Deep learning encodes robust discriminative neuroimaging representations to outperform standard machine learning

Single-shot echo planar time-resolved imaging for multi-echo functional MRI and distortion-free diffusion imaging

PURPOSE

To develop a single-shot SNR-efficient distortion-free multi-echo imaging technique for dynamic imaging applications.

METHODS

Echo planar time-resolved imaging (EPTI) was first introduced as a multi-shot technique for distortion-free multi-echo imaging. This work aims to develop single-shot EPTI (ss-EPTI) to achieve improved robustness to motion/physiological noise, increased temporal resolution, and higher SNR efficiency. A new spatiotemporal encoding that enables reduced phase-encoding blips and minimized echo spacing under the single-shot regime was developed, which improves sampling efficiency and enhances spatiotemporal correlation in the k-TE space for improved reconstruction. A continuous readout with minimized deadtime was employed to optimize SNR efficiency. Moreover, k-TE partial Fourier and simultaneous multi-slice acquisition were integrated for further acceleration.

RESULTS

ss-EPTI provided distortion-free imaging with densely sampled multi-echo images at standard resolutions (e.g., ˜1.25 to 3 mm) in a single-shot. Improved SNR efficiency was observed in ss-EPTI due to improved motion/physiological-noise robustness and efficient continuous readout. Its ability to eliminate dynamic distortions-geometric changes across dynamics due to field changes induced by physiological variations or eddy currents-further improved the data's temporal stability. For multi-echo fMRI, ss-EPTI's multi-echo images recovered signal dropout in short-T 2 * $$ {\mathrm{T}}_2^{\ast } $$ regions and provided TE-dependent functional information to distinguish non-BOLD noise for further tSNR improvement. For diffusion MRI, it achieved shortened TEs for improved SNR and provided images free from both B0-induced and diffusion-encoding-dependent eddy-current-induced distortions with multi-TE diffusion metrics.

CONCLUSION

ss-EPTI provides SNR-efficient distortion-free multi-echo imaging with comparable temporal resolutions to ss-EPI, offering a new acquisition tool for dynamic imaging.

Can artificial intelligence be the future solution to the enormous challenges and suffering caused by Schizophrenia?

This study evaluated the potential of artificial intelligence (AI) in the diagnosis, treatment, and prognostic assessment of schizophrenia (SZ) and explored collaborative directions for AI applications in future medical innovations. SZ is a severe mental disorder that causes significant suffering and imposes challenges on patients. With the rapid advancement of machine learning and deep learning technologies, AI has demonstrated notable advantages in the early diagnosis of high-risk populations. By integrating multidimensional biomarkers and linguistic behavior data of patients, AI can provide further objective and precise diagnostic criteria. Moreover, it aids in formulating personalized treatment plans, enhancing therapeutic outcomes, and offering new therapeutic strategies for patients with treatment-resistant SZ. Furthermore, AI excels in developing individualized prognostic plans, which enables the rapid identification of disease progression, accurate prediction of disease trajectory, and timely adjustment of treatment strategies, thereby improving prognosis and facilitating recovery. Despite the immense potential of AI in SZ management, its role as an auxiliary tool must be emphasized, with clinical judgment and compassionate care from healthcare professionals remaining crucial. Future research should focus on optimizing human-machine interactions to achieve efficient AI application in SZ management. The in-depth integration of AI technology into clinical practice will advance the field of SZ, ultimately improving the quality of life and treatment outcomes of patients.

Explainable Self-Supervised Dynamic Neuroimaging Using Time Reversal

OBJECTIVE

Functional magnetic resonance imaging data pose significant challenges due to their inherently noisy and complex nature, making traditional statistical models less effective in capturing predictive features. While deep learning models offer superior performance through their non-linear capabilities, they often lack transparency, reducing trust in their predictions. This study introduces the Time Reversal (TR) pretraining method to address these challenges. TR aims to learn temporal dependencies in data, leveraging large datasets for pretraining and applying this knowledge to improve schizophrenia classification on smaller datasets.

METHODS

We pretrained an LSTM-based model with attention using the TR approach, focusing on learning the direction of time in fMRI data, achieving over 98 % accuracy on HCP and UK Biobank datasets. For downstream schizophrenia classification, TR-pretrained weights were transferred to models evaluated on FBIRN, COBRE, and B-SNIP datasets. Saliency maps were generated using Integrated Gradients (IG) to provide post hoc explanations for pretraining, while Earth Mover's Distance (EMD) quantified the temporal dynamics of salient features in the downstream tasks.

RESULTS

TR pretraining significantly improved schizophrenia classification performance across all datasets: median AUC scores increased from 0.7958 to 0.8359 (FBIRN), 0.6825 to 0.7778 (COBRE), and 0.6341 to 0.7224 (B-SNIP). The saliency maps revealed more concentrated and biologically meaningful salient features along the time axis, aligning with the episodic nature of schizophrenia. TR consistently outperformed baseline pretraining methods, including OCP and PCL, in terms of AUC, balanced accuracy, and robustness.

CONCLUSIONS

This study demonstrates the dual benefits of the TR method: enhanced predictive performance and improved interpretability. By aligning model predictions with meaningful temporal patterns in brain activity, TR bridges the gap between deep learning and clinical relevance. These findings emphasize the potential of explainable AI tools for aiding clinicians in diagnostics and treatment planning, especially in conditions characterized by disrupted temporal dynamics.

Texture analysis combined with machine learning in radiographs of the knee joint: potential to identify tibial plateau occult fractures

Background

Missed or delayed diagnosis of occult fractures of tibial plateau may cause adverse effects on patients. The objective of this study was to evaluate the diagnostic performance of texture analysis (TA) of knee joint radiographs combined with machine learning (ML) in identifying patients at risk of tibial plateau occult fractures.

Methods

A total of 169 patients with negative fracture on knee X-ray films from 2018 to 2022 who were diagnosed with occult tibial plateau fractures or no fractures by subsequent magnetic resonance imaging (MRI) examination were retrospectively enrolled. The X-ray images of the patient's knee joint were used for texture feature extraction. A total of 9 ML feature selection methods (including 6 mainstream methods and 3 methods provided by MaZda software) combined with 3 classification methods were used to build the best diagnostic model. The performance of each model was evaluated by accuracy, F1-value, and area under the curve (AUC).

Results

The least absolute shrinkage and selection operator (LASSO) method had the best performance of the 6 mainstream methods, with an accuracy of 0.81, an F1 value of 0.80, and an AUC of 0.920, all of which were higher than those of the other five methods (accuracy range: 0.65-0.80, F1 score range: 0.61-0.79, AUC range: 0.722-0.895). Among the three feature selection models in MaZda software, the most ideal method for accuracy measurement was the MI method, reaching 0.77. In the measurement of the F1 value and AUC, MaZda's best method was Fisher, reaching 0.78 and 0.888, respectively. All indicators were lower than those of the LASSO method. The combination of LASSO and support vector machine (SVM) yielded the best classification performance, while the performance of the combination of LASSO and logistic regression was slightly inferior, but the difference was not statistically significant.

Conclusions

TA of knee joint radiography combined with ML has achieved high performance in identifying patients at risk of occult fractures of the tibial plateau. Considering both the model performance and computational complexity, the LASSO feature selection method combined with the logistic regression classifier yielded the best classification performance in this process.

The Transition From Homogeneous to Heterogeneous Machine Learning in Neuropsychiatric Research

Despite the advantage of neuroimaging-based machine learning (ML) models as pivotal tools for investigating brain-behavior relationships in neuropsychiatric studies, these data-driven predictive approaches have yet to yield substantial, clinically actionable insights for mental health care. A notable impediment lies in the inadequate accommodation of most ML research to the natural heterogeneity within large samples. Although commonly thought of as individual-level analyses, many ML algorithms are unimodal and homogeneous and thus incapable of capturing the potentially heterogeneous relationships between biology and psychopathology. We review the current landscape of computational research targeting population heterogeneity and argue that there is a need to expand from brain subtyping and behavioral phenotyping to analyses that focus on heterogeneity at the relational level. To this end, we review and suggest several existing ML models with the capacity to discern how external environmental and sociodemographic factors moderate the brain-behavior mapping function in a data-driven fashion. These heterogeneous ML models hold promise for enhancing the discovery of individualized brain-behavior associations and advancing precision psychiatry.

Explainable brain age prediction: a comparative evaluation of morphometric and deep learning pipelines

Brain age, a biomarker reflecting brain health relative to chronological age, is increasingly used in neuroimaging to detect early signs of neurodegenerative diseases and support personalized treatment plans. Two primary approaches for brain age prediction have emerged: morphometric feature extraction from MRI scans and deep learning (DL) applied to raw MRI data. However, a systematic comparison of these methods regarding performance, interpretability, and clinical utility has been limited. In this study, we present a comparative evaluation of two pipelines: one using morphometric features from FreeSurfer and the other employing 3D convolutional neural networks (CNNs). Using a multisite neuroimaging dataset, we assessed both model performance and the interpretability of predictions through eXplainable Artificial Intelligence (XAI) methods, applying SHAP to the feature-based pipeline and Grad-CAM and DeepSHAP to the CNN-based pipeline. Our results show comparable performance between the two pipelines in Leave-One-Site-Out (LOSO) validation, achieving state-of-the-art performance on the independent test set ( M A E = 3.21 with DNN and morphometric features and M A E = 3.08 with a DenseNet-121 architecture). SHAP provided the most consistent and interpretable results, while DeepSHAP exhibited greater variability. Further work is needed to assess the clinical utility of Grad-CAM. This study addresses a critical gap by systematically comparing the interpretability of multiple XAI methods across distinct brain age prediction pipelines. Our findings underscore the importance of integrating XAI into clinical practice, offering insights into how XAI outputs vary and their potential utility for clinicians.

A multimodal vision transformer for interpretable fusion of functional and structural neuroimaging data

Multimodal neuroimaging is an emerging field that leverages multiple sources of information to diagnose specific brain disorders, especially when deep learning-based AI algorithms are applied. The successful combination of different brain imaging modalities using deep learning remains a challenging yet crucial research topic. The integration of structural and functional modalities is particularly important for the diagnosis of various brain disorders, where structural information plays a crucial role in diseases such as Alzheimer's, while functional imaging is more critical for disorders such as schizophrenia. However, the combination of functional and structural imaging modalities can provide a more comprehensive diagnosis. In this work, we present MultiViT, a novel diagnostic deep learning model that utilizes vision transformers and cross-attention mechanisms to effectively fuse information from 3D gray matter maps derived from structural MRI with functional network connectivity matrices obtained from functional MRI using the ICA algorithm. MultiViT achieves an AUC of 0.833, outperforming both our unimodal and multimodal baselines, enabling more accurate classification and diagnosis of schizophrenia. In addition, using vision transformer's unique attentional maps in combination with cross-attentional mechanisms and brain function information, we identify critical brain regions in 3D gray matter space associated with the characteristics of schizophrenia. Our research not only significantly improves the accuracy of AI-based automated imaging diagnostics for schizophrenia, but also pioneers a rational and advanced data fusion approach by replacing complex, high-dimensional fMRI information with functional network connectivity, integrating it with representative structural data from 3D gray matter images, and further providing interpretative biomarker localization in a 3D structural space.

A Method for Multimodal IVA Fusion Within a MISA Unified Model Reveals Markers of Age, Sex, Cognition, and Schizophrenia in Large Neuroimaging Studies

With the increasing availability of large-scale multimodal neuroimaging datasets, it is necessary to develop data fusion methods which can extract cross-modal features. A general framework, multidataset independent subspace analysis (MISA), has been developed to encompass multiple blind source separation approaches and identify linked cross-modal sources in multiple datasets. In this work, we utilized the multimodal independent vector analysis (MMIVA) model in MISA to directly identify meaningful linked features across three neuroimaging modalities-structural magnetic resonance imaging (MRI), resting state functional MRI and diffusion MRI-in two large independent datasets, one comprising of control subjects and the other including patients with schizophrenia. Results show several linked subject profiles (sources) that capture age-associated decline, schizophrenia-related biomarkers, sex effects, and cognitive performance. For sources associated with age, both shared and modality-specific brain-age deltas were evaluated for association with non-imaging variables. In addition, each set of linked sources reveals a corresponding set of cross-modal spatial patterns that can be studied jointly. We demonstrate that the MMIVA fusion model can identify linked sources across multiple modalities, and that at least one set of linked, age-related sources replicates across two independent and separately analyzed datasets. The same set also presented age-adjusted group differences, with schizophrenia patients indicating lower multimodal source levels. Linked sets associated with sex and cognition are also reported for the UK Biobank dataset.

A simple but tough-to-beat baseline for fMRI time-series classification

Current neuroimaging studies frequently use complex machine learning models to classify human fMRI data, distinguishing healthy and disordered brains, often to validate new methods or enhance prediction accuracy. Yet, where prediction accuracy is a concern, our results suggest that precision in prediction does not always require such sophistication. When a classifier as simple as logistic regression is applied to feature-engineered fMRI data, it can match or even outperform more sophisticated recent models. Classification of the raw time series fMRI data generally benefits from complex parameter-rich models. However, this complexity often pushes them into the class of black-box models. Yet, we found that a relatively simple model can consistently outperform much more complex classifiers in both accuracy and speed. This model applies the same multi-layer perceptron repeatedly across time and averages the results. Thus, the complexity and black-box nature of the parameter rich models, often perceived as a necessary trade-off for higher performance, do not invariably yield superior results on fMRI. Given the success of straightforward approaches, we challenge the merit of research that concentrates solely on complex model development driven by classification. Instead, we advocate for increased focus on designing models that prioritize the explainability of fMRI data or pursue applicable objectives beyond mere classification accuracy, unless they significantly outperform logistic regression or our proposed model. To validate our claim, we explore possible reasons for the superior performance of our straightforward model by examining the innate characteristics of fMRI time series data. Our findings suggest that the sequential information hidden in the temporal order may be far less important for the accurate fMRI classification than the stand-alone pieces of information scattered across the frames of the time series.

NeuroGraph: Benchmarks for Graph Machine Learning in Brain Connectomics

Machine learning provides a valuable tool for analyzing high-dimensional functional neuroimaging data, and is proving effective in predicting various neurological conditions, psychiatric disorders, and cognitive patterns. In functional magnetic resonance imaging (MRI) research, interactions between brain regions are commonly modeled using graph-based representations. The potency of graph machine learning methods has been established across myriad domains, marking a transformative step in data interpretation and predictive modeling. Yet, despite their promise, the transposition of these techniques to the neuroimaging domain has been challenging due to the expansive number of potential preprocessing pipelines and the large parameter search space for graph-based dataset construction. In this paper, we introduce NeuroGraph, a collection of graph-based neuroimaging datasets, and demonstrated its utility for predicting multiple categories of behavioral and cognitive traits. We delve deeply into the dataset generation search space by crafting 35 datasets that encompass static and dynamic brain connectivity, running in excess of 15 baseline methods for benchmarking. Additionally, we provide generic frameworks for learning on both static and dynamic graphs. Our extensive experiments lead to several key observations. Notably, using correlation vectors as node features, incorporating larger number of regions of interest, and employing sparser graphs lead to improved performance. To foster further advancements in graph-based data driven neuroimaging analysis, we offer a comprehensive open-source Python package that includes the benchmark datasets, baseline implementations, model training, and standard evaluation.

Predictive and Explainable Artificial Intelligence for Neuroimaging Applications

BACKGROUND

The aim of this review is to highlight the new advance of predictive and explainable artificial intelligence for neuroimaging applications.

METHODS

Data came from 30 original studies in PubMed with the following search terms: "neuroimaging" (title) together with "machine learning" (title) or "deep learning" (title). The 30 original studies were eligible according to the following criteria: the participants with the dependent variable of brain image or associated disease; the interventions/comparisons of artificial intelligence; the outcomes of accuracy, the area under the curve (AUC), and/or variable importance; the publication year of 2019 or later; and the publication language of English.

RESULTS

The performance outcomes reported were within 58-96 for accuracy (%), 66-97 for sensitivity (%), 76-98 for specificity (%), and 70-98 for the AUC (%). The support vector machine and the convolutional neural network registered the best performance (AUC 98%) for the classifications of low- vs. high-grade glioma and brain conditions, respectively. Likewise, the random forest delivered the best performance (root mean square error 1) for the regression of brain conditions. The following factors were discovered to be major predictors of brain image or associated disease: (demographic) age, education, sex; (health-related) alpha desynchronization, Alzheimer's disease stage, CD4, depression, distress, mild behavioral impairment, RNA sequencing; (neuroimaging) abnormal amyloid-β, amplitude of low-frequency fluctuation, cortical thickness, functional connectivity, fractal dimension measure, gray matter volume, left amygdala activity, left hippocampal volume, plasma neurofilament light, right cerebellum, regional homogeneity, right middle occipital gyrus, surface area, sub-cortical volume.

CONCLUSION

Predictive and explainable artificial intelligence provide an effective, non-invasive decision support system for neuroimaging applications.

Advances and challenges in neuroimaging-based pain biomarkers

Identifying neural biomarkers of pain has long been a central theme in pain neuroscience. Here, we review the state-of-the-art candidates for neural biomarkers of acute and chronic pain. We classify these potential neural biomarkers into five categories based on the nature of their target variables, including neural biomarkers of (1) within-individual perception, (2) between-individual sensitivity, and (3) discriminability for acute pain, as well as (4) assessment and (5) prospective neural biomarkers for chronic pain. For each category, we provide a synthesized review of candidate biomarkers developed using neuroimaging techniques including functional magnetic resonance imaging (fMRI), structural magnetic resonance imaging (sMRI), and electroencephalography (EEG). We also discuss the conceptual and practical challenges in developing neural biomarkers of pain. Addressing these challenges, optimal biomarkers of pain can be developed to deepen our understanding of how the brain represents pain and ultimately help alleviate patients' suffering and improve their well-being.

Adding the third dimension: 3D convolutional neural network diagnosis of temporal lobe epilepsy

Convolutional neural networks (CNN) show great promise for translating decades of research on structural abnormalities in temporal lobe epilepsy into clinical practice. Three-dimensional CNNs typically outperform two-dimensional CNNs in medical imaging. Here we explore for the first time whether a three-dimensional CNN outperforms a two-dimensional CNN for identifying temporal lobe epilepsy-specific features on MRI. Using 1178 T1-weighted images (589 temporal lobe epilepsy, 589 healthy controls) from 12 surgical centres, we trained 3D and 2D CNNs for temporal lobe epilepsy versus healthy control classification, using feature visualization to identify important regions. The 3D CNN was compared to the 2D model and to a randomized model (comparison to chance). Further, we explored the effect of sample size with subsampling, examined model performance based on single-subject clinical characteristics, and tested the impact of image harmonization on model performance. Across 50 datapoints (10 runs with 5-folds each) the 3D CNN median accuracy was 86.4% (35.3% above chance) and the median F1-score was 86.1% (33.3% above chance). The 3D model yielded higher accuracy compared to the 2D model on 84% of datapoints (median 2D accuracy, 83.0%), a significant outperformance for the 3D model (binomial test: P < 0.001). This advantage of the 3D model was only apparent at the highest sample size. Saliency maps exhibited the importance of medial-ventral temporal, cerebellar, and midline subcortical regions across both models for classification. However, the 3D model had higher salience in the most important regions, the ventral-medial temporal and midline subcortical regions. Importantly, the model achieved high accuracy (82% accuracy) even in patients without MRI-identifiable hippocampal sclerosis. Finally, applying ComBat for harmonization did not improve performance. These findings highlight the value of 3D CNNs for identifying subtle structural abnormalities on MRI, especially in patients without clinically identified temporal lobe epilepsy lesions. Our findings also reveal that the advantage of 3D CNNs relies on large sample sizes for model training.

A perspective on brain-age estimation and its clinical promise

Brain-age estimation has gained increased attention in the neuroscientific community owing to its potential use as a biomarker of brain health. The difference between estimated and chronological age based on neuroimaging data enables a unique perspective on brain development and aging, with multiple open questions still remaining in the brain-age research field. This Perspective presents an overview of current advancements in the field and envisions the future evolution of the brain-age framework before its potential deployment in hospital settings.

Attention-Enhanced Fusion of Structural and Functional MRI for Analyzing HIV-Associated Asymptomatic Neurocognitive Impairment

Asymptomatic neurocognitive impairment (ANI) is a predominant form of cognitive impairment among individuals infected with human immunodeficiency virus (HIV). The current diagnostic criteria for ANI primarily rely on subjective clinical assessments, possibly leading to different interpretations among clinicians. Some recent studies leverage structural or functional MRI containing objective biomarkers for ANI analysis, offering clinicians companion diagnostic tools. However, they mainly utilize a single imaging modality, neglecting complementary information provided by structural and functional MRI. To this end, we propose an attention-enhanced structural and functional MRI fusion (ASFF) framework for HIV-associated ANI analysis. Specifically, the ASFF first extracts data-driven and human-engineered features from structural MRI, and also captures functional MRI features via a graph isomorphism network and Transformer. A mutual cross-attention fusion module is then designed to model the underlying relationship between structural and functional MRI. Additionally, a semantic inter-modality constraint is introduced to encourage consistency of multimodal features, facilitating effective feature fusion. Experimental results on 137 subjects from an HIV-associated ANI dataset with T1-weighted MRI and resting-state functional MRI show the effectiveness of our ASFF in ANI identification. Furthermore, our method can identify both modality-shared and modality-specific brain regions, which may advance our understanding of the structural and functional pathology underlying ANI.

Efficient federated learning for distributed neuroimaging data

Recent advancements in neuroimaging have led to greater data sharing among the scientific community. However, institutions frequently maintain control over their data, citing concerns related to research culture, privacy, and accountability. This creates a demand for innovative tools capable of analyzing amalgamated datasets without the need to transfer actual data between entities. To address this challenge, we propose a decentralized sparse federated learning (FL) strategy. This approach emphasizes local training of sparse models to facilitate efficient communication within such frameworks. By capitalizing on model sparsity and selectively sharing parameters between client sites during the training phase, our method significantly lowers communication overheads. This advantage becomes increasingly pronounced when dealing with larger models and accommodating the diverse resource capabilities of various sites. We demonstrate the effectiveness of our approach through the application to the Adolescent Brain Cognitive Development (ABCD) dataset.

H-Net: Heterogeneous Neural Network for Multi-Classification of Neuropsychiatric Disorders

Clinical studies have proved that both structural magnetic resonance imaging (sMRI) and functional magnetic resonance imaging (fMRI) are implicitly associated with neuropsychiatric disorders (NDs), and integrating multi-modal to the binary classification of NDs has been thoroughly explored. However, accurately classifying multiple classes of NDs remains a challenge due to the complexity of disease subclass. In our study, we develop a heterogeneous neural network (H-Net) that integrates sMRI and fMRI modes for classifying multi-class NDs. To account for the differences between the two modes, H-Net adopts a heterogeneous neural network strategy to extract information from each mode. Specifically, H-Net includes an multi-layer perceptron based (MLP-based) encoder, a graph attention network based (GAT-based) encoder, and a cross-modality transformer block. The MLP-based and GAT-based encoders extract semantic features from sMRI and features from fMRI, respectively, while the cross-modality transformer block models the attention of two types of features. In H-Net, the proposed MLP-mixer block and cross-modality alignment are powerful tools for improving the multi-classification performance of NDs. H-Net is validate on the public dataset (CNP), where H-Net achieves 90% classification accuracy in diagnosing multi-class NDs. Furthermore, we demonstrate the complementarity of the two MRI modalities in improving the identification of multi-class NDs. Both visual and statistical analyses show the differences between ND subclasses.

Associations of plasma proteomics and age-related outcomes with brain age in a diverse cohort

Machine learning models are increasingly being used to estimate "brain age" from neuroimaging data. The gap between chronological age and the estimated brain age gap (BAG) is potentially a measure of accelerated and resilient brain aging. Brain age calculated in this fashion has been shown to be associated with mortality, measures of physical function, health, and disease. Here, we estimate the BAG using a voxel-based elastic net regression approach, and then, we investigate its associations with mortality, cognitive status, and measures of health and disease in participants from Atherosclerosis Risk in Communities (ARIC) study who had a brain MRI at visit 5 of the study. Finally, we used the SOMAscan assay containing 4877 proteins to examine the proteomic associations with the MRI-defined BAG. Among N = 1849 participants (age, 76.4 (SD 5.6)), we found that increased values of BAG were strongly associated with increased mortality and increased severity of the cognitive status. Strong associations with mortality persisted when the analyses were performed in cognitively normal participants. In addition, it was strongly associated with BMI, diabetes, measures of physical function, hypertension, prevalent heart disease, and stroke. Finally, we found 33 proteins associated with BAG after a correction for multiple comparisons. The top proteins with positive associations to brain age were growth/differentiation factor 15 (GDF-15), Sushi, von Willebrand factor type A, EGF, and pentraxin domain-containing protein 1 (SEVP 1), matrilysin (MMP7), ADAMTS-like protein 2 (ADAMTS), and heat shock 70 kDa protein 1B (HSPA1B) while EGF-receptor (EGFR), mast/stem-cell-growth-factor-receptor (KIT), coagulation-factor-VII, and cGMP-dependent-protein-kinase-1 (PRKG1) were negatively associated to brain age. Several of these proteins were previously associated with dementia in ARIC. These results suggest that circulating proteins implicated in biological aging, cellular senescence, angiogenesis, and coagulation are associated with a neuroimaging measure of brain aging.

Exploring the potential of representation and transfer learning for anatomical neuroimaging: Application to psychiatry

The perspective of personalized medicine for brain disorders requires efficient learning models for anatomical neuroimaging-based prediction of clinical conditions. There is now a consensus on the benefit of deep learning (DL) in addressing many medical imaging tasks, such as image segmentation. However, for single-subject prediction problems, recent studies yielded contradictory results when comparing DL with Standard Machine Learning (SML) on top of classical feature extraction. Most existing comparative studies were limited in predicting phenotypes of little clinical interest, such as sex and age, and using a single dataset. Moreover, they conducted a limited analysis of the employed image pre-processing and feature selection strategies. This paper extensively compares DL and SML prediction capacity on five multi-site problems, including three increasingly complex clinical applications in psychiatry namely schizophrenia, bipolar disorder, and Autism Spectrum Disorder (ASD) diagnosis. To compensate for the relative scarcity of neuroimaging data on these clinical datasets, we also evaluate three pre-training strategies for transfer learning from brain imaging of the general healthy population: self-supervised learning, generative modeling and supervised learning with age. Overall, we find similar performance between randomly initialized DL and SML for the three clinical tasks and a similar scaling trend for sex prediction. This was replicated on an external dataset. We also show highly correlated discriminative brain regions between DL and linear ML models in all problems. Nonetheless, we demonstrate that self-supervised pre-training on large-scale healthy population imaging datasets (N≈10k), along with Deep Ensemble, allows DL to learn robust and transferable representations to smaller-scale clinical datasets (N≤1k). It largely outperforms SML on 2 out of 3 clinical tasks both in internal and external test sets. These findings suggest that the improvement of DL over SML in anatomical neuroimaging mainly comes from its capacity to learn meaningful and useful abstract representations of the brain anatomy, and it sheds light on the potential of transfer learning for personalized medicine in psychiatry.

Multimodal active subspace analysis for computing assessment oriented subspaces from neuroimaging data

BACKGROUND

For successful biomarker discovery, it is essential to develop computational frameworks that summarize high-dimensional neuroimaging data in terms of involved sub-systems of the brain, while also revealing underlying heterogeneous functional and structural changes covarying with specific cognitive and biological traits. However, unsupervised decompositions do not inculcate clinical assessment information, while supervised approaches extract only individual feature importance, thereby impeding qualitative interpretation at the level of subspaces.

NEW METHOD

We present a novel framework to extract robust multimodal brain subspaces associated with changes in a given cognitive or biological trait. Our approach involves active subspace learning on the gradients of a trained machine learning model followed by clustering to extract and summarize the most salient and consistent subspaces associated with the target variable.

RESULTS

Through a rigorous cross-validation procedure on an Alzheimer's disease (AD) dataset, our framework successfully extracts multimodal subspaces specific to a given clinical assessment (e.g., memory and other cognitive skills), and also retains predictive performance in standard machine learning algorithms. We also show that the salient active subspace directions occur consistently across randomly sub-sampled repetitions of the analysis.

COMPARISON WITH EXISTING METHOD(S)

Compared to existing unsupervised decompositions based on principle component analysis, the subspace components in our framework retain higher predictive information.

CONCLUSIONS

As an important step towards biomarker discovery, our framework not only uncovers AD-related brain regions in the associated brain subspaces, but also enables automated identification of multiple underlying structural and functional sub-systems of the brain that collectively characterize changes in memory and proficiency in cognitive skills related to brain disorders like AD.

Efficient Federated Learning for distributed NeuroImaging Data

Recent advancements in neuroimaging have led to greater data sharing among the scientific community. However, institutions frequently maintain control over their data, citing concerns related to research culture, privacy, and accountability. This creates a demand for innovative tools capable of analyzing amalgamated datasets without the need to transfer actual data between entities. To address this challenge, we propose a decentralized sparse federated learning (FL) strategy. This approach emphasizes local training of sparse models to facilitate efficient communication within such frameworks. By capitalizing on model sparsity and selectively sharing parameters between client sites during the training phase, our method significantly lowers communication overheads. This advantage becomes increasingly pronounced when dealing with larger models and accommodating the diverse resource capabilities of various sites. We demonstrate the effectiveness of our approach through the application to the Adolescent Brain Cognitive Development (ABCD) dataset.

White matter diffusion estimates in obsessive-compulsive disorder across 1653 individuals: machine learning findings from the ENIGMA OCD Working Group

White matter pathways, typically studied with diffusion tensor imaging (DTI), have been implicated in the neurobiology of obsessive-compulsive disorder (OCD). However, due to limited sample sizes and the predominance of single-site studies, the generalizability of OCD classification based on diffusion white matter estimates remains unclear. Here, we tested classification accuracy using the largest OCD DTI dataset to date, involving 1336 adult participants (690 OCD patients and 646 healthy controls) and 317 pediatric participants (175 OCD patients and 142 healthy controls) from 18 international sites within the ENIGMA OCD Working Group. We used an automatic machine learning pipeline (with feature engineering and selection, and model optimization) and examined the cross-site generalizability of the OCD classification models using leave-one-site-out cross-validation. Our models showed low-to-moderate accuracy in classifying (1) "OCD vs. healthy controls" (Adults, receiver operator characteristic-area under the curve = 57.19 ± 3.47 in the replication set; Children, 59.8 ± 7.39), (2) "unmedicated OCD vs. healthy controls" (Adults, 62.67 ± 3.84; Children, 48.51 ± 10.14), and (3) "medicated OCD vs. unmedicated OCD" (Adults, 76.72 ± 3.97; Children, 72.45 ± 8.87). There was significant site variability in model performance (cross-validated ROC AUC ranges 51.6-79.1 in adults; 35.9-63.2 in children). Machine learning interpretation showed that diffusivity measures of the corpus callosum, internal capsule, and posterior thalamic radiation contributed to the classification of OCD from HC. The classification performance appeared greater than the model trained on grey matter morphometry in the prior ENIGMA OCD study (our study includes subsamples from the morphometry study). Taken together, this study points to the meaningful multivariate patterns of white matter features relevant to the neurobiology of OCD, but with low-to-moderate classification accuracy. The OCD classification performance may be constrained by site variability and medication effects on the white matter integrity, indicating room for improvement for future research.

A Learnable Counter-Condition Analysis Framework for Functional Connectivity-Based Neurological Disorder Diagnosis

To understand the biological characteristics of neurological disorders with functional connectivity (FC), recent studies have widely utilized deep learning-based models to identify the disease and conducted post-hoc analyses via explainable models to discover disease-related biomarkers. Most existing frameworks consist of three stages, namely, feature selection, feature extraction for classification, and analysis, where each stage is implemented separately. However, if the results at each stage lack reliability, it can cause misdiagnosis and incorrect analysis in afterward stages. In this study, we propose a novel unified framework that systemically integrates diagnoses (i.e., feature selection and feature extraction) and explanations. Notably, we devised an adaptive attention network as a feature selection approach to identify individual-specific disease-related connections. We also propose a functional network relational encoder that summarizes the global topological properties of FC by learning the inter-network relations without pre-defined edges between functional networks. Last but not least, our framework provides a novel explanatory power for neuroscientific interpretation, also termed counter-condition analysis. We simulated the FC that reverses the diagnostic information (i.e., counter-condition FC): converting a normal brain to be abnormal and vice versa. We validated the effectiveness of our framework by using two large resting-state functional magnetic resonance imaging (fMRI) datasets, Autism Brain Imaging Data Exchange (ABIDE) and REST-meta-MDD, and demonstrated that our framework outperforms other competing methods for disease identification. Furthermore, we analyzed the disease-related neurological patterns based on counter-condition analysis.

Diagnosis of cognitive and motor disorders levels in stroke patients through explainable machine learning based on MRI

BACKGROUND

Globally, stroke is the third most significant cause of disability. A stroke may produce motor, sensory, perceptual, or cognitive disorders that result in disability and affect the likelihood of recovery, affecting a person's ability to function. Evaluation post-stroke is critical for optimal stroke care.

PURPOSE

Traditional methods for classifying the clinical disorders of cognitive and motor in stroke patients use assessment and interrogative measures, which are time-consuming, complex, and labor-intensive. In response to the current situation, this study develops an algorithm to automatically classify motor and cognitive disorders in stroke patients by 3D brain MRI to assist physicians in diagnosis.

METHODS

First, radiomics and fusion features are extracted from the OAx T2 Propeller of 3D brain MRI. Then, we use 14 machine learning models and one model ensemble method to predict Fugl-Meyer and MMSE levels of stroke patients. Next, we evaluate the models using accuracy, recall, f1-score, and area under the curve (AUC). Finally, we employ SHAP to explain the output of the model.

RESULTS

The best predictive models come from Random Forest (RF) Classifier with fusion features in cognitive classification and Linear Discriminant Analysis (LDA) with radiomics features in motor classification. The highest accuracies are 92.0 and 82.5% for cognitive and motor disorders.

CONCLUSIONS

MRI brain maps can classify the cognitive and motor disorders of stroke patients. Radiomics features demonstrate its merits. The proposed algorithms with MRI images can efficiently assist physicians in diagnosing the cognitive and motor disorders of stroke patients in clinical practice. Additionally, this lessens labor costs, improves diagnostic effectiveness, and avoids the subjective difference that comes with manual assessment.

BrainAGE: Revisited and reframed machine learning workflow

Since the introduction of the BrainAGE method, novel machine learning methods for brain age prediction have continued to emerge. The idea of estimating the chronological age from magnetic resonance images proved to be an interesting field of research due to the relative simplicity of its interpretation and its potential use as a biomarker of brain health. We revised our previous BrainAGE approach, originally utilising relevance vector regression (RVR), and substituted it with Gaussian process regression (GPR), which enables more stable processing of larger datasets, such as the UK Biobank (UKB). In addition, we extended the global BrainAGE approach to regional BrainAGE, providing spatially specific scores for five brain lobes per hemisphere. We tested the performance of the new algorithms under several different conditions and investigated their validity on the ADNI and schizophrenia samples, as well as on a synthetic dataset of neocortical thinning. The results show an improved performance of the reframed global model on the UKB sample with a mean absolute error (MAE) of less than 2 years and a significant difference in BrainAGE between healthy participants and patients with Alzheimer's disease and schizophrenia. Moreover, the workings of the algorithm show meaningful effects for a simulated neocortical atrophy dataset. The regional BrainAGE model performed well on two clinical samples, showing disease-specific patterns for different levels of impairment. The results demonstrate that the new improved algorithms provide reliable and valid brain age estimations.

Data-centric artificial olfactory system based on the eigengraph

Recent studies of electronic nose system tend to waste significant amount of important data in odor identification. Until now, the sensitivity-oriented data composition has made it difficult to discover meaningful data to apply artificial intelligence in terms of in-depth analysis for odor attributes specifying the identities of gas molecules, ultimately resulting in hindering the advancement of the artificial olfactory technology. Here, we realize a data-centric approach to implement standardized artificial olfactory systems inspired by human olfactory mechanisms by formally defining and utilizing the concept of Eigengraph in electrochemisty. The implicit odor attributes of the eigengraphs were mathematically substantialized as the Fourier transform-based Mel-Frequency Cepstral Coefficient feature vectors. Their effectiveness and applicability in deep learning processes for gas classification have been clearly demonstrated through experiments on complex mixed gases and automobile exhaust gases. We suggest that our findings can be widely applied as source technologies to develop standardized artificial olfactory systems.

Single-shot Echo Planar Time-resolved Imaging for multi-echo functional MRI and distortion-free diffusion imaging

Purpose

To develop EPTI, a multi-shot distortion-free multi-echo imaging technique, into a single-shot acquisition to achieve improved robustness to motion and physiological noise, increased temporal resolution, and high SNR efficiency for dynamic imaging applications.

Methods

A new spatiotemporal encoding was developed to achieve single-shot EPTI by enhancing spatiotemporal correlation in k-t space. The proposed single-shot encoding improves reconstruction conditioning and sampling efficiency, with additional optimization under various accelerations to achieve optimized performance. To achieve high SNR efficiency, continuous readout with minimized deadtime was employed that begins immediately after excitation and extends for an SNR-optimized length. Moreover, k-t partial Fourier and simultaneous multi-slice acquisition were integrated to further accelerate the acquisition and achieve high spatial and temporal resolution.

Results

We demonstrated that ss-EPTI achieves higher tSNR efficiency than multi-shot EPTI, and provides distortion-free imaging with densely-sampled multi-echo images at resolutions ~1.25-3 mm at 3T and 7T-with high SNR efficiency and with comparable temporal resolutions to ss-EPI. The ability of ss-EPTI to eliminate dynamic distortions common in EPI also further improves temporal stability. For fMRI, ss-EPTI also provides early-TE images (e.g., 2.9ms) to recover signal-intensity and functional-sensitivity dropout in challenging regions. The multi-echo images provide TE-dependent information about functional fluctuations, successfully distinguishing noise-components from BOLD signals and further improving tSNR. For diffusion MRI, ss-EPTI provides high-quality distortion-free diffusion images and multi-echo diffusion metrics.

Conclusion

ss-EPTI provides distortion-free imaging with high image quality, rich multi-echo information, and enhanced efficiency within comparable temporal resolution to ss-EPI, offering a robust and efficient acquisition for dynamic imaging.

Performance reserves in brain-imaging-based phenotype prediction

This study examines the impact of sample size on predicting cognitive and mental health phenotypes from brain imaging via machine learning. Our analysis shows a 3- to 9-fold improvement in prediction performance when sample size increases from 1,000 to 1 M participants. However, despite this increase, the data suggest that prediction accuracy remains worryingly low and far from fully exploiting the predictive potential of brain imaging data. Additionally, we find that integrating multiple imaging modalities boosts prediction accuracy, often equivalent to doubling the sample size. Interestingly, the most informative imaging modality often varied with increasing sample size, emphasizing the need to consider multiple modalities. Despite significant performance reserves for phenotype prediction, achieving substantial improvements may necessitate prohibitively large sample sizes, thus casting doubt on the practical or clinical utility of machine learning in some areas of neuroimaging.

Discovering individual fingerprints in resting-state functional connectivity using deep neural networks

Non-negligible idiosyncrasy due to interindividual differences is an ongoing issue in resting-state functional MRI (rfMRI) analysis. We show that a deep neural network (DNN) can be employed for individual identification by learning important features from the time-varying functional connectivity (FC) of rfMRI in the Human Connectome Project. We employed the trained DNN to identify individuals from an independent dataset acquired at our institution. The results revealed that the DNN could successfully identify 300 individuals with an error rate of 2.9% using 15 s time-window and 870 individuals with an error rate of 6.7%. A trained DNN with nonlinear hidden layers led to the proposal of the "fingerprint of FC" (fpFC) as representative edges of individual FC. The fpFCs for individuals exhibited commonly important and individual-specific edges across time-window lengths (from 5 min to 15 s). Furthermore, the utility of our model for another group of subjects was validated, supporting the feasibility of our technique in the context of transfer learning. In conclusion, our study offers an insight into the discovery of the intrinsic mode of the human brain using whole-brain resting-state FC and DNNs.

Decoding fMRI data with support vector machines and deep neural networks

BACKGROUND

Multivoxel pattern analysis (MVPA) examines fMRI activation patterns associated with different cognitive conditions. Support vector machines (SVMs) are the predominant method in MVPA. While SVM is intuitive and easy to apply, it is mainly suitable for analyzing data that are linearly separable. Convolutional neural networks (CNNs) are known to have the ability to approximate nonlinear relationships. Applications of CNN to fMRI data are beginning to appear with increasing frequency, but our understanding of the similarities and differences between CNN models and SVM models is limited.

NEW METHOD

We compared the two methods when they are applied to the same datasets. Two datasets were considered: (1) fMRI data collected from participants during a cued visual spatial attention task and (2) fMRI data collected from participants viewing natural images containing varying degrees of affective content.

RESULTS

We found that (1) both SVM and CNN are able to achieve above-chance decoding accuracies for attention control and emotion processing in both the primary visual cortex and the whole brain, (2) the CNN decoding accuracies are consistently higher than that of the SVM, (3) the SVM and CNN decoding accuracies are generally not correlated, and (4) the heatmaps derived from SVM and CNN are not significantly overlapping.

COMPARISON WITH EXISTING METHODS

By comparing SVM and CNN we pointed out the similarities and differences between the two methods.

CONCLUSIONS

SVM and CNN rely on different neural features for classification. Applying both to the same data may yield a more comprehensive understanding of neuroimaging data.

Development of Efficient Brain Age Estimation Method Based on Regional Brain Volume From Structural Magnetic Resonance Imaging

OBJECTIVE

We aimed to create an efficient and valid predicting model which can estimate individuals' brain age by quantifying their regional brain volumes.

METHODS

A total of 2,560 structural brain magnetic resonance imaging (MRI) scans, along with demographic and clinical data, were obtained. Pretrained deep-learning models were employed to automatically segment the MRI data, which enabled fast calculation of regional brain volumes. Brain age gaps for each subject were estimated using volumetric values from predefined 12 regions of interest (ROIs): bilateral frontal, parietal, occipital, and temporal lobes, as well as bilateral hippocampus and lateral ventricles. A larger weight was given to the ROIs having a larger mean volumetric difference between the cognitively unimpaired (CU) and cognitively impaired group including mild cognitive impairment (MCI), and dementia groups. The brain age was predicted by adding or subtracting the brain age gap to the chronological age according to the presence or absence of the atrophy region.

RESULTS

The study showed significant differences in brain age gaps among CU, MCI, and dementia groups. Furthermore, the brain age gaps exhibited significant correlations with education level and measures of cognitive function, including the clinical dementia rating sum-of-boxes and the Korean version of the Mini-Mental State Examination.

CONCLUSION

The brain age that we developed enabled fast and efficient brain age calculations, and it also reflected individual's cognitive function and cognitive reserve. Thus, our study suggested that the brain age might be an important marker of brain health that can be used effectively in real clinical settings.

Self-supervised multimodal learning for group inferences from MRI data: Discovering disorder-relevant brain regions and multimodal links

In recent years, deep learning approaches have gained significant attention in predicting brain disorders using neuroimaging data. However, conventional methods often rely on single-modality data and supervised models, which provide only a limited perspective of the intricacies of the highly complex brain. Moreover, the scarcity of accurate diagnostic labels in clinical settings hinders the applicability of the supervised models. To address these limitations, we propose a novel self-supervised framework for extracting multiple representations from multimodal neuroimaging data to enhance group inferences and enable analysis without resorting to labeled data during pre-training. Our approach leverages Deep InfoMax (DIM), a self-supervised methodology renowned for its efficacy in learning representations by estimating mutual information without the need for explicit labels. While DIM has shown promise in predicting brain disorders from single-modality MRI data, its potential for multimodal data remains untapped. This work extends DIM to multimodal neuroimaging data, allowing us to identify disorder-relevant brain regions and explore multimodal links. We present compelling evidence of the efficacy of our multimodal DIM analysis in uncovering disorder-relevant brain regions, including the hippocampus, caudate, insula, - and multimodal links with the thalamus, precuneus, and subthalamus hypothalamus. Our self-supervised representations demonstrate promising capabilities in predicting the presence of brain disorders across a spectrum of Alzheimer's phenotypes. Comparative evaluations against state-of-the-art unsupervised methods based on autoencoders, canonical correlation analysis, and supervised models highlight the superiority of our proposed method in achieving improved classification performance, capturing joint information, and interpretability capabilities. The computational efficiency of the decoder-free strategy enhances its practical utility, as it saves compute resources without compromising performance. This work offers a significant step forward in addressing the challenge of understanding multimodal links in complex brain disorders, with potential applications in neuroimaging research and clinical diagnosis.

Voxel-wise Fusion of Resting fMRI Networks and Gray Matter Volume for Alzheimer's Disease Classification using Deep Multimodal Learning

Alzheimer's disease (AD) is a prevalent neurodegenerative disorder requiring accurate and early diagnosis for effective treatment. Resting-state functional magnetic resonance imaging (rs-fMRI) and gray matter volume analysis from structural MRI have emerged as valuable tools for investigating AD-related brain alterations. However, the potential benefits of integrating these modalities using deep learning techniques remain unexplored. In this study, we propose a novel framework that fuses composite images of multiple rs-fMRI networks (called voxelwise intensity projection) and gray matter segmentation images through a deep learning approach for improved AD classification. We demonstrate the superiority of fMRI networks over commonly used metrics such as amplitude of low-frequency fluctuations (ALFF) and fractional ALFF in capturing spatial maps critical for AD classification. We use a multi-channel convolutional neural network incorporating the AlexNet dropout architecture to effectively model spatial and temporal dependencies in the integrated data. Extensive experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset of AD patients and cognitively normal (CN) validate the efficacy of our approach, showcasing improved classification performance of 94.12% test accuracy and an area under the curve (AUC) score of 97.79 compared to existing methods. Our results show that the fusion results generally outperformed the unimodal results. The saliency visualizations also show significant differences in the hippocampus, amygdala, putamen, caudate nucleus, and regions of basal ganglia which are in line with the previous neurobiological literature. Our research offers a novel method to enhance our grasp of AD pathology. By integrating data from various functional networks with structural MRI insights, we significantly improve diagnostic accuracy. This accuracy is further boosted by the effective visualization of this combined information. This lays the groundwork for further studies focused on providing a more accurate and personalized approach to AD diagnosis. The proposed framework and insights gained from fMRI networks provide a promising avenue for future research in deep multimodal fusion and neuroimaging analysis.

Gaussian Process-based prediction of memory performance and biomarker status in ageing and Alzheimer's disease-A systematic model evaluation

Neuroimaging markers based on Magnetic Resonance Imaging (MRI) combined with various other measures (such as genetic covariates, biomarkers, vascular risk factors, neuropsychological tests etc.) might provide useful predictions of clinical outcomes during the progression towards Alzheimer's disease (AD). The use of multiple features in predictive frameworks for clinical outcomes has become increasingly prevalent in AD research. However, many studies do not focus on systematically and accurately evaluating combinations of multiple input features. Hence, the aim of the present work is to explore and assess optimal combinations of various features for MR-based prediction of (1) cognitive status and (2) biomarker positivity with a multi-kernel learning Gaussian process framework. The explored features and parameters included (A) combinations of brain tissues, modulation, smoothing, and image resolution; (B) incorporating demographics & clinical covariates; (C) the impact of the size of the training data set; (D) the influence of dimensionality reduction and the choice of kernel types. The approach was tested in a large German cohort including 959 subjects from the multicentric longitudinal study of cognitive impairment and dementia (DELCODE). Our evaluation suggests the best prediction of memory performance was obtained for a combination of neuroimaging markers, demographics, genetic information (ApoE4) and CSF biomarkers explaining 57% of outcome variance in out-of-sample predictions. The highest performance for Aβ42/40 status classification was achieved for a combination of demographics, ApoE4, and a memory score while usage of structural MRI further improved the classification of individual patient's pTau status.

dcSBM: A federated constrained source-based morphometry approach for multivariate brain structure mapping

The examination of multivariate brain morphometry patterns has gained attention in recent years, especially for their powerful exploratory capabilities in the study of differences between patients and controls. Among the many existing methods and tools for the analysis of brain anatomy based on structural magnetic resonance imaging data, data-driven source-based morphometry (SBM) focuses on the exploratory detection of such patterns. Here, we implement a semi-blind extension of SBM, called constrained source-based morphometry (constrained SBM), which enables the extraction of maximally independent reference-alike sources using the constrained independent component analysis (ICA) approach. To do this, we combine SBM with a set of reference components covering the full brain, derived from a large independent data set (UKBiobank), to provide a fully automated SBM framework. This also allows us to implement a federated version of constrained SBM (cSBM) to allow analysis of data that is not locally accessible. In our proposed decentralized constrained source-based morphometry (dcSBM), the original data never leaves the local site. Each site operates constrained ICA on its private local data using a common distributed computation platform. Next, an aggregator/master node aggregates the results estimated from each local site and applies statistical analysis to estimate the significance of the sources. Finally, we utilize two additional multisite patient data sets to validate our model by comparing the resulting group difference estimates from both cSBM and dcSBM.

3DCNN predicting brain age using diffusion tensor imaging

Neuroimaging-based brain age prediction using deep learning is gaining popularity. However, few studies have attempted to leverage diffusion tensor imaging (DTI) to predict brain age. In this study, we proposed a 3D convolutional neural network model (3DCNN) and trained it on fractional anisotropy (FA) data from six publicly available datasets (n = 2406, age = 17-60) to estimate brain age. Implementing a two-loop nested cross-validation scheme with a tenfold cross-validation procedure, we achieved a robust prediction performance of a mean absolute error (MAE) of 2.785 and a correlation coefficient of 0.932. We also employed Grad-Cam++ to visualize the salient features of the proposed model. We identified a few highly salient fiber tracts, including the genu of corpus callosum and the left cerebellar peduncle, among others that play a pivotal role in our model. In sum, our model reliably predicted brain age and provided novel insight into age-related changes in brains' axonal structure.

Predicting the efficacy of non-steroidal anti-inflammatory drugs in migraine using deep learning and three-dimensional T1-weighted images

Deep learning (DL) models based on individual images could contribute to tailored therapies and personalized treatment strategies. We aimed to construct a DL model using individual 3D structural images for predicting the efficacy of non-steroidal anti-inflammatory drugs (NSAIDs) in migraine. A 3D convolutional neural network model was constructed, with ResNet18 as the classification backbone, to link structural images to predict the efficacy of NSAIDs. In total, 111 patients were included and allocated to the training and testing sets in a 4:1 ratio. The prediction accuracies of the ResNet34, ResNet50, ResNeXt50, DenseNet121, and 3D ResNet18 models were 0.65, 0.74, 0.65, 0.70, and 0.78, respectively. This model, based on individual 3D structural images, demonstrated better predictive performance in comparison to conventional models. Our study highlights the feasibility of the DL algorithm based on brain structural images and suggests that it can be applied to predict the efficacy of NSAIDs in migraine treatment.

Human-to-monkey transfer learning identifies the frontal white matter as a key determinant for predicting monkey brain age

The application of artificial intelligence (AI) to summarize a whole-brain magnetic resonance image (MRI) into an effective "brain age" metric can provide a holistic, individualized, and objective view of how the brain interacts with various factors (e.g., genetics and lifestyle) during aging. Brain age predictions using deep learning (DL) have been widely used to quantify the developmental status of human brains, but their wider application to serve biomedical purposes is under criticism for requiring large samples and complicated interpretability. Animal models, i.e., rhesus monkeys, have offered a unique lens to understand the human brain - being a species in which aging patterns are similar, for which environmental and lifestyle factors are more readily controlled. However, applying DL methods in animal models suffers from data insufficiency as the availability of animal brain MRIs is limited compared to many thousands of human MRIs. We showed that transfer learning can mitigate the sample size problem, where transferring the pre-trained AI models from 8,859 human brain MRIs improved monkey brain age estimation accuracy and stability. The highest accuracy and stability occurred when transferring the 3D ResNet [mean absolute error (MAE) = 1.83 years] and the 2D global-local transformer (MAE = 1.92 years) models. Our models identified the frontal white matter as the most important feature for monkey brain age predictions, which is consistent with previous histological findings. This first DL-based, anatomically interpretable, and adaptive brain age estimator could broaden the application of AI techniques to various animal or disease samples and widen opportunities for research in non-human primate brains across the lifespan.

A Siamese Network With Node Convolution for Individualized Predictions Based on Connectivity Maps Extracted From Resting-State fMRI Data

Deep learning has demonstrated great potential for objective diagnosis of neuropsychiatric disorders based on neuroimaging data, which includes the promising resting-state functional magnetic resonance imaging (RS-fMRI). However, the insufficient sample size has long been a bottleneck for deep model training for the purpose. In this study, we proposed a Siamese network with node convolution (SNNC) for individualized predictions based on RS-fMRI data. With the involvement of Siamese network, which uses sample pair (rather than a single sample) as input, the problem of insufficient sample size can largely be alleviated. To adapt to connectivity maps extracted from RS-fMRI data, we applied node convolution to each of the two branches of the Siamese network. For regression purposes, we replaced the contrastive loss in classic Siamese network with the mean square error loss and thus enabled Siamese network to quantitatively predict label differences. The label of a test sample can be predicted based on any of the training samples, by adding the label of the training sample to the predicted label difference between them. The final prediction for a test sample in this study was made by averaging the predictions based on each of the training samples. The performance of the proposed SNNC was evaluated with age and IQ predictions based on a public dataset (Cam-CAN). The results indicated that SNNC can make effective predictions even with a sample size of as small as 40, and SNNC achieved state-of-the-art accuracy among a variety of deep models and standard machine learning approaches.

Ideography insight from facial recognition and neuroimaging

One novel example and/or perspective in support of "Why the learning account fails" is the impressive ability of humans to recognize and memorize facial features and accurately and reliably connect those to related identities. Furthermore, neuroimaging analysis presents an example in support of the crucial role of standardization in the lack of adoption of ideography.

Real-world prediction of preclinical Alzheimer's disease with a deep generative model

Amyloid positivity is an early indicator of Alzheimer's disease and is necessary to determine the disease. In this study, a deep generative model is utilized to predict the amyloid positivity of cognitively normal individuals using proxy measures, such as structural MRI scans, demographic variables, and cognitive scores, instead of invasive direct measurements. Through its remarkable efficacy in handling imperfect datasets caused by missing data or labels, and imbalanced classes, the model outperforms previous studies and widely used machine learning approaches with an AUROC of 0.8609. Furthermore, this study illuminates the model's adaptability to diverse clinical scenarios, even when feature sets or diagnostic criteria differ from the training data. We identify the brain regions and variables that contribute most to classification, including the lateral occipital lobes, posterior temporal lobe, and APOE ϵ4 allele. Taking advantage of deep generative models, our approach can not only provide inexpensive, non-invasive, and accurate diagnostics for preclinical Alzheimer's disease, but also meet real-world requirements for clinical translation of a deep learning model, including transferability and interpretability.

Accounting for temporal variability in functional magnetic resonance imaging improves prediction of intelligence

Neuroimaging-based prediction methods for intelligence have seen a rapid development. Among different neuroimaging modalities, prediction using functional connectivity (FC) has shown great promise. Most literature has focused on prediction using static FC, with limited investigations on the merits of such analysis compared to prediction using dynamic FC or region-level functional magnetic resonance imaging (fMRI) times series that encode temporal variability. To account for the temporal dynamics in fMRI, we propose a bi-directional long short-term memory (bi-LSTM) approach that incorporates feature selection mechanism. The proposed pipeline is implemented via an efficient algorithm and applied for predicting intelligence using region-level time series and dynamic FC. We compare the prediction performance using different fMRI features acquired from the Adolescent Brain Cognitive Development (ABCD) study involving nearly 7000 individuals. Our detailed analysis illustrates the consistently inferior performance of static FC compared to region-level time series or dynamic FC for single and combined rest and task fMRI experiments. The joint analysis of task and rest fMRI leads to improved intelligence prediction under all models compared to using fMRI from only one experiment. In addition, the proposed bi-LSTM pipeline based on region-level time series identifies several shared and differential important brain regions across fMRI experiments that drive intelligence prediction. A test-retest analysis of the selected regions shows strong reliability across cross-validation folds. Given the large sample size of ABCD study, our results provide strong evidence that superior prediction of intelligence can be achieved by accounting for temporal variations in fMRI.

Replicable brain-phenotype associations require large-scale neuroimaging data

Numerous neuroimaging studies have investigated the neural basis of interindividual differences but the replicability of brain-phenotype associations remains largely unknown. We used the UK Biobank neuroimaging dataset (N = 37,447) to examine associations with six variables related to physical and mental health: age, body mass index, intelligence, memory, neuroticism and alcohol consumption, and assessed the improvement of replicability for brain-phenotype associations with increasing sampling sizes. Age may require only 300 individuals to provide highly replicable associations but other phenotypes required 1,500 to 3,900 individuals. The required sample size showed a negative power law relation with the estimated effect size. When only comparing the upper and lower quarters, the minimally required sample sizes for imaging decreased by 15-75%. Our findings demonstrate that large-scale neuroimaging data are required for replicable brain-phenotype associations, that this can be mitigated by preselection of individuals and that small-scale studies may have reported false positive findings.

Automatic brain MRI motion artifact detection based on end-to-end deep learning is similarly effective as traditional machine learning trained on image quality metrics

Head motion artifacts in magnetic resonance imaging (MRI) are an important confounding factor concerning brain research as well as clinical practice. For this reason, several machine learning-based methods have been developed for the automatic quality control of structural MRI scans. Deep learning offers a promising solution to this problem, however, given its data-hungry nature and the scarcity of expert-annotated datasets, its advantage over traditional machine learning methods in identifying motion-corrupted brain scans is yet to be determined. In the present study, we investigated the relative advantage of the two methods in structural MRI quality control. To this end, we collected publicly available T1-weighted images and scanned subjects in our own lab under conventional and active head motion conditions. The quality of the images was rated by a team of radiologists from the point of view of clinical diagnostic use. We present a relatively simple, lightweight 3D convolutional neural network trained in an end-to-end manner that achieved a test set (N = 411) balanced accuracy of 94.41% in classifying brain scans into clinically usable or unusable categories. A support vector machine trained on image quality metrics achieved a balanced accuracy of 88.44% on the same test set. Statistical comparison of the two models yielded no significant difference in terms of confusion matrices, error rates, or receiver operating characteristic curves. Our results suggest that these machine learning methods are similarly effective in identifying severe motion artifacts in brain MRI scans, and underline the efficacy of end-to-end deep learning-based systems in brain MRI quality control, allowing the rapid evaluation of diagnostic utility without the need for elaborate image pre-processing.

The future of diagnosis in clinical neurosciences: Comparing multiple sclerosis and schizophrenia

The ongoing developments of psychiatric classification systems have largely improved reliability of diagnosis, including that of schizophrenia. However, with an unknown pathophysiology and lacking biomarkers, its validity still remains low, requiring further advancements. Research has helped establish multiple sclerosis (MS) as the central nervous system (CNS) disorder with an established pathophysiology, defined biomarkers and therefore good validity and significantly improved treatment options. Before proposing next steps in research that aim to improve the diagnostic process of schizophrenia, it is imperative to recognize its clinical heterogeneity. Indeed, individuals with schizophrenia show high interindividual variability in terms of symptomatic manifestation, response to treatment, course of illness and functional outcomes. There is also a multiplicity of risk factors that contribute to the development of schizophrenia. Moreover, accumulating evidence indicates that several dimensions of psychopathology and risk factors cross current diagnostic categorizations. Schizophrenia shares a number of similarities with MS, which is a demyelinating disease of the CNS. These similarities appear in the context of age of onset, geographical distribution, involvement of immune-inflammatory processes, neurocognitive impairment and various trajectories of illness course. This article provides a critical appraisal of diagnostic process in schizophrenia, taking into consideration advancements that have been made in the diagnosis and management of MS. Based on the comparison between the two disorders, key directions for studies that aim to improve diagnostic process in schizophrenia are formulated. All of them converge on the necessity to deconstruct the psychosis spectrum and adopt dimensional approaches with deep phenotyping to refine current diagnostic boundaries.

Identifying healthy individuals with Alzheimer's disease neuroimaging phenotypes in the UK Biobank

BACKGROUND

Identifying prediagnostic neurodegenerative disease is a critical issue in neurodegenerative disease research, and Alzheimer's disease (AD) in particular, to identify populations suitable for preventive and early disease-modifying trials. Evidence from genetic and other studies suggests the neurodegeneration of Alzheimer's disease measured by brain atrophy starts many years before diagnosis, but it is unclear whether these changes can be used to reliably detect prediagnostic sporadic disease.

METHODS

We trained a Bayesian machine learning neural network model to generate a neuroimaging phenotype and AD score representing the probability of AD using structural MRI data in the Alzheimer's Disease Neuroimaging Initiative (ADNI) Cohort (cut-off 0.5, AUC 0.92, PPV 0.90, NPV 0.93). We go on to validate the model in an independent real-world dataset of the National Alzheimer's Coordinating Centre (AUC 0.74, PPV 0.65, NPV 0.80) and demonstrate the correlation of the AD-score with cognitive scores in those with an AD-score above 0.5. We then apply the model to a healthy population in the UK Biobank study to identify a cohort at risk for Alzheimer's disease.

RESULTS

We show that the cohort with a neuroimaging Alzheimer's phenotype has a cognitive profile in keeping with Alzheimer's disease, with strong evidence for poorer fluid intelligence, and some evidence of poorer numeric memory, reaction time, working memory, and prospective memory. We found some evidence in the AD-score positive cohort for modifiable risk factors of hypertension and smoking.

CONCLUSIONS

This approach demonstrates the feasibility of using AI methods to identify a potentially prediagnostic population at high risk for developing sporadic Alzheimer's disease.

Interpretable LSTM model reveals transiently-realized patterns of dynamic brain connectivity that predict patient deterioration or recovery from very mild cognitive impairment

Alzheimer's Disease (AZD) is a neurodegenerative disease for which there is now no known effective treatment. Mild cognitive impairment (MCI) is considered a precursor to AZD and affects cognitive abilities. Patients with MCI have the potential to recover cognitive health, can remain mildly cognitively impaired indefinitely or eventually progress to AZD. Identifying imaging-based predictive biomarkers for disease progression in patients presenting with evidence of very mild/questionable MCI (qMCI) can play an important role in triggering early dementia intervention. Dynamic functional network connectivity (dFNC) estimated from resting-state functional magnetic resonance imaging (rs-fMRI) has been increasingly studied in brain disorder diseases. In this work, employing a recent developed a time-attention long short-term memory (TA-LSTM) network to classify multivariate time series data. A gradient-based interpretation framework, transiently-realized event classifier activation map (TEAM) is introduced to localize the group-defining "activated" time intervals over the full time series and generate the class difference map. To test the trustworthiness of TEAM, we did a simulation study to validate the model interpretative power of TEAM. We then applied this simulation-validated framework to a well-trained TA-LSTM model which predicts the progression or recovery from questionable/mild cognitive impairment (qMCI) subjects after three years from windowless wavelet-based dFNC (WWdFNC). The FNC class difference map points to potentially important predictive dynamic biomarkers. Moreover, the more highly time-solved dFNC (WWdFNC) achieves better performance in both TA-LSTM and a multivariate CNN model than dFNC based on windowed correlations between timeseries, suggesting that better temporally resolved measures can enhance the model's performance.

Brain age prediction using the graph neural network based on resting-state functional MRI in Alzheimer's disease

Introduction

Alzheimer's disease (AD) is a neurodegenerative disease that significantly impacts the quality of life of patients and their families. Neuroimaging-driven brain age prediction has been proposed as a potential biomarker to detect mental disorders, such as AD, aiding in studying its effects on functional brain networks. Previous studies have shown that individuals with AD display impaired resting-state functional connections. However, most studies on brain age prediction have used structural magnetic resonance imaging (MRI), with limited studies based on resting-state functional MRI (rs-fMRI).

Methods

In this study, we applied a graph neural network (GNN) model on controls to predict brain ages using rs-fMRI in patients with AD. We compared the performance of the GNN model with traditional machine learning models. Finally, the post hoc model was also used to identify the critical brain regions in AD.

Results

The experimental results demonstrate that our GNN model can predict brain ages of normal controls using rs-fMRI data from the ADNI database. Moreover the differences between brain ages and chronological ages were more significant in AD patients than in normal controls. Our results also suggest that AD is associated with accelerated brain aging and that the GNN model based on resting-state functional connectivity is an effective tool for predicting brain age.

Discussion

Our study provides evidence that rs-fMRI is a promising modality for brain age prediction in AD research, and the GNN model proves to be effective in predicting brain age. Furthermore, the effects of the hippocampus, parahippocampal gyrus, and amygdala on brain age prediction are verified.

Deep-learning-enabled brain hemodynamic mapping using resting-state fMRI

Cerebrovascular disease is a leading cause of death globally. Prevention and early intervention are known to be the most effective forms of its management. Non-invasive imaging methods hold great promises for early stratification, but at present lack the sensitivity for personalized prognosis. Resting-state functional magnetic resonance imaging (rs-fMRI), a powerful tool previously used for mapping neural activity, is available in most hospitals. Here we show that rs-fMRI can be used to map cerebral hemodynamic function and delineate impairment. By exploiting time variations in breathing pattern during rs-fMRI, deep learning enables reproducible mapping of cerebrovascular reactivity (CVR) and bolus arrival time (BAT) of the human brain using resting-state CO2 fluctuations as a natural "contrast media". The deep-learning network is trained with CVR and BAT maps obtained with a reference method of CO2-inhalation MRI, which includes data from young and older healthy subjects and patients with Moyamoya disease and brain tumors. We demonstrate the performance of deep-learning cerebrovascular mapping in the detection of vascular abnormalities, evaluation of revascularization effects, and vascular alterations in normal aging. In addition, cerebrovascular maps obtained with the proposed method exhibit excellent reproducibility in both healthy volunteers and stroke patients. Deep-learning resting-state vascular imaging has the potential to become a useful tool in clinical cerebrovascular imaging.