine learning methods have been used to automate anomaly detection. In this work, we propose to use Denoising Diffusion Probabilistic Model (DDPM) for anomaly detection in MRI. DDPM learns to reconstruct healthy brain MRI and generates anomaly maps to facilitate the classification of the input MRI as healthy or unhealthy. The proposed model correctly differentiated all healthy or unhealthy test patients. Additionally, localized regions of interest in the generated anomaly maps, and the diffusion model achieved a Dice coefficient of 0.376, outperforming other generative models such as generative adversarial network (GAN).

Predicting Cognitive Functioning for Patients with a High-Grade Glioma: Evaluating Different Representations of Tumor Location in a Common Space

Cognitive functioning is increasingly considered when making treatment decisions for patients with a brain tumor in view of a personalized onco-functional balance. Ideally, one can predict cognitive functioning of individual patients to make treatment decisions considering this balance. To make accurate predictions, an informative representation of tumor location is pivotal, yet comparisons of representations are lacking. Therefore, this study compares brain atlases and principal component analysis (PCA) to represent voxel-wise tumor location. Pre-operative cognitive functioning was predicted for 246 patients with a high-grade glioma across eight cognitive tests while using different representations of voxel-wise tumor location as predictors. Voxel-wise tumor location was represented using 13 different frequently-used population average atlases, 13 randomly generated atlases, and 13 representations based on PCA. ElasticNet predictions were compared between representations and against a model solely using tumor volume. Preoperative cognitive functioning could only partly be predicted from tumor location. Performances of different representations were largely similar. Population average atlases did not result in better predictions compared to random atlases. PCA-based representation did not clearly outperform other representations, although summary metrics indicated that PCA-based representations performed somewhat better in our sample. Representations with more regions or components resulted in less accurate predictions. Population average atlases possibly cannot distinguish between functionally distinct areas when applied to patients with a glioma. This stresses the need to develop and validate methods for individual parcellations in the presence of lesions. Future studies may test if the observed small advantage of PCA-based representations generalizes to other data.

Stepwise Transfer Learning for Expert-level Pediatric Brain Tumor MRI Segmentation in a Limited Data Scenario

Purpose To develop, externally test, and evaluate clinical acceptability of a deep learning pediatric brain tumor segmentation model using stepwise transfer learning. Materials and Methods In this retrospective study, the authors leveraged two T2-weighted MRI datasets (May 2001 through December 2015) from a national brain tumor consortium (n = 184; median age, 7 years [range, 1-23 years]; 94 male patients) and a pediatric cancer center (n = 100; median age, 8 years [range, 1-19 years]; 47 male patients) to develop and evaluate deep learning neural networks for pediatric low-grade glioma segmentation using a stepwise transfer learning approach to maximize performance in a limited data scenario. The best model was externally tested on an independent test set and subjected to randomized blinded evaluation by three clinicians, wherein they assessed clinical acceptability of expert- and artificial intelligence (AI)-generated segmentations via 10-point Likert scales and Turing tests. Results The best AI model used in-domain stepwise transfer learning (median Dice score coefficient, 0.88 [IQR, 0.72-0.91] vs 0.812 [IQR, 0.56-0.89] for baseline model; P = .049). With external testing, the AI model yielded excellent accuracy using reference standards from three clinical experts (median Dice similarity coefficients: expert 1, 0.83 [IQR, 0.75-0.90]; expert 2, 0.81 [IQR, 0.70-0.89]; expert 3, 0.81 [IQR, 0.68-0.88]; mean accuracy, 0.82). For clinical benchmarking (n = 100 scans), experts rated AI-based segmentations higher on average compared with other experts (median Likert score, 9 [IQR, 7-9] vs 7 [IQR 7-9]) and rated more AI segmentations as clinically acceptable (80.2% vs 65.4%). Experts correctly predicted the origin of AI segmentations in an average of 26.0% of cases. Conclusion Stepwise transfer learning enabled expert-level automated pediatric brain tumor autosegmentation and volumetric measurement with a high level of clinical acceptability. Keywords: Stepwise Transfer Learning, Pediatric Brain Tumors, MRI Segmentation, Deep Learning Supplemental material is available for this article. © RSNA, 2024.

Two-Stage Training Framework Using Multicontrast MRI Radiomics for IDH Mutation Status Prediction in Glioma

Purpose To develop a radiomics framework for preoperative MRI-based prediction of isocitrate dehydrogenase (IDH) mutation status, a crucial glioma prognostic indicator. Materials and Methods Radiomics features (shape, first-order statistics, and texture) were extracted from the whole tumor or the combination of nonenhancing, necrosis, and edema regions. Segmentation masks were obtained via the federated tumor segmentation tool or the original data source. Boruta, a wrapper-based feature selection algorithm, identified relevant features. Addressing the imbalance between mutated and wild-type cases, multiple prediction models were trained on balanced data subsets using random forest or XGBoost and assembled to build the final classifier. The framework was evaluated using retrospective MRI scans from three public datasets (The Cancer Imaging Archive [TCIA, 227 patients], the University of California San Francisco Preoperative Diffuse Glioma MRI dataset [UCSF, 495 patients], and the Erasmus Glioma Database [EGD, 456 patients]) and internal datasets collected from the University of Texas Southwestern Medical Center (UTSW, 356 patients), New York University (NYU, 136 patients), and University of Wisconsin-Madison (UWM, 174 patients). TCIA and UTSW served as separate training sets, while the remaining data constituted the test set (1617 or 1488 testing cases, respectively). Results The best performing models trained on the TCIA dataset achieved area under the receiver operating characteristic curve (AUC) values of 0.89 for UTSW, 0.86 for NYU, 0.93 for UWM, 0.94 for UCSF, and 0.88 for EGD test sets. The best performing models trained on the UTSW dataset achieved slightly higher AUCs: 0.92 for TCIA, 0.88 for NYU, 0.96 for UWM, 0.93 for UCSF, and 0.90 for EGD. Conclusion This MRI radiomics-based framework shows promise for accurate preoperative prediction of IDH mutation status in patients with glioma. Keywords: Glioma, Isocitrate Dehydrogenase Mutation, IDH Mutation, Radiomics, MRI Supplemental material is available for this article. Published under a CC BY 4.0 license. See also commentary by Moassefi and Erickson in this issue.

Brain functional networks and structures that categorize type 2 bipolar disorder and major depression

BACKGROUND

Distinguishing between type 2 bipolar disorder (BD II) and major depressive disorder (MDD) poses a significant clinical challenge due to their overlapping symptomatology. This study aimed to investigate neurobiological markers that differentiate BD II from MDD using multimodal neuroimaging techniques.

METHODS

Fifty-nine individuals with BD II, 114 with MDD, and 117 healthy controls participated in the study, undergoing structural and functional magnetic resonance imaging. Functional connectivity (FC) analysis used regions from Shen's whole-brain FC-based atlas. Feature selection was carried out using independent t-tests and ReliefF algorithms, followed by classification using Support Vector Machine and wide neural network.

RESULTS

Significant differences in brain structure and function were observed among patients with BD II, MDD, and healthy controls. Both structural and functional alterations were more pronounced in BD II compared to MDD, particularly in regions associated with sensory processing, motor function, and the cerebellum. Classification based on neurobiological markers achieved a mean testing accuracy of 88.24%, with the t-test selected features outperforming those selected by ReliefF. Dysconnectivity patterns correlated with symptom severity and functioning in BD II but not MDD.

CONCLUSION

Our findings suggest that neurobiological markers derived from multimodal imaging techniques can effectively differentiate patients with BD II from those with MDD. The identified alterations in brain structure and function, particularly in sensory-motor processing networks, may serve as potential biomarkers for distinguishing between these mood disorders. However, the influence of psychotropic medications and daily functioning severity on these neurobiological markers warrants further investigation.

Fast refacing of MR images with a generative neural network lowers re-identification risk and preserves volumetric consistency

With the rise of open data, identifiability of individuals based on 3D renderings obtained from routine structural magnetic resonance imaging (MRI) scans of the head has become a growing privacy concern. To protect subject privacy, several algorithms have been developed to de-identify imaging data using blurring, defacing or refacing. Completely removing facial structures provides the best re-identification protection but can significantly impact post-processing steps, like brain morphometry. As an alternative, refacing methods that replace individual facial structures with generic templates have a lower effect on the geometry and intensity distribution of original scans, and are able to provide more consistent post-processing results by the price of higher re-identification risk and computational complexity. In the current study, we propose a novel method for anonymized face generation for defaced 3D T1-weighted scans based on a 3D conditional generative adversarial network. To evaluate the performance of the proposed de-identification tool, a comparative study was conducted between several existing defacing and refacing tools, with two different segmentation algorithms (FAST and Morphobox). The aim was to evaluate (i) impact on brain morphometry reproducibility, (ii) re-identification risk, (iii) balance between (i) and (ii), and (iv) the processing time. The proposed method takes 9 s for face generation and is suitable for recovering consistent post-processing results after defacing.

nnSegNeXt: A 3D Convolutional Network for Brain Tissue Segmentation Based on Quality Evaluation

Accurate and automated segmentation of brain tissue images can significantly streamline clinical diagnosis and analysis. Manual delineation needs improvement due to its laborious and repetitive nature, while automated techniques encounter challenges stemming from disparities in magnetic resonance imaging (MRI) acquisition equipment and accurate labeling. Existing software packages, such as FSL and FreeSurfer, do not fully replace ground truth segmentation, highlighting the need for an efficient segmentation tool. To better capture the essence of cerebral tissue, we introduce nnSegNeXt, an innovative segmentation architecture built upon the foundations of quality assessment. This pioneering framework effectively addresses the challenges posed by missing and inaccurate annotations. To enhance the model's discriminative capacity, we integrate a 3D convolutional attention mechanism instead of conventional convolutional blocks, enabling simultaneous encoding of contextual information through the incorporation of multiscale convolutional features. Our methodology was evaluated on four multi-site T1-weighted MRI datasets from diverse sources, magnetic field strengths, scanning parameters, temporal instances, and neuropsychiatric conditions. Empirical evaluations on the HCP, SALD, and IXI datasets reveal that nnSegNeXt surpasses the esteemed nnUNet, achieving Dice coefficients of 0.992, 0.987, and 0.989, respectively, and demonstrating superior generalizability across four distinct projects with Dice coefficients ranging from 0.967 to 0.983. Additionally, extensive ablation studies have been implemented to corroborate the effectiveness of the proposed model. These findings represent a notable advancement in brain tissue analysis, suggesting that nnSegNeXt holds the promise to significantly refine clinical workflows.

Generative AI in glioma: Ensuring diversity in training image phenotypes to improve diagnostic performance for IDH mutation prediction

BACKGROUND

This study evaluated whether generative artificial intelligence (AI)-based augmentation (GAA) can provide diverse and realistic imaging phenotypes and improve deep learning-based classification of isocitrate dehydrogenase (IDH) type in glioma compared with neuroradiologists.

METHODS

For model development, 565 patients (346 IDH-wildtype, 219 IDH-mutant) with paired contrast-enhanced T1 and FLAIR MRI scans were collected from tertiary hospitals and The Cancer Imaging Archive. Performance was tested on internal (119, 78 IDH-wildtype, 41 IDH-mutant [IDH1 and 2]) and external test sets (108, 72 IDH-wildtype, 36 IDH-mutant). GAA was developed using a score-based diffusion model and ResNet50 classifier. The optimal GAA was selected in comparison with the null model. Two neuroradiologists (R1, R2) assessed realism, diversity of imaging phenotypes, and predicted IDH mutation. The performance of a classifier trained with optimal GAA was compared with that of neuroradiologists using the area under the receiver operating characteristics curve (AUC). The effect of tumor size and contrast enhancement on GAA performance was tested.

RESULTS

Generated images demonstrated realism (Turing's test: 47.5-50.5%) and diversity indicating IDH type. Optimal GAA was achieved with augmentation with 110 000 generated slices (AUC: 0.938). The classifier trained with optimal GAA demonstrated significantly higher AUC values than neuroradiologists in both the internal (R1, P = .003; R2, P < .001) and external test sets (R1, P < .01; R2, P < .001). GAA with large-sized tumors or predominant enhancement showed comparable performance to optimal GAA (internal test: AUC 0.956 and 0.922; external test: 0.810 and 0.749).

CONCLUSIONS

The application of generative AI with realistic and diverse images provided better diagnostic performance than neuroradiologists for predicting IDH type in glioma.

Cluster-based prognostication in glioblastoma: Unveiling heterogeneity based on diffusion and perfusion similarities

BACKGROUND

While the association between diffusion and perfusion magnetic resonance imaging (MRI) and survival in glioblastoma is established, prognostic models for patients are lacking. This study employed clustering of functional imaging to identify distinct functional phenotypes in untreated glioblastomas, assessing their prognostic significance for overall survival.

METHODS

A total of 289 patients with glioblastoma who underwent preoperative multimodal MR imaging were included. Mean values of apparent diffusion coefficient normalized relative cerebral blood volume and relative cerebral blood flow were calculated for different tumor compartments and the entire tumor. Distinct imaging patterns were identified using partition around medoids (PAM) clustering on the training dataset, and their ability to predict overall survival was assessed. Additionally, tree-based machine-learning models were trained to ascertain the significance of features pertaining to cluster membership.

RESULTS

Using the training dataset (231/289) we identified 2 stable imaging phenotypes through PAM clustering with significantly different overall survival (OS). Validation in an independent test set revealed a high-risk group with a median OS of 10.2 months and a low-risk group with a median OS of 26.6 months (P = 0.012). Patients in the low-risk cluster had high diffusion and low perfusion values throughout, while the high-risk cluster displayed the reverse pattern. Including cluster membership in all multivariate Cox regression analyses improved performance (P ≤ 0.004 each).

CONCLUSIONS

Our research demonstrates that data-driven clustering can identify clinically relevant, distinct imaging phenotypes, highlighting the potential role of diffusion, and perfusion MRI in predicting survival rates of glioblastoma patients.

Assessing the Accuracy and Reproducibility of PARIETAL: A Deep Learning Brain Extraction Algorithm

BACKGROUND

Manual brain extraction from magnetic resonance (MR) images is time-consuming and prone to intra- and inter-rater variability. Several automated approaches have been developed to alleviate these constraints, including deep learning pipelines. However, these methods tend to reduce their performance in unseen magnetic resonance imaging (MRI) scanner vendors and different imaging protocols.

PURPOSE

To present and evaluate for clinical use PARIETAL, a pre-trained deep learning brain extraction method. We compare its reproducibility in a scan/rescan analysis and its robustness among scanners of different manufacturers.

STUDY TYPE

Retrospective.

POPULATION

Twenty-one subjects (12 women) with age range 22-48 years acquired using three different MRI scanner machines including scan/rescan in each of them.

FIELD STRENGTH/SEQUENCE

T1-weighted images acquired in a 3-T Siemens with magnetization prepared rapid gradient-echo sequence and two 1.5 T scanners, Philips and GE, with spin-echo and spoiled gradient-recalled (SPGR) sequences, respectively.

ASSESSMENT

Analysis of the intracranial cavity volumes obtained for each subject on the three different scanners and the scan/rescan acquisitions.

STATISTICAL TESTS

Parametric permutation tests of the differences in volumes to rank and statistically evaluate the performance of PARIETAL compared to state-of-the-art methods.

RESULTS

The mean absolute intracranial volume differences obtained by PARIETAL in the scan/rescan analysis were 1.88 mL, 3.91 mL, and 4.71 mL for Siemens, GE, and Philips scanners, respectively. PARIETAL was the best-ranked method on Siemens and GE scanners, while decreasing to Rank 2 on the Philips images. Intracranial differences for the same subject between scanners were 5.46 mL, 27.16 mL, and 30.44 mL for GE/Philips, Siemens/Philips, and Siemens/GE comparison, respectively. The permutation tests revealed that PARIETAL was always in Rank 1, obtaining the most similar volumetric results between scanners.

DATA CONCLUSION

PARIETAL accurately segments the brain and it generalizes to images acquired at different sites without the need of training or fine-tuning it again. PARIETAL is publicly available.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY STAGE: 2.