Advancing cerebral small vessel disease diagnosis: Integrating quantitative susceptibility mapping with MRI-based radiomics.

Cerebral small vessel disease (CSVD) is a neurodegenerative disease with hidden symptoms and difficult to diagnose. The diagnosis mainly depends on clinical symptoms and neuroimaging. Therefore, we explored the potential of combining clinical detection with MRI-based radiomics features for the diagnosis of CSVD in a large cohort. A total of 118 CSVD patients and 127 healthy controls underwent quantitative susceptibility mapping and 3D-T1 scans, and all completed multiple cognitive tests. Lasso regression was used to select features, and the radiomics model was constructed based on the regression coefficients of these features. Clinical cognitive and motor tests were added to the model to construct a hybrid model. All models were cross-validated to analyze the generalization ability of the models. The AUCs of the radiomics and hybrid models in the internal test set were 0.80 and 0.87, respectively. In the validation set, the AUCs were 0.77 and 0.79, respectively. The hybrid model demonstrated higher decision efficiency. The Trail Making Test, which enhances the diagnostic performance of the model, is associated with multiple brain regions, particularly the right cortical nuclei and the right fimbria. The hybrid model based on radiomics features and cognitive tests can achieve quantitative diagnosis of CSVD and improve the diagnostic efficiency. Furthermore, the reduced processing capacity due to atrophy of the right cortical nucleus and right fimbria suggests the importance of these regions in improving the diagnostic accuracy of the model.

Neuro-XAI: Explainable deep learning framework based on deeplabV3+ and bayesian optimization for segmentation and classification of brain tumor in MRI scans.

The prevalence of brain tumor disorders is currently a global issue. In general, radiography, which includes a large number of images, is an efficient method for diagnosing these life-threatening disorders. The biggest issue in this area is that it takes a radiologist a long time and is physically strenuous to look at all the images. As a result, research into developing systems based on machine learning to assist radiologists in diagnosis continues to rise daily. Convolutional neural networks (CNNs), one type of deep learning approach, have been pivotal in achieving state-of-the-art results in several medical imaging applications, including the identification of brain tumors. CNN hyperparameters are typically set manually for segmentation and classification, which might take a while and increase the chance of using suboptimal hyperparameters for both tasks. Bayesian optimization is a useful method for updating the deep CNN's optimal hyperparameters. The CNN network, however, can be considered a "black box" model because of how difficult it is to comprehend the information it stores because of its complexity. Therefore, this problem can be solved by using Explainable Artificial Intelligence (XAI) tools, which provide doctors with a realistic explanation of CNN's assessments. Implementation of deep learning-based systems in real-time diagnosis is still rare. One of the causes could be that these methods don't quantify the Uncertainty in the predictions, which could undermine trust in the AI-based diagnosis of diseases. To be used in real-time medical diagnosis, CNN-based models must be realistic and appealing, and uncertainty needs to be evaluated. So, a novel three-phase strategy is proposed for segmenting and classifying brain tumors. Segmentation of brain tumors using the DeeplabV3+ model is first performed with tuning of hyperparameters using Bayesian optimization. For classification, features from state-of-the-art deep learning models Darknet53 and mobilenetv2 are extracted and fed to SVM for classification, and hyperparameters of SVM are also optimized using a Bayesian approach. The second step is to understand whatever portion of the images CNN uses for feature extraction using XAI algorithms. Using confusion entropy, the Uncertainty of the Bayesian optimized classifier is finally quantified. Based on a Bayesian-optimized deep learning framework, the experimental findings demonstrate that the proposed method outperforms earlier techniques, achieving a 97 % classification accuracy and a 0.98 global accuracy.

Segmentation of stroke lesions using transformers-augmented MRI analysis.

Accurate segmentation of chronic stroke lesions from mono-spectral magnetic resonance imaging scans (e.g., T1-weighted images) is a difficult task due to the arbitrary shape, complex texture, variable size and intensities, and varied locations of the lesions. Due to this inherent spatial heterogeneity, existing machine learning methods have shown moderate performance for chronic lesion delineation. In this study, we introduced: (1) a method that integrates transformers' deformable feature attention mechanism with convolutional deep learning architecture to improve the accuracy and generalizability of stroke lesion segmentation, and (2) an ecological data augmentation technique based on inserting real lesions into intact brain regions. Our combination of these two approaches resulted in a significant increase in segmentation performance, with a Dice index of 0.82 (±0.39), outperforming the existing methods trained and tested on the same Anatomical Tracings of Lesions After Stroke (ATLAS) 2022 dataset. Our method performed relatively well even for cases with small stroke lesions. We validated the robustness of our method through an ablation study and by testing it on new unseen brain scans from the Ischemic Stroke Lesion Segmentation (ISLES) 2015 dataset. Overall, our proposed approach of transformers with ecological data augmentation offers a robust way to delineate chronic stroke lesions with clinically relevant accuracy. Our method can be extended to other challenging tasks that require automated detection and segmentation of diverse brain abnormalities from clinical scans.

Development and validation of a perivascular space segmentation method in multi-center datasets.

BACKGROUND: Perivascular spaces (PVS) visible on magnetic resonance imaging (MRI) are significant markers associated with various neurological diseases. Although quantitative analysis of PVS may enhance sensitivity and improve consistency across studies, the field lacks a universally validated method for analyzing images from multi-center studies. METHODS: We annotated PVS on multi-center 3D T1-weighted (T1w) images acquired using scanners from three major vendors (Siemens, General Electric, and Philips). A neural network, mcPVS-Net (multi-center PVS segmentation network), was trained using data from 40 subjects and then tested in a separate cohort of 15 subjects. We assessed segmentation accuracy against ground truth masks tailored for each scanner vendor. Additionally, we evaluated the agreement between segmented PVS volumes and visual scores for each scanner. We also explored correlations between PVS volumes and various clinical factors such as age, hypertension, and white matter hyperintensities (WMH) in a larger sample of 1020 subjects. Furthermore, mcPVS-Net was applied to a new dataset comprising both T1w and T2-weighted (T2w) images from a United Imaging scanner to investigate if PVS volumes could discriminate between subjects with differing visual scores. We also compared the mcPVS-Net with a previously published method that segments PVS from T1 images. RESULTS: In the test dataset, mcPVS-Net achieved a mean DICE coefficient of 0.80, with an average Precision of 0.81 and Recall of 0.79, indicating good specificity and sensitivity. The segmented PVS volumes were significantly associated with visual scores in both the basal ganglia (r = 0.541, p < 0.001) and white matter regions (r = 0.706, p < 0.001), and PVS volumes were significantly different among subjects with varying visual scores. Segmentation performance was consistent across different scanner vendors. PVS volumes exhibited significant associations with age, hypertension, and WMH. In the United Imaging scanner dataset, PVS volumes showed good associations with PVS visual scores evaluated on either T1w or T2w images. Compared to a previously published method, mcPVS-Net showed a higher accuracy and improved PVS segmentation in the basal ganglia region. CONCLUSION: The mcPVS-Net demonstrated good accuracy for segmenting PVS from 3D T1w images. It may serve as a useful tool for future PVS research.

Age-appropriate or delayed myelination? Scoring myelination in routine clinical MRI.

BACKGROUND: Assessment of myelination is a core issue in paediatric neuroimaging and can be challenging, particularly in settings without dedicated paediatric neuroradiologists. Deep learning models have recently been shown to be able to estimate myelination age in children with normal MRI, but currently lack validation for patients with myelination delay and implementation including pre-processing suitable for local imaging is not trivial. Standardized myelination scores, which have been successfully used as biomarkers for myelination in hypomyelinating diseases, rely on visual, semiquantitative scoring of myelination on routine clinical MRI and may offer an easy-to-use alternative for assessment of myelination. METHODS: Myelination was scored in 13 anatomic sites (items) on conventional T2w and T1w images in controls (n = 253, 0-2 years). Items for the score were selected based on inter-rater variability, practicability of scoring, and importance for correctly identifying validation scans. RESULTS: The resulting myelination score consisting of 7 T2- and 5 T1-items delineated myelination from term-equivalent to advanced, incomplete myelination which 50 % and 99 % of controls had reached by 19.1 and 32.7 months, respectively. It correctly identified 20/20 new control MRIs and 40/43 with myelination delay, missing one patient with borderline myelination delay at 8.6 months and 2 patients with incomplete T2-myelination of subcortical temporopolar white matter at 28 and 34 months. CONCLUSIONS: The proposed myelination score provides an easy to use, standardized, and versatile tool to delineate myelination normally occurring during the first 1.5 years of life.

Fully Automated Hippocampus Segmentation using T2-informed Deep Convolutional Neural Networks.

Hippocampal atrophy (tissue loss) has become a fundamental outcome parameter in clinical trials on Alzheimer's disease. To accurately estimate hippocampus volume and track its volume loss, a robust and reliable segmentation is essential. Manual hippocampus segmentation is considered the gold standard but is extensive, time-consuming, and prone to rater bias. Therefore, it is often replaced by automated programs like FreeSurfer, one of the most commonly used tools in clinical research. Recently, deep learning-based methods have also been successfully applied to hippocampus segmentation. The basis of all approaches are clinically used T1-weighted whole-brain MR images with approximately 1 mm isotropic resolution. However, such T1 images show low contrast-to-noise ratios (CNRs), particularly for many hippocampal substructures, limiting delineation reliability. To overcome these limitations, high-resolution T2-weighted scans are suggested for better visualization and delineation, as they show higher CNRs and usually allow for higher resolutions. Unfortunately, such time-consuming T2-weighted sequences are not feasible in a clinical routine. We propose an automated hippocampus segmentation pipeline leveraging deep learning with T2-weighted MR images for enhanced hippocampus segmentation of clinical T1-weighted images based on a series of 3D convolutional neural networks and a specifically acquired multi-contrast dataset. This dataset consists of corresponding pairs of T1- and high-resolution T2-weighted images, with the T2 images only used to create more accurate manual ground truth annotations and to train the segmentation network. The T2-based ground truth labels were also used to evaluate all experiments by comparing the masks visually and by various quantitative measures. We compared our approach with four established state-of-the-art hippocampus segmentation algorithms (FreeSurfer, ASHS, HippoDeep, HippMapp3r) and demonstrated a superior segmentation performance. Moreover, we found that the automated segmentation of T1-weighted images benefits from the T2-based ground truth data. In conclusion, this work showed the beneficial use of high-resolution, T2-based ground truth data for training an automated, deep learning-based hippocampus segmentation and provides the basis for a reliable estimation of hippocampal atrophy in clinical studies.

Brain-age prediction: Systematic evaluation of site effects, and sample age range and size.

Structural neuroimaging data have been used to compute an estimate of the biological age of the brain (brain-age) which has been associated with other biologically and behaviorally meaningful measures of brain development and aging. The ongoing research interest in brain-age has highlighted the need for robust and publicly available brain-age models pre-trained on data from large samples of healthy individuals. To address this need we have previously released a developmental brain-age model. Here we expand this work to develop, empirically validate, and disseminate a pre-trained brain-age model to cover most of the human lifespan. To achieve this, we selected the best-performing model after systematically examining the impact of seven site harmonization strategies, age range, and sample size on brain-age prediction in a discovery sample of brain morphometric measures from 35,683 healthy individuals (age range: 5-90 years; 53.59% female). The pre-trained models were tested for cross-dataset generalizability in an independent sample comprising 2101 healthy individuals (age range: 8-80 years; 55.35% female) and for longitudinal consistency in a further sample comprising 377 healthy individuals (age range: 9-25 years; 49.87% female). This empirical examination yielded the following findings: (1) the accuracy of age prediction from morphometry data was higher when no site harmonization was applied; (2) dividing the discovery sample into two age-bins (5-40 and 40-90 years) provided a better balance between model accuracy and explained age variance than other alternatives; (3) model accuracy for brain-age prediction plateaued at a sample size exceeding 1600 participants. These findings have been incorporated into CentileBrain (https://centilebrain.org/#/brainAGE2), an open-science, web-based platform for individualized neuroimaging metrics.

Stress testing the Centiloid: Precision and variability of PET quantification of amyloid pathology.

INTRODUCTION: Assessing the potential sources of bias and variability of the Centiloid (CL) scale is fundamental for its appropriate clinical application. METHODS: We included 533 participants from AMYloid imaging to Prevent Alzheimer's Disease (AMYPAD DPMS) and Alzheimer's Disease Neuroimaging Initiative (ADNI) cohorts. Thirty-two CL pipelines were created using different combinations of reference region (RR), RR and target types, and quantification spaces. Generalized estimating equations stratified by amyloid positivity were used to assess the impact of the quantification pipeline, radiotracer, age, brain atrophy, and harmonization status on CL. RESULTS: RR selection and RR type impact CL the most, particularly in amyloid-negative individuals. The standard CL pipeline with the whole cerebellum as RR is robust against brain atrophy and differences in image resolution, with 95% confidence intervals below ± 3.95 CL for amyloid beta positivity cutoffs (CL < 24). DISCUSSION: The standard CL pipeline is recommended for most scenarios. Confidence intervals should be considered when operationalizing CL cutoffs in clinical and research settings. HIGHLIGHTS: We developed a framework for evaluating Centiloid (CL) variability to different factors. Reference region selection and delineation had the highest impact on CL values. Whole cerebellum (WCB) and whole cerebellum plus brainstem (WCB+BSTM) as reference regions yielded consistent results across tracers. The standard CL pipeline is robust against atrophy and image resolution variation. Estimated within- and between-pipeline variability (95% confidence interval) in absolute CL units.

DeepComBat: A statistically motivated, hyperparameter-robust, deep learning approach to harmonization of neuroimaging data.

Neuroimaging data acquired using multiple scanners or protocols are increasingly available. However, such data exhibit technical artifacts across batches which introduce confounding and decrease reproducibility. This is especially true when multi-batch data are analyzed using complex downstream models which are more likely to pick up on and implicitly incorporate batch-related information. Previously proposed image harmonization methods have sought to remove these batch effects; however, batch effects remain detectable in the data after applying these methods. We present DeepComBat, a deep learning harmonization method based on a conditional variational autoencoder and the ComBat method. DeepComBat combines the strengths of statistical and deep learning methods in order to account for the multivariate relationships between features while simultaneously relaxing strong assumptions made by previous deep learning harmonization methods. As a result, DeepComBat can perform multivariate harmonization while preserving data structure and avoiding the introduction of synthetic artifacts. We apply this method to cortical thickness measurements from a cognitive-aging cohort and show DeepComBat qualitatively and quantitatively outperforms existing methods in removing batch effects while preserving biological heterogeneity. Additionally, DeepComBat provides a new perspective for statistically motivated deep learning harmonization methods.

Identification of schizophrenia by applying interpretable radiomics modeling with structural magnetic resonance imaging of the cerebellum.

AIMS: The cerebellum is involved in higher-order mental processing as well as sensorimotor functions. Although structural abnormalities in the cerebellum have been demonstrated in schizophrenia, neuroimaging techniques are not yet applicable to identify them given the lack of biomarkers. We aimed to develop a robust diagnostic model for schizophrenia using radiomic features from T1-weighted magnetic resonance imaging (T1-MRI) of the cerebellum. METHODS: A total of 336 participants (174 schizophrenia; 162 healthy controls [HCs]) were allocated to training (122 schizophrenia; 115 HCs) and test (52 schizophrenia; 47 HCs) cohorts. We obtained 2568 radiomic features from T1-MRI of the cerebellar subregions. After feature selection, a light gradient boosting machine classifier was trained. The discrimination and calibration of the model were evaluated. SHapley Additive exPlanations (SHAP) was applied to determine model interpretability. RESULTS: We identified 17 radiomic features to differentiate participants with schizophrenia from HCs. In the test cohort, the radiomics model had an area under the curve, accuracy, sensitivity, and specificity of 0.89 (95% confidence interval: 0.82-0.95), 78.8%, 88.5%, and 75.4%, respectively. The model explanation by SHAP suggested that the second-order size zone non-uniformity feature from the right lobule IX and first-order energy feature from the right lobules V and VI were highly associated with the risk of schizophrenia. CONCLUSION: The radiomics model focused on the cerebellum demonstrates robustness in diagnosing schizophrenia. Our results suggest that microcircuit disruption in the posterior cerebellum is a disease-defining feature of schizophrenia, and radiomics modeling has potential for supporting biomarker-based decision-making in clinical practice.

The role of the Alzheimer's Disease Neuroimaging Initiative in establishing the Dominantly Inherited Alzheimer Network.

The Dominantly Inherited Alzheimer Network (DIAN) initially was funded by the National Institute on Aging (NIA) in 2008 and thus was able to adopt and incorporate the protocols developed by the Alzheimer's Disease Neuroimaging Initiative (ADNI) that had been established by the NIA in 2004. The use of ADNI protocols for DIAN neuroimaging studies and assays of biological fluids for Alzheimer disease (AD) biomarkers permitted examination of the hypothesis that autosomal dominant AD (ADAD), studied by DIAN, and "sporadic" late-onset AD (LOAD), studied by ADNI, shared the same pathobiological construct. In a collaborative effort, the longitudinal DIAN and ADNI databases were compared and the findings supported the conclusion that ADAD and LOAD share a similar pathophysiology. The importance of the DIAN study thus is amplified by its relevance to LOAD, as characterized by the "parent" ADNI program.

An XAI-enhanced efficientNetB0 framework for precision brain tumor detection in MRI imaging.

BACKGROUND: Accurately diagnosing brain tumors from MRI scans is crucial for effective treatment planning. While traditional methods heavily rely on radiologist expertise, the integration of AI, particularly Convolutional Neural Networks (CNNs), has shown promise in improving accuracy. However, the lack of transparency in AI decision-making processes presents a challenge for clinical adoption. METHODS: Recent advancements in deep learning, particularly the utilization of CNNs, have facilitated the development of models for medical image analysis. In this study, we employed the EfficientNetB0 architecture and integrated explainable AI techniques to enhance both accuracy and interpretability. Grad-CAM visualization was utilized to highlight significant areas in MRI scans influencing classification decisions. RESULTS: Our model achieved a classification accuracy of 98.72 % across four categories of brain tumors (Glioma, Meningioma, No Tumor, Pituitary), with precision and recall exceeding 97 % for all categories. The incorporation of explainable AI techniques was validated through visual inspection of Grad-CAM heatmaps, which aligned well with established diagnostic markers in MRI scans. CONCLUSION: The AI-enhanced EfficientNetB0 framework with explainable AI techniques significantly improves brain tumor classification accuracy to 98.72 %, offering clear visual insights into the decision-making process. This method enhances diagnostic reliability and trust, demonstrating substantial potential for clinical adoption in medical diagnostics.

Amyloid-related imaging abnormalities (ARIA): diagnosis, management, and care in the setting of amyloid-modifying therapy.

Amyloid-related imaging abnormalities, were originally described by dementia care experts. The wider use of aducanumab and now lecanemab warrant broader understanding by the health care provider continuum. The optimal care approach for patients with Alzheimer's dementia, treated with amyloid-targeted therapy, includes proper clinical diagnosis, complication surveillance, specific imaging protocols, expert specialty consultation, integrated treatment strategies, and proper facility system planning. Improved awareness and understanding of amyloid-modifying therapy, both benefits and potential complications, among the health care provider continuum is paramount to the success of complex care programs. Specifically, recognition of treatment high risk, high benefit groups, and the interface of concurrent antiplatelet and anticoagulation. This integrated acute, specialty, and primary care approach should improve patient care quality and outcome.

Latent diffusion model-based MRI superresolution enhances mild cognitive impairment prognostication and Alzheimer's disease classification.

INTRODUCTION: Timely diagnosis and prognostication of Alzheimer's disease (AD) and mild cognitive impairment (MCI) are pivotal for effective intervention. Artificial intelligence (AI) in neuroradiology may aid in such appropriate diagnosis and prognostication. This study aimed to evaluate the potential of novel diffusion model-based AI for enhancing AD and MCI diagnosis through superresolution (SR) of brain magnetic resonance (MR) images. METHODS: 1.5T brain MR scans of patients with AD or MCI and healthy controls (NC) from Alzheimer's Disease Neuroimaging Initiative 1 (ADNI1) were superresolved to 3T using a novel diffusion model-based generative AI (d3T*) and a convolutional neural network-based model (c3T*). Comparisons of image quality to actual 1.5T and 3T MRI were conducted based on signal-to-noise ratio (SNR), naturalness image quality evaluator (NIQE), and Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE). Voxel-based volumetric analysis was then conducted to study whether 3T* images offered more accurate volumetry than 1.5T images. Binary and multiclass classifications of AD, MCI, and NC were conducted to evaluate whether 3T* images offered superior AD classification performance compared to actual 1.5T MRI. Moreover, CNN-based classifiers were used to predict conversion of MCI to AD, to evaluate the prognostication performance of 3T* images. The classification performances were evaluated using accuracy, sensitivity, specificity, F1 score, Matthews correlation coefficient (MCC), and area under the receiver-operating curves (AUROC). RESULTS: Analysis of variance (ANOVA) detected significant differences in image quality among the 1.5T, c3T*, d3T*, and 3T groups across all metrics. Both c3T* and d3T* showed superior image quality compared to 1.5T MRI in NIQE and BRISQUE with statistical significance. While the hippocampal volumes measured in 3T* and 3T images were not significantly different, the hippocampal volume measured in 1.5T images showed significant difference. 3T*-based AD classifications showed superior performance across all performance metrics compared to 1.5T-based AD classification. Classification performance between d3T* and actual 3T was not significantly different. 3T* images offered superior accuracy in predicting the conversion of MCI to AD than 1.5T images did. CONCLUSIONS: The diffusion model-based MRI SR enhances the resolution of brain MR images, significantly improving diagnostic and prognostic accuracy for AD and MCI. Superresolved 3T* images closely matched actual 3T MRIs in quality and volumetric accuracy, and notably improved the prediction performance of conversion from MCI to AD.

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.

MRIO: the Magnetic Resonance Imaging Acquisition and Analysis Ontology.

Magnetic resonance imaging of the brain is a useful tool in both the clinic and research settings, aiding in the diagnosis and treatments of neurological disease and expanding our knowledge of the brain. However, there are many challenges inherent in managing and analyzing MRI data, due in large part to the heterogeneity of data acquisition. To address this, we have developed MRIO, the Magnetic Resonance Imaging Acquisition and Analysis Ontology. MRIO provides well-reasoned classes and logical axioms for the acquisition of several MRI acquisition types and well-known, peer-reviewed analysis software, facilitating the use of MRI data. These classes provide a common language for the neuroimaging research process and help standardize the organization and analysis of MRI data for reproducible datasets. We also provide queries for automated assignment of analyses for given MRI types. MRIO aids researchers in managing neuroimaging studies by helping organize and annotate MRI data and integrating with existing standards such as Digital Imaging and Communications in Medicine and the Brain Imaging Data Structure, enhancing reproducibility and interoperability. MRIO was constructed according to Open Biomedical Ontologies Foundry principles and has contributed several classes to the Ontology for Biomedical Investigations to help bridge neuroimaging data to other domains. MRIO addresses the need for a "common language" for MRI that can help manage the neuroimaging research, by enabling researchers to identify appropriate analyses for sets of scans and facilitating data organization and reporting.

Disease-driven domain generalization for neuroimaging-based assessment of Alzheimer's disease.

Development of deep learning models to evaluate structural brain changes caused by cognitive impairment in MRI scans holds significant translational value. The efficacy of these models often encounters challenges due to variabilities arising from different data generation protocols, imaging equipment, radiological artifacts, and shifts in demographic distributions. Domain generalization (DG) techniques show promise in addressing these challenges by enabling the model to learn from one or more source domains and apply this knowledge to new, unseen target domains. Here we present a framework that utilizes model interpretability to enhance the generalizability of classification models across various cohorts. We used MRI scans and clinical diagnoses from four independent cohorts: Alzheimer's Disease Neuroimaging Initiative (ADNI, n = 1821), the Framingham Heart Study (FHS, n = 304), the Australian Imaging Biomarkers & Lifestyle Study of Ageing (AIBL, n = 661), and the National Alzheimer's Coordinating Center (NACC, n = 4647). With this data, we trained a deep neural network to focus on areas of the brain identified as relevant to the disease for model training. Our approach involved training a classifier to differentiate between structural neurodegeneration in individuals with normal cognition (NC), mild cognitive impairment (MCI), and dementia due to Alzheimer's disease (AD). This was achieved by aligning class-wise attention with a unified visual saliency prior, which was computed offline for each class using all the training data. Our method not only competes with state-of-the-art approaches but also shows improved correlation with postmortem histology. This alignment with the gold standard evidence is a significant step towards validating the effectiveness of DG frameworks, paving the way for their broader application in the field.

SAN: Mitigating spatial covariance heterogeneity in cortical thickness data collected from multiple scanners or sites.

In neuroimaging studies, combining data collected from multiple study sites or scanners is becoming common to increase the reproducibility of scientific discoveries. At the same time, unwanted variations arise by using different scanners (inter-scanner biases), which need to be corrected before downstream analyses to facilitate replicable research and prevent spurious findings. While statistical harmonization methods such as ComBat have become popular in mitigating inter-scanner biases in neuroimaging, recent methodological advances have shown that harmonizing heterogeneous covariances results in higher data quality. In vertex-level cortical thickness data, heterogeneity in spatial autocorrelation is a critical factor that affects covariance heterogeneity. Our work proposes a new statistical harmonization method called spatial autocorrelation normalization (SAN) that preserves homogeneous covariance vertex-level cortical thickness data across different scanners. We use an explicit Gaussian process to characterize scanner-invariant and scanner-specific variations to reconstruct spatially homogeneous data across scanners. SAN is computationally feasible, and it easily allows the integration of existing harmonization methods. We demonstrate the utility of the proposed method using cortical thickness data from the Social Processes Initiative in the Neurobiology of the Schizophrenia(s) (SPINS) study. SAN is publicly available as an R package.

The Current Diagnostic Performance of MRI-Based Radiomics for Glioma Grading: A Meta-Analysis.

BACKGROUND: Multiple radiomics models have been proposed for grading glioma using different algorithms, features, and sequences of magnetic resonance imaging. The research seeks to assess the present overall performance of radiomics for grading glioma. METHODS: A systematic literature review of the databases Ovid MEDLINE PubMed, and Ovid EMBASE for publications published on radiomics for glioma grading between 2012 and 2023 was performed. The systematic review was carried out following the criteria of Preferred Reporting Items for Systematic Reviews and Meta-Analysis. RESULTS: In the meta-analysis, a total of 7654 patients from 40 articles, were assessed. R-package mada was used for modeling the joint estimates of specificity (SPE) and sensitivity (SEN). Pooled event rates across studies were performed with a random-effects meta-analysis. The heterogeneity of SPE and SEN were based on the χ2 test. Overall values for SPE and SEN in the differentiation between high-grade gliomas (HGGs) and low-grade gliomas (LGGs) were 84% and 91%, respectively. With regards to the discrimination between World Health Organization (WHO) grade 4 and WHO grade 3, the overall SPE was 81% and the SEN was 89%. The modern non-linear classifiers showed a better trend, whereas textural features tend to be the best-performing (29%) and the most used. CONCLUSIONS: Our findings confirm that present radiomics' diagnostic performance for glioma grading is superior in terms of SEN and SPE for the HGGs vs. LGGs discrimination task when compared to the WHO grade 4 vs. 3 task.

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.