Bias in machine learning models can be significantly mitigated by careful training: Evidence from neuroimaging studies.

Despite the great promise that machine learning has offered in many fields of medicine, it has also raised concerns about potential biases and poor generalization across genders, age distributions, races and ethnicities, hospitals, and data acquisition equipment and protocols. In the current study, and in the context of three brain diseases, we provide evidence which suggests that when properly trained, machine learning models can generalize well across diverse conditions and do not necessarily suffer from bias. Specifically, by using multistudy magnetic resonance imaging consortia for diagnosing Alzheimer's disease, schizophrenia, and autism spectrum disorder, we find that well-trained models have a high area-under-the-curve (AUC) on subjects across different subgroups pertaining to attributes such as gender, age, racial groups and different clinical studies and are unbiased under multiple fairness metrics such as demographic parity difference, equalized odds difference, equal opportunity difference, etc. We find that models that incorporate multisource data from demographic, clinical, genetic factors, and cognitive scores are also unbiased. These models have a better predictive AUC across subgroups than those trained only with imaging features, but there are also situations when these additional features do not help.

ASLPrep: a platform for processing of arterial spin labeled MRI and quantification of regional brain perfusion.

Arterial spin labeled (ASL) magnetic resonance imaging (MRI) is the primary method for noninvasively measuring regional brain perfusion in humans. We introduce ASLPrep, a suite of software pipelines that ensure the reproducible and generalizable processing of ASL MRI data.

Proteomic Indicators of Health Predict Alzheimer's Disease Biomarker Levels and Dementia Risk.

OBJECTIVE: Few studies have comprehensively examined how health and disease risk influence Alzheimer's disease (AD) biomarkers. The present study examined the association of 14 protein-based health indicators with plasma and neuroimaging biomarkers of AD and neurodegeneration. METHODS: In 706 cognitively normal adults, we examined whether 14 protein-based health indices (ie, SomaSignal® tests) were associated with concurrently measured plasma-based biomarkers of AD pathology (amyloid-ß [Aß]42/40 , tau phosphorylated at threonine-181 [pTau-181]), neuronal injury (neurofilament light chain [NfL]), and reactive astrogliosis (glial fibrillary acidic protein [GFAP]), brain volume, and cortical Aß and tau. In a separate cohort (n = 11,285), we examined whether protein-based health indicators associated with neurodegeneration also predict 25-year dementia risk. RESULTS: Greater protein-based risk for cardiovascular disease, heart failure mortality, and kidney disease was associated with lower Aß42/40 and higher pTau-181, NfL, and GFAP levels, even in individuals without cardiovascular or kidney disease. Proteomic indicators of body fat percentage, lean body mass, and visceral fat were associated with pTau-181, NfL, and GFAP, whereas resting energy rate was negatively associated with NfL and GFAP. Together, these health indicators predicted 12, 31, 50, and 33% of plasma Aß42/40 , pTau-181, NfL, and GFAP levels, respectively. Only protein-based measures of cardiovascular risk were associated with reduced regional brain volumes; these measures predicted 25-year dementia risk, even among those without clinically defined cardiovascular disease. INTERPRETATION: Subclinical peripheral health may influence AD and neurodegenerative disease processes and relevant biomarker levels, particularly NfL. Cardiovascular health, even in the absence of clinically defined disease, plays a central role in brain aging and dementia. ANN NEUROL 2024;95:260-273.

Sex differences in the functional topography of association networks in youth.

Prior work has shown that there is substantial interindividual variation in the spatial distribution of functional networks across the cerebral cortex, or functional topography. However, it remains unknown whether there are sex differences in the topography of individualized networks in youth. Here, we leveraged an advanced machine learning method (sparsity-regularized non-negative matrix factorization) to define individualized functional networks in 693 youth (ages 8 to 23 y) who underwent functional MRI as part of the Philadelphia Neurodevelopmental Cohort. Multivariate pattern analysis using support vector machines classified participant sex based on functional topography with 82.9% accuracy (P < 0.0001). Brain regions most effective in classifying participant sex belonged to association networks, including the ventral attention, default mode, and frontoparietal networks. Mass univariate analyses using generalized additive models with penalized splines provided convergent results. Furthermore, transcriptomic data from the Allen Human Brain Atlas revealed that sex differences in multivariate patterns of functional topography were spatially correlated with the expression of genes on the X chromosome. These results highlight the role of sex as a biological variable in shaping functional topography.

Penalized decomposition using residuals (PeDecURe) for feature extraction in the presence of nuisance variables.

Neuroimaging data are an increasingly important part of etiological studies of neurological and psychiatric disorders. However, mitigating the influence of nuisance variables, including confounders, remains a challenge in image analysis. In studies of Alzheimer's disease, for example, an imbalance in disease rates by age and sex may make it difficult to distinguish between structural patterns in the brain (as measured by neuroimaging scans) attributable to disease progression and those characteristic of typical human aging or sex differences. Concerningly, when not properly accounted for, nuisance variables pose threats to the generalizability and interpretability of findings from these studies. Motivated by this critical issue, in this work, we examine the impact of nuisance variables on feature extraction methods and propose Penalized Decomposition Using Residuals (PeDecURe), a new method for obtaining nuisance variable-adjusted features. PeDecURe estimates primary directions of variation which maximize covariance between partially residualized imaging features and a variable of interest (e.g., Alzheimer's diagnosis) while simultaneously mitigating the influence of nuisance variation through a penalty on the covariance between partially residualized imaging features and those variables. Using features derived using PeDecURe's first direction of variation, we train a highly accurate and generalizable predictive model, as evidenced by its robustness in testing samples with different underlying nuisance variable distributions. We compare PeDecURe to commonly used decomposition methods (principal component analysis (PCA) and partial least squares) as well as a confounder-adjusted variation of PCA. We find that features derived from PeDecURe offer greater accuracy and generalizability and lower correlations with nuisance variables compared with the other methods. While PeDecURe is primarily motivated by challenges that arise in the analysis of neuroimaging data, it is broadly applicable to data sets with highly correlated features, where novel methods to handle nuisance variables are warranted.

QSIPrep: an integrative platform for preprocessing and reconstructing diffusion MRI data.

Diffusion-weighted magnetic resonance imaging (dMRI) is the primary method for noninvasively studying the organization of white matter in the human brain. Here we introduce QSIPrep, an integrative software platform for the processing of diffusion images that is compatible with nearly all dMRI sampling schemes. Drawing on a diverse set of software suites to capitalize on their complementary strengths, QSIPrep facilitates the implementation of best practices for processing of diffusion images.

Alzheimer's Disease Genetic Risk, Cognition, and Brain Aging in Midlife.

We examined associations of an Alzheimer's disease (AD) Genetic Risk Score (AD-GRS) and midlife cognitive and neuroimaging outcomes in 1,252 middle-aged participants (311 with brain MRI). A higher AD-GRS based on 25 previously identified loci (excluding apolipoprotein E [APOE]) was associated with worse Montreal Cognitive Assessment (-0.14 standard deviation [SD] [95% confidence interval {CI}: -0.26, -0.02]), older machine learning predicted brain age (2.35 years[95%CI: 0.01, 4.69]), and white matter hyperintensity volume (0.35 SD [95% CI: 0.00, 0.71]), but not with a composite cognitive outcome, total brain, or hippocampal volume. APOE ε4 allele was not associated with any outcomes. AD risk genes beyond APOE may contribute to subclinical differences in cognition and brain health in midlife. ANN NEUROL 2023;93:629-634.

Psychosis brain subtypes validated in first-episode cohorts and related to illness remission: results from the PHENOM consortium.

Using machine learning, we recently decomposed the neuroanatomical heterogeneity of established schizophrenia to discover two volumetric subgroups-a 'lower brain volume' subgroup (SG1) and an 'higher striatal volume' subgroup (SG2) with otherwise normal brain structure. In this study, we investigated whether the MRI signatures of these subgroups were also already present at the time of the first-episode of psychosis (FEP) and whether they were related to clinical presentation and clinical remission over 1-, 3-, and 5-years. We included 572 FEP and 424 healthy controls (HC) from 4 sites (Sao Paulo, Santander, London, Melbourne) of the PHENOM consortium. Our prior MRI subgrouping models (671 participants; USA, Germany, and China) were applied to both FEP and HC. Participants were assigned into 1 of 4 categories: subgroup 1 (SG1), subgroup 2 (SG2), no subgroup membership ('None'), and mixed SG1 + SG2 subgroups ('Mixed'). Voxel-wise analyses characterized SG1 and SG2 subgroups. Supervised machine learning analyses characterized baseline and remission signatures related to SG1 and SG2 membership. The two dominant patterns of 'lower brain volume' in SG1 and 'higher striatal volume' (with otherwise normal neuromorphology) in SG2 were identified already at the first episode of psychosis. SG1 had a significantly higher proportion of FEP (32%) vs. HC (19%) than SG2 (FEP, 21%; HC, 23%). Clinical multivariate signatures separated the SG1 and SG2 subgroups (balanced accuracy = 64%; p < 0.0001), with SG2 showing higher education but also greater positive psychosis symptoms at first presentation, and an association with symptom remission at 1-year, 5-year, and when timepoints were combined. Neuromorphological subtypes of schizophrenia are already evident at illness onset, separated by distinct clinical presentations, and differentially associated with subsequent remission. These results suggest that the subgroups may be underlying risk phenotypes that could be targeted in future treatment trials and are critical to consider when interpreting neuroimaging literature.

Preference matrix guided sparse canonical correlation analysis for mining brain imaging genetic associations in Alzheimer's disease.

Investigating the relationship between genetic variation and phenotypic traits is a key issue in quantitative genetics. Specifically for Alzheimer's disease, the association between genetic markers and quantitative traits remains vague while, once identified, will provide valuable guidance for the study and development of genetics-based treatment approaches. Currently, to analyze the association of two modalities, sparse canonical correlation analysis (SCCA) is commonly used to compute one sparse linear combination of the variable features for each modality, giving a pair of linear combination vectors in total that maximizes the cross-correlation between the analyzed modalities. One drawback of the plain SCCA model is that the existing findings and knowledge cannot be integrated into the model as priors to help extract interesting correlations as well as identify biologically meaningful genetic and phenotypic markers. To bridge this gap, we introduce preference matrix guided SCCA (PM-SCCA) that not only takes priors encoded as a preference matrix but also maintains computational simplicity. A simulation study and a real-data experiment are conducted to investigate the effectiveness of the model. Both experiments demonstrate that the proposed PM-SCCA model can capture not only genotype-phenotype correlation but also relevant features effectively.

Proteomic analyses reveal plasma EFEMP1 and CXCL12 as biomarkers and determinants of neurodegeneration.

INTRODUCTION: Plasma proteomic analyses of unique brain atrophy patterns may illuminate peripheral drivers of neurodegeneration and identify novel biomarkers for predicting clinically relevant outcomes. METHODS: We identified proteomic signatures associated with machine learning-derived aging- and Alzheimer's disease (AD) -related brain atrophy patterns in the Baltimore Longitudinal Study of Aging (n = 815). Using data from five cohorts, we examined whether candidate proteins were associated with AD endophenotypes and long-term dementia risk. RESULTS: Plasma proteins associated with distinct patterns of age- and AD-related atrophy were also associated with plasma/cerebrospinal fluid (CSF) AD biomarkers, cognition, AD risk, as well as mid-life (20-year) and late-life (8-year) dementia risk. EFEMP1 and CXCL12 showed the most consistent associations across cohorts and were mechanistically implicated as determinants of brain structure using genetic methods, including Mendelian randomization. DISCUSSION: Our findings reveal plasma proteomic signatures of unique aging- and AD-related brain atrophy patterns and implicate EFEMP1 and CXCL12 as important molecular drivers of neurodegeneration. HIGHLIGHTS: Plasma proteomic signatures are associated with unique patterns of brain atrophy. Brain atrophy-related proteins predict clinically relevant outcomes across cohorts. Genetic variation underlying plasma EFEMP1 and CXCL12 influences brain structure. EFEMP1 and CXCL12 may be important molecular drivers of neurodegeneration.

Tau-neurodegeneration mismatch reveals vulnerability and resilience to comorbidities in Alzheimer's continuum.

INTRODUCTION: Variability in relationship of tau-based neurofibrillary tangles (T) and neurodegeneration (N) in Alzheimer's disease (AD) arises from non-specific nature of N, modulated by non-AD co-pathologies, age-related changes, and resilience factors. METHODS: We used regional T-N residual patterns to partition 184 patients within the Alzheimer's continuum into data-driven groups. These were compared with groups from 159 non-AD (amyloid "negative") patients partitioned using cortical thickness, and groups in 98 patients with ante mortem MRI and post mortem tissue for measuring N and T, respectively. We applied the initial T-N residual model to classify 71 patients in an independent cohort into predefined groups. RESULTS: AD groups displayed spatial T-N mismatch patterns resembling neurodegeneration patterns in non-AD groups, similarly associated with non-AD factors and diverging cognitive outcomes. In the autopsy cohort, limbic T-N mismatch correlated with TDP-43 co-pathology. DISCUSSION: T-N mismatch may provide a personalized approach for determining non-AD factors associated with resilience/vulnerability in AD.

Identification of Clinically Relevant Brain Endothelial Cell Biomarkers in Plasma.

BACKGROUND: Proteins expressed by brain endothelial cells (BECs), the primary cell type of the blood-brain barrier, may serve as sensitive plasma biomarkers for neurological and neurovascular conditions, including cerebral small vessel disease. METHODS: Using data from the BLSA (Baltimore Longitudinal Study of Aging; n=886; 2009-2020), BEC-enriched proteins were identified among 7268 plasma proteins (measured with SomaScanv4.1) using an automated annotation algorithm that filtered endothelial cell transcripts followed by cross-referencing with BEC-specific transcripts reported in single-cell RNA-sequencing studies. To identify BEC-enriched proteins in plasma most relevant to the maintenance of neurological and neurovascular health, we selected proteins significantly associated with 3T magnetic resonance imaging-defined white matter lesion volumes. We then examined how these candidate BEC biomarkers related to white matter lesion volumes, cerebral microhemorrhages, and lacunar infarcts in the ARIC study (Atherosclerosis Risk in Communities; US multisite; 1990-2017). Finally, we determined whether these candidate BEC biomarkers, when measured during midlife, were related to dementia risk over a 25-year follow-up period. RESULTS: Of the 28 proteins identified as BEC-enriched, 4 were significantly associated with white matter lesion volumes (CDH5 [cadherin 5], CD93 [cluster of differentiation 93], ICAM2 [intracellular adhesion molecule 2], GP1BB [glycoprotein 1b platelet subunit beta]), while another approached significance (RSPO3 [R-Spondin 3]). A composite score based on 3 of these BEC proteins accounted for 11% of variation in white matter lesion volumes in BLSA participants. We replicated the associations between the BEC composite score, CDH5, and RSPO3 with white matter lesion volumes in ARIC, and further demonstrated that the BEC composite score and RSPO3 were associated with the presence of ≥1 cerebral microhemorrhages. We also showed that the BEC composite score, CDH5, and RSPO3 were associated with 25-year dementia risk. CONCLUSIONS: In addition to identifying BEC proteins in plasma that relate to cerebral small vessel disease and dementia risk, we developed a composite score of plasma BEC proteins that may be used to estimate blood-brain barrier integrity and risk for adverse neurovascular outcomes.

Associations of baseline and longitudinal change in cerebellum volume with age-related changes in verbal learning and memory.

The cerebellum is involved in higher-order cognitive functions, e.g., learning and memory, and is susceptible to age-related atrophy. Yet, the cerebellum's role in age-related cognitive decline remains largely unknown. We investigated cross-sectional and longitudinal associations between cerebellar volume and verbal learning and memory. Linear mixed effects models and partial correlations were used to examine the relationship between changes in cerebellum volumes (total cerebellum, cerebellum white matter [WM], cerebellum hemisphere gray matter [GM], and cerebellum vermis subregions) and changes in verbal learning and memory performance among 549 Baltimore Longitudinal Study of Aging participants (2,292 visits). All models were adjusted by baseline demographic characteristics (age, sex, race, education), and APOE e4 carrier status. In examining associations between change with change, we tested an additional model that included either hippocampal (HC), cuneus, or postcentral gyrus (PoCG) volumes to assess whether cerebellar volumes were uniquely associated with verbal learning and memory. Cross-sectionally, the association of baseline cerebellum GM and WM with baseline verbal learning and memory was age-dependent, with the oldest individuals showing the strongest association between volume and performance. Baseline volume was not significantly associated with change in learning and memory. However, analysis of associations between change in volumes and changes in verbal learning and memory showed that greater declines in verbal memory were associated with greater volume loss in cerebellum white matter, and preserved GM volume in cerebellum vermis lobules VI-VII. The association between decline in verbal memory and decline in cerebellar WM volume remained after adjustment for HC, cuneus, and PoCG volume. Our findings highlight that associations between cerebellum volume and verbal learning and memory are age-dependent and regionally specific.

Brain-wide genome-wide colocalization study for integrating genetics, transcriptomics and brain morphometry in Alzheimer's disease.

Alzheimer's disease (AD) is one of the most common neurodegenerative diseases. However, the AD mechanism has not yet been fully elucidated to date, hindering the development of effective therapies. In our work, we perform a brain imaging genomics study to link genetics, single-cell gene expression data, tissue-specific gene expression data, brain imaging-derived volumetric endophenotypes, and disease diagnosis to discover potential underlying neurobiological pathways for AD. To do so, we perform brain-wide genome-wide colocalization analyses to integrate multidimensional imaging genomic biobank data. Specifically, we use (1) the individual-level imputed genotyping data and magnetic resonance imaging (MRI) data from the UK Biobank, (2) the summary statistics of the genome-wide association study (GWAS) from multiple European ancestry cohorts, and (3) the tissue-specific cis-expression quantitative trait loci (cis-eQTL) summary statistics from the GTEx project. We apply a Bayes factor colocalization framework and mediation analysis to these multi-modal imaging genomic data. As a result, we derive the brain regional level GWAS summary statistics for 145 brain regions with 482,831 single nucleotide polymorphisms (SNPs) followed by posthoc functional annotations. Our analysis yields the discovery of a potential AD causal pathway from a systems biology perspective: the SNP chr10:124165615:G>A (rs6585827) mutation upregulates the expression of BTBD16 gene in oligodendrocytes, a specialized glial cells, in the brain cortex, leading to a reduced risk of volumetric loss in the entorhinal cortex, resulting in the protective effect on AD. We substantiate our findings with multiple evidence from existing imaging, genetic and genomic studies in AD literature. Our study connects genetics, molecular and cellular signatures, regional brain morphologic endophenotypes, and AD diagnosis, providing new insights into the mechanistic understanding of the disease. Our findings can provide valuable guidance for subsequent therapeutic target identification and drug discovery in AD.

Applications of generative adversarial networks in neuroimaging and clinical neuroscience.

Generative adversarial networks (GANs) are one powerful type of deep learning models that have been successfully utilized in numerous fields. They belong to the broader family of generative methods, which learn to generate realistic data with a probabilistic model by learning distributions from real samples. In the clinical context, GANs have shown enhanced capabilities in capturing spatially complex, nonlinear, and potentially subtle disease effects compared to traditional generative methods. This review critically appraises the existing literature on the applications of GANs in imaging studies of various neurological conditions, including Alzheimer's disease, brain tumors, brain aging, and multiple sclerosis. We provide an intuitive explanation of various GAN methods for each application and further discuss the main challenges, open questions, and promising future directions of leveraging GANs in neuroimaging. We aim to bridge the gap between advanced deep learning methods and neurology research by highlighting how GANs can be leveraged to support clinical decision making and contribute to a better understanding of the structural and functional patterns of brain diseases.

Image harmonization: A review of statistical and deep learning methods for removing batch effects and evaluation metrics for effective harmonization.

Magnetic resonance imaging and computed tomography from multiple batches (e.g. sites, scanners, datasets, etc.) are increasingly used alongside complex downstream analyses to obtain new insights into the human brain. However, significant confounding due to batch-related technical variation, called batch effects, is present in this data; direct application of downstream analyses to the data may lead to biased results. Image harmonization methods seek to remove these batch effects and enable increased generalizability and reproducibility of downstream results. In this review, we describe and categorize current approaches in statistical and deep learning harmonization methods. We also describe current evaluation metrics used to assess harmonization methods and provide a standardized framework to evaluate newly-proposed methods for effective harmonization and preservation of biological information. Finally, we provide recommendations to end-users to advocate for more effective use of current methods and to methodologists to direct future efforts and accelerate development of the field.

Multiscale functional connectivity patterns of the aging brain learned from harmonized rsfMRI data of the multi-cohort iSTAGING study.

To learn multiscale functional connectivity patterns of the aging brain, we built a brain age prediction model of functional connectivity measures at seven scales on a large fMRI dataset, consisting of resting-state fMRI scans of 4186 individuals with a wide age range (22 to 97 years, with an average of 63) from five cohorts. We computed multiscale functional connectivity measures of individual subjects using a personalized functional network computational method, harmonized the functional connectivity measures of subjects from multiple datasets in order to build a functional brain age model, and finally evaluated how functional brain age gap correlated with cognitive measures of individual subjects. Our study has revealed that functional connectivity measures at multiple scales were more informative than those at any single scale for the brain age prediction, the data harmonization significantly improved the brain age prediction performance, and the data harmonization in the functional connectivity measures' tangent space worked better than in their original space. Moreover, brain age gap scores of individual subjects derived from the brain age prediction model were significantly correlated with clinical and cognitive measures. Overall, these results demonstrated that multiscale functional connectivity patterns learned from a large-scale multi-site rsfMRI dataset were informative for characterizing the aging brain and the derived brain age gap was associated with cognitive and clinical measures.

Benchmarking the generalizability of brain age models: Challenges posed by scanner variance and prediction bias.

Machine learning has been increasingly applied to neuroimaging data to predict age, deriving a personalized biomarker with potential clinical applications. The scientific and clinical value of these models depends on their applicability to independently acquired scans from diverse sources. Accordingly, we evaluated the generalizability of two brain age models that were trained across the lifespan by applying them to three distinct early-life samples with participants aged 8-22 years. These models were chosen based on the size and diversity of their training data, but they also differed greatly in their processing methods and predictive algorithms. Specifically, one brain age model was built by applying gradient tree boosting (GTB) to extracted features of cortical thickness, surface area, and brain volume. The other model applied a 2D convolutional neural network (DBN) to minimally preprocessed slices of T1-weighted scans. Additional model variants were created to understand how generalizability changed when each model was trained with data that became more similar to the test samples in terms of age and acquisition protocols. Our results illustrated numerous trade-offs. The GTB predictions were relatively more accurate overall and yielded more reliable predictions when applied to lower quality scans. In contrast, the DBN displayed the most utility in detecting associations between brain age gaps and cognitive functioning. Broadly speaking, the largest limitations affecting generalizability were acquisition protocol differences and biased brain age estimates. If such confounds could eventually be removed without post-hoc corrections, brain age predictions may have greater utility as personalized biomarkers of healthy aging.

Automated analysis of low-field brain MRI in cerebral malaria.

A central challenge of medical imaging studies is to extract biomarkers that characterize disease pathology or outcomes. Modern automated approaches have found tremendous success in high-resolution, high-quality magnetic resonance images. These methods, however, may not translate to low-resolution images acquired on magnetic resonance imaging (MRI) scanners with lower magnetic field strength. In low-resource settings where low-field scanners are more common and there is a shortage of radiologists to manually interpret MRI scans, it is critical to develop automated methods that can augment or replace manual interpretation, while accommodating reduced image quality. We present a fully automated framework for translating radiological diagnostic criteria into image-based biomarkers, inspired by a project in which children with cerebral malaria (CM) were imaged using low-field 0.35 Tesla MRI. We integrate multiatlas label fusion, which leverages high-resolution images from another sample as prior spatial information, with parametric Gaussian hidden Markov models based on image intensities, to create a robust method for determining ventricular cerebrospinal fluid volume. We also propose normalized image intensity and texture measurements to determine the loss of gray-to-white matter tissue differentiation and sulcal effacement. These integrated biomarkers have excellent classification performance for determining severe brain swelling due to CM.

Association of partial T2-FLAIR mismatch sign and isocitrate dehydrogenase mutation in WHO grade 4 gliomas: results from the ReSPOND consortium.

PURPOSE: While the T2-FLAIR mismatch sign is highly specific for isocitrate dehydrogenase (IDH)-mutant, 1p/19q-noncodeleted astrocytomas among lower-grade gliomas, its utility in WHO grade 4 gliomas is not well-studied. We derived the partial T2-FLAIR mismatch sign as an imaging biomarker for IDH mutation in WHO grade 4 gliomas. METHODS: Preoperative MRI scans of adult WHO grade 4 glioma patients (n = 2165) from the multi-institutional ReSPOND (Radiomics Signatures for PrecisiON Diagnostics) consortium were analyzed. Diagnostic performance of the partial T2-FLAIR mismatch sign was evaluated. Subset analyses were performed to assess associations of imaging markers with overall survival (OS). RESULTS: One hundred twenty-one (5.6%) of 2165 grade 4 gliomas were IDH-mutant. Partial T2-FLAIR mismatch was present in 40 (1.8%) cases, 32 of which were IDH-mutant, yielding 26.4% sensitivity, 99.6% specificity, 80.0% positive predictive value, and 95.8% negative predictive value. Multivariate logistic regression demonstrated IDH mutation was significantly associated with partial T2-FLAIR mismatch (odds ratio [OR] 5.715, 95% CI [1.896, 17.221], p = 0.002), younger age (OR 0.911 [0.895, 0.927], p < 0.001), tumor centered in frontal lobe (OR 3.842, [2.361, 6.251], p < 0.001), absence of multicentricity (OR 0.173, [0.049, 0.612], p = 0.007), and presence of cystic (OR 6.596, [3.023, 14.391], p < 0.001) or non-enhancing solid components (OR 6.069, [3.371, 10.928], p < 0.001). Multivariate Cox analysis demonstrated cystic components (p = 0.024) and non-enhancing solid components (p = 0.003) were associated with longer OS, while older age (p < 0.001), frontal lobe center (p = 0.008), multifocality (p < 0.001), and multicentricity (p < 0.001) were associated with shorter OS. CONCLUSION: Partial T2-FLAIR mismatch sign is highly specific for IDH mutation in WHO grade 4 gliomas.