Fully Connected Cascade Artificial Neural Network Architecture for Attention Deficit Hyperactivity Disorder Classification From Functional Magnetic Resonance Imaging Data

A review of ADHD detection studies with machine learning methods using rsfMRI data

Attention deficit hyperactivity disorder (ADHD) is a common mental health condition that significantly affects school-age children, causing difficulties with learning and daily functioning. Early identification is crucial, and reliable and objective diagnostic tools are necessary. However, current clinical evaluations of behavioral symptoms can be inconsistent and subjective. Functional magnetic resonance imaging (fMRI) is a non-invasive technique that has proven effective in detecting brain abnormalities in individuals with ADHD. Recent studies have shown promising outcomes in using resting state fMRI (rsfMRI)-based brain functional networks to diagnose various brain disorders, including ADHD. Several review papers have examined the detection of other diseases using fMRI data and machine learning or deep learning methods. However, no review paper has specifically addressed ADHD. Therefore, this study aims to contribute to the literature by reviewing the use of rsfMRI data and machine learning methods for detection of ADHD. The study provides general information about fMRI databases and detailed knowledge of the ADHD-200 database, which is commonly used for ADHD detection. It also emphasizes the importance of examining all stages of the process, including network and atlas selection, feature extraction, and feature selection, before the classification stage. The study compares the performance, advantages, and disadvantages of previous studies in detail. This comprehensive approach may be a useful starting point for new researchers in this area.

Diagnosis of mental disorders using machine learning: Literature review and bibliometric mapping from 2012 to 2023

Background

Mental disorders (MDs) are becoming a leading burden in non-communicable diseases (NCDs). As per the World Health Organization's 2022 assessment report, there was a steep increase of 25 % in MDs during the COVID-19 pandemic. Early diagnosis of MDs can significantly improve treatment outcome and save disability-adjusted life years (DALYs). In recent times, the application of machine learning (ML) and deep learning (DL)) has shown promising results in the diagnosis of MDs, and the field has witnessed a huge research output in the form of research publications. Therefore, a bibliometric mapping along with a review of recent advancements is required.

Methods

This study presents a bibliometric analysis and review of the research, published over the last 10 years. Literature searches were conducted in the Scopus database for the period from January 1, 2012, to June 9, 2023. The data was filtered and screened to include only relevant and reliable publications. A total of 2811 journal articles were found. The data was exported to a comma-separated value (CSV) format for further analysis. Furthermore, a review of 40 selected studies was performed.

Results

The popularity of ML techniques in diagnosing MDs has been growing, with an annual research growth rate of 17.05 %. The Journal of Affective Disorders published the most documents (n = 97), while Wang Y. (n = 64) has published the most articles. Lotka's law is observed, with a minority of authors contributing the majority of publications. The top affiliating institutes are the West China Hospital of Sichuan University followed by the University of California, with China and the US dominating the top 10 institutes. While China has more publications, papers affiliated with the US receive more citations. Depression and schizophrenia are the primary focuses of ML and deep learning (DL) in mental disease detection. Co-occurrence network analysis reveals that ML is associated with depression, schizophrenia, autism, anxiety, ADHD, obsessive-compulsive disorder, and PTSD. Popular algorithms include support vector machine (SVM) classifier, decision tree classifier, and random forest classifier. Furthermore, DL is linked to neuroimaging techniques such as MRI, fMRI, and EEG, as well as bipolar disorder. Current research trends encompass DL, LSTM, generalized anxiety disorder, feature fusion, and convolutional neural networks.

Structural or/and functional MRI-based machine learning techniques for attention-deficit/hyperactivity disorder diagnosis: A systematic review and meta-analysis

BACKGROUND

The aim of this study was to investigate the diagnostic value of ML techniques based on sMRI or/and fMRI for ADHD.

METHODS

We conducted a comprehensive search (from database creation date to March 2024) for relevant English articles on sMRI or/and fMRI-based ML techniques for diagnosing ADHD. The pooled sensitivity, specificity, positive likelihood ratio (LR+), negative likelihood ratio (LR-), summary receiver operating characteristic (SROC) curve and area under the curve (AUC) were calculated to assess the diagnostic value of sMRI or/and fMRI-based ML techniques. The I2 test was used to assess heterogeneity and the source of heterogeneity was investigated by performing a meta-regression analysis. Publication bias was assessed using the Deeks funnel plot asymmetry test.

RESULTS

Forty-three studies were included in the systematic review, 27 of which were included in our meta-analysis. The pooled sensitivity and specificity of sMRI or/and fMRI-based ML techniques for the diagnosis of ADHD were 0.74 (95 % CI 0.65-0.81) and 0.75 (95 % CI 0.67-0.81), respectively. SROC curve showed that AUC was 0.81 (95 % CI 0.77-0.84). Based on these findings, the sMRI or/and fMRI-based ML techniques have relatively good diagnostic value for ADHD.

LIMITATIONS

Our meta-analysis specifically focused on ML techniques based on sMRI or/and fMRI studies. Since EEG-based ML techniques are also used for diagnosing ADHD, further systematic analyses are necessary to explore ML methods based on multimodal medical data.

CONCLUSION

sMRI or/and fMRI-based ML technique is a promising objective diagnostic method for ADHD.

Bayesian Estimation of Inverted Beta Mixture Models With Extended Stochastic Variational Inference for Positive Vector Classification

The finite inverted beta mixture model (IBMM) has been proven to be efficient in modeling positive vectors. Under the traditional variational inference framework, the critical challenge in Bayesian estimation of the IBMM is that the computational cost of performing inference with large datasets is prohibitively expensive, which often limits the use of Bayesian approaches to small datasets. An efficient alternative provided by the recently proposed stochastic variational inference (SVI) framework allows for efficient inference on large datasets. Nevertheless, when using the SVI framework to address the non-Gaussian statistical models, the evidence lower bound (ELBO) cannot be explicitly calculated due to the intractable moment computation. Therefore, the algorithm under the SVI framework cannot directly use stochastic optimization to optimize the ELBO, and an analytically tractable solution cannot be derived. To address this problem, we propose an extended version of the SVI framework with more flexibility, namely, the extended SVI (ESVI) framework. This framework can be used in many non-Gaussian statistical models. First, some approximation strategies are applied to further lower the ELBO to avoid intractable moment calculations. Then, stochastic optimization with noisy natural gradients is used to optimize the lower bound. The excellent performance and effectiveness of the proposed method are verified in real data evaluation.

A review of the applications of generative adversarial networks to structural and functional MRI based diagnostic classification of brain disorders

Structural and functional MRI (magnetic resonance imaging) based diagnostic classification using machine learning has long held promise, but there are many roadblocks to achieving their potential. While traditional machine learning models suffered from their inability to capture the complex non-linear mapping, deep learning models tend to overfit the model. This is because there is data scarcity and imbalanced classes in neuroimaging; it is expensive to acquire data from human subjects and even more so in clinical populations. Due to their ability to augment data by learning underlying distributions, generative adversarial networks (GAN) provide a potential solution to this problem. Here, we provide a methodological primer on GANs and review the applications of GANs to classification of mental health disorders from neuroimaging data such as functional MRI and showcase the progress made thus far. We also highlight gaps in methodology as well as interpretability that are yet to be addressed. This provides directions about how the field can move forward. We suggest that since there are a range of methodological choices available to users, it is critical for users to interact with method developers so that the latter can tailor their development according to the users' needs. The field can be enriched by such synthesis between method developers and users in neuroimaging.

Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry

BACKGROUND

The development of machine learning models for aiding in the diagnosis of mental disorder is recognized as a significant breakthrough in the field of psychiatry. However, clinical practice of such models remains a challenge, with poor generalizability being a major limitation.

METHODS

Here, we conducted a pre-registered meta-research assessment on neuroimaging-based models in the psychiatric literature, quantitatively examining global and regional sampling issues over recent decades, from a view that has been relatively underexplored. A total of 476 studies (n = 118,137) were included in the current assessment. Based on these findings, we built a comprehensive 5-star rating system to quantitatively evaluate the quality of existing machine learning models for psychiatric diagnoses.

RESULTS

A global sampling inequality in these models was revealed quantitatively (sampling Gini coefficient (G) = 0.81, p < .01), varying across different countries (regions) (e.g., China, G = 0.47; the USA, G = 0.58; Germany, G = 0.78; the UK, G = 0.87). Furthermore, the severity of this sampling inequality was significantly predicted by national economic levels (β =  - 2.75, p < .001, R²_adj = 0.40; r =  - .84, 95% CI: - .41 to - .97), and was plausibly predictable for model performance, with higher sampling inequality for reporting higher classification accuracy. Further analyses showed that lack of independent testing (84.24% of models, 95% CI: 81.0-87.5%), improper cross-validation (51.68% of models, 95% CI: 47.2-56.2%), and poor technical transparency (87.8% of models, 95% CI: 84.9-90.8%)/availability (80.88% of models, 95% CI: 77.3-84.4%) are prevailing in current diagnostic classifiers despite improvements over time. Relating to these observations, model performances were found decreased in studies with independent cross-country sampling validations (all p < .001, BF₁₀ > 15). In light of this, we proposed a purpose-built quantitative assessment checklist, which demonstrated that the overall ratings of these models increased by publication year but were negatively associated with model performance.

CONCLUSIONS

Together, improving sampling economic equality and hence the quality of machine learning models may be a crucial facet to plausibly translating neuroimaging-based diagnostic classifiers into clinical practice.

Bibliometric Analysis of Granger Causality Studies

Granger causality provides a framework that uses predictability to identify causation between time series variables. This is important to policymakers for effective policy management and recommendations. Granger causality is recognized as the primary advance on the causation problem. The objective of this paper is to conduct a bibliometric analysis of Granger causality publications indexed in the Web of Science database. Harzing's Publish or Perish and VOSviewer were used for performance analysis and science mapping. The first paper indexed was published in 1981 and there has been an upward trend in the annual publication of Granger causality studies which are shifting towards the areas of environmental science, energy, and economics. Most of the publications are articles and proceeding papers under the areas of business economics, environmental science ecology, and neurosciences/neurology. China has the highest number of publications while the United States has the highest number of citations. England has the highest citation impact. This paper also constructed country co-authorship, co-analysis of cited references, cited sources, and cited authors, keyword co-occurrence, and keyword overlay visualization maps.

Evaluation of Risk of Bias in Neuroimaging-Based Artificial Intelligence Models for Psychiatric Diagnosis: A Systematic Review

IMPORTANCE

Neuroimaging-based artificial intelligence (AI) diagnostic models have proliferated in psychiatry. However, their clinical applicability and reporting quality (ie, feasibility) for clinical practice have not been systematically evaluated.

OBJECTIVE

To systematically assess the risk of bias (ROB) and reporting quality of neuroimaging-based AI models for psychiatric diagnosis.

EVIDENCE REVIEW

PubMed was searched for peer-reviewed, full-length articles published between January 1, 1990, and March 16, 2022. Studies aimed at developing or validating neuroimaging-based AI models for clinical diagnosis of psychiatric disorders were included. Reference lists were further searched for suitable original studies. Data extraction followed the CHARMS (Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies) and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines. A closed-loop cross-sequential design was used for quality control. The PROBAST (Prediction Model Risk of Bias Assessment Tool) and modified CLEAR (Checklist for Evaluation of Image-Based Artificial Intelligence Reports) benchmarks were used to systematically evaluate ROB and reporting quality.

FINDINGS

A total of 517 studies presenting 555 AI models were included and evaluated. Of these models, 461 (83.1%; 95% CI, 80.0%-86.2%) were rated as having a high overall ROB based on the PROBAST. The ROB was particular high in the analysis domain, including inadequate sample size (398 of 555 models [71.7%; 95% CI, 68.0%-75.6%]), poor model performance examination (with 100% of models lacking calibration examination), and lack of handling data complexity (550 of 555 models [99.1%; 95% CI, 98.3%-99.9%]). None of the AI models was perceived to be applicable to clinical practices. Overall reporting completeness (ie, number of reported items/number of total items) for the AI models was 61.2% (95% CI, 60.6%-61.8%), and the completeness was poorest for the technical assessment domain with 39.9% (95% CI, 38.8%-41.1%).

CONCLUSIONS AND RELEVANCE

This systematic review found that the clinical applicability and feasibility of neuroimaging-based AI models for psychiatric diagnosis were challenged by a high ROB and poor reporting quality. Particularly in the analysis domain, ROB in AI diagnostic models should be addressed before clinical application.

Population level multimodal neuroimaging correlates of attention-deficit hyperactivity disorder among children

Objectives

Leveraging a large population-level morphologic, microstructural, and functional neuroimaging dataset, we aimed to elucidate the underlying neurobiology of attention-deficit hyperactivity disorder (ADHD) in children. In addition, we evaluated the applicability of machine learning classifiers to predict ADHD diagnosis based on imaging and clinical information.

Methods

From the Adolescents Behavior Cognitive Development (ABCD) database, we included 1,798 children with ADHD diagnosis and 6,007 without ADHD. In multivariate logistic regression adjusted for age and sex, we examined the association of ADHD with different neuroimaging metrics. The neuroimaging metrics included fractional anisotropy (FA), neurite density (ND), mean-(MD), radial-(RD), and axial diffusivity (AD) of white matter (WM) tracts, cortical region thickness and surface areas from T1-MPRAGE series, and functional network connectivity correlations from resting-state fMRI.

Results

Children with ADHD showed markers of pervasive reduced microstructural integrity in white matter (WM) with diminished neural density and fiber-tracks volumes - most notable in the frontal and parietal lobes. In addition, ADHD diagnosis was associated with reduced cortical volume and surface area, especially in the temporal and frontal regions. In functional MRI studies, ADHD children had reduced connectivity among default-mode network and the central and dorsal attention networks, which are implicated in concentration and attention function. The best performing combination of feature selection and machine learning classifier could achieve a receiver operating characteristics area under curve of 0.613 (95% confidence interval = 0.580-0.645) to predict ADHD diagnosis in independent validation, using a combination of multimodal imaging metrics and clinical variables.

Conclusion

Our study highlights the neurobiological implication of frontal lobe cortex and associate WM tracts in pathogenesis of childhood ADHD. We also demonstrated possible potentials and limitations of machine learning models to assist with ADHD diagnosis in a general population cohort based on multimodal neuroimaging metrics.

Machine Learning and MRI-based Diagnostic Models for ADHD: Are We There Yet?

OBJECTIVE

Machine learning (ML) has been applied to develop magnetic resonance imaging (MRI)-based diagnostic classifiers for attention-deficit/hyperactivity disorder (ADHD). This systematic review examines this literature to clarify its clinical significance and to assess the implications of the various analytic methods applied.

METHODS

A comprehensive literature search on MRI-based diagnostic classifiers for ADHD was performed and data regarding the utilized models and samples were gathered.

RESULTS

We found that, although most studies reported the classification accuracies, they varied in choice of MRI modalities, ML models, cross-validation and testing methods, and sample sizes. We found that the accuracies of cross-validation methods inflated the performance estimation compared with those of a held-out test, compromising the model generalizability. Test accuracies have increased with publication year but were not associated with training sample sizes. Improved test accuracy over time was likely due to the use of better ML methods along with strategies to deal with data imbalances.

CONCLUSION

Ultimately, large multi-modal imaging datasets, and potentially the combination with other types of data, like cognitive data and/or genetics, will be essential to achieve the goal of developing clinically useful imaging classification tools for ADHD in the future.

Deep learning in neuroimaging data analysis: Applications, challenges, and solutions

Methods for the analysis of neuroimaging data have advanced significantly since the beginning of neuroscience as a scientific discipline. Today, sophisticated statistical procedures allow us to examine complex multivariate patterns, however most of them are still constrained by assuming inherent linearity of neural processes. Here, we discuss a group of machine learning methods, called deep learning, which have drawn much attention in and outside the field of neuroscience in recent years and hold the potential to surpass the mentioned limitations. Firstly, we describe and explain the essential concepts in deep learning: the structure and the computational operations that allow deep models to learn. After that, we move to the most common applications of deep learning in neuroimaging data analysis: prediction of outcome, interpretation of internal representations, generation of synthetic data and segmentation. In the next section we present issues that deep learning poses, which concerns multidimensionality and multimodality of data, overfitting and computational cost, and propose possible solutions. Lastly, we discuss the current reach of DL usage in all the common applications in neuroimaging data analysis, where we consider the promise of multimodality, capability of processing raw data, and advanced visualization strategies. We identify research gaps, such as focusing on a limited number of criterion variables and the lack of a well-defined strategy for choosing architecture and hyperparameters. Furthermore, we talk about the possibility of conducting research with constructs that have been ignored so far or/and moving toward frameworks, such as RDoC, the potential of transfer learning and generation of synthetic data.

SSPNet: An interpretable 3D-CNN for classification of schizophrenia using phase maps of resting-state complex-valued fMRI data

Convolutional neural networks (CNNs) have shown promising results in classifying individuals with mental disorders such as schizophrenia using resting-state fMRI data. However, complex-valued fMRI data is rarely used since additional phase data introduces high-level noise though it is potentially useful information for the context of classification. As such, we propose to use spatial source phase (SSP) maps derived from complex-valued fMRI data as the CNN input. The SSP maps are not only less noisy, but also more sensitive to spatial activation changes caused by mental disorders than magnitude maps. We build a 3D-CNN framework with two convolutional layers (named SSPNet) to fully explore the 3D structure and voxel-level relationships from the SSP maps. Two interpretability modules, consisting of saliency map generation and gradient-weighted class activation mapping (Grad-CAM), are incorporated into the well-trained SSPNet to provide additional information helpful for understanding the output. Experimental results from classifying schizophrenia patients (SZs) and healthy controls (HCs) show that the proposed SSPNet significantly improved accuracy and AUC compared to CNN using magnitude maps extracted from either magnitude-only (by 23.4 and 23.6% for DMN) or complex-valued fMRI data (by 10.6 and 5.8% for DMN). SSPNet captured more prominent HC-SZ differences in saliency maps, and Grad-CAM localized all contributing brain regions with opposite strengths for HCs and SZs within SSP maps. These results indicate the potential of SSPNet as a sensitive tool that may be useful for the development of brain-based biomarkers of mental disorders.

Multikernel Capsule Network for Schizophrenia Identification

Schizophrenia seriously affects the quality of life. To date, both simple (e.g., linear discriminant analysis) and complex (e.g., deep neural network) machine-learning methods have been utilized to identify schizophrenia based on functional connectivity features. The existing simple methods need two separate steps (i.e., feature extraction and classification) to achieve the identification, which disables simultaneous tuning for the best feature extraction and classifier training. The complex methods integrate two steps and can be simultaneously tuned to achieve optimal performance, but these methods require a much larger amount of data for model training. To overcome the aforementioned drawbacks, we proposed a multikernel capsule network (MKCapsnet), which was developed by considering the brain anatomical structure. Kernels were set to match partition sizes of the brain anatomical structure in order to capture interregional connectivities at the varying scales. With the inspiration of the widely used dropout strategy in deep learning, we developed capsule dropout in the capsule layer to prevent overfitting of the model. The comparison results showed that the proposed method outperformed the state-of-the-art methods. Besides, we compared performances using different parameters and illustrated the routing process to reveal characteristics of the proposed method. MKCapsnet is promising for schizophrenia identification. Our study first utilized the capsule neural network for analyzing functional connectivity of magnetic resonance imaging (MRI) and proposed a novel multikernel capsule structure with the consideration of brain anatomical parcellation, which could be a new way to reveal brain mechanisms. In addition, we provided useful information in the parameter setting, which is informative for further studies using a capsule network for other neurophysiological signal classification.

Multimodal Vigilance Estimation Using Deep Learning

The phenomenon of increasing accidents caused by reduced vigilance does exist. In the future, the high accuracy of vigilance estimation will play a significant role in public transportation safety. We propose a multimodal regression network that consists of multichannel deep autoencoders with subnetwork neurons (MCDAE _sn ). After we define two thresholds of "0.35" and "0.70" from the percentage of eye closure, the output values are in the continuous range of 0-0.35, 0.36-0.70, and 0.71-1 representing the awake state, the tired state, and the drowsy state, respectively. To verify the efficiency of our strategy, we first applied the proposed approach to a single modality. Then, for the multimodality, since the complementary information between forehead electrooculography and electroencephalography features, we found the performance of the proposed approach using features fusion significantly improved, demonstrating the effectiveness and efficiency of our method.

Machine Learning Techniques for the Diagnosis of Attention-Deficit/Hyperactivity Disorder from Magnetic Resonance Imaging: A Concise Review

Background

Attention-deficit/hyperactivity disorder (ADHD) is a neuro-developmental disease commonly seen in children and it is diagnosed via extensive interview procedures, behavioral studies, third-party observations, and comprehensive personal history. ADHD causes regional atrophy in brain regions and alters the pattern of functional brain connectivity networks. Automated/computerized methods based on magnetic resonance imaging (MRI) can replace subjective methods for the identification of ADHD.

Objectives

The aim of this study was to analyze various machine-learning algorithms for ADHD by feeding in vital input features extracted from functional brain connectivity and different existing methods and to review factors crucial for the diagnosis of ADHD.

Methods

This paper is a concise review of machine learning methods for the diagnosis of ADHD from MRI. Techniques for feature extraction, dimensionality reduction/feature selection, and classification, employed in the computerized techniques for the diagnosis of ADHD from MRI and the accuracy of classification offered by the individual methods, are focussed on the review.

Conclusions

Machine learning algorithms with features of functional brain connectivity networks as input, with hierarchical sparse feature elimination, exhibits the highest accuracy. Augmentation of the behavioral features does not contribute much to increased accuracy. The level of accuracy offered by the frameworks meant for the computer-aided diagnosis of ADHD, available in the literature, does not justify their feasibility in clinical practice. Computerized methods that exploit highly specific biomarkers of ADHD like brain iron concentration in Globus Pallidus, Putamen, Caudate nucleus, and thalamus as features are not available.

Trends, Limits, and Challenges of Computer Technologies in Attention Deficit Hyperactivity Disorder Diagnosis and Treatment

Attention deficit hyperactivity disorder (ADHD) is a neurobiological condition that appears during an individual's childhood and may follow her/him for life. The research objective was to understand better how and which computer technologies have been applied to support ADHD diagnosis and treatment. The research used the systematic literature review method: a rigorous, verifiable, and repeatable approach that follows well-defined steps. Six well-known academic data sources have been consulted, including search engines and bibliographic databases, from technology and health care areas. After a rigorous research protocol, 1,239 articles were analyzed. For the diagnosis, the use of machine learning techniques was verified in 61 percent of the articles. Neurofeedback was ranked second with 9.3 percent participation, followed by serious games and eye tracking with 5.6 percent each. For the treatment, neurofeedback was present in 50 percent of the articles, whereas some studies combined both approaches, accounting for 31 percent of the total. Nine percent of the articles reported remote assistance technology, whereas another 9 percent have used virtual reality. By highlighting the leading computer technologies used, their applications, results, and challenges, this literature review breaks ground for further investigations. Moreover, the study highlighted the lack of consensus on ADHD biomarkers. The approaches using machine learning call attention to the probable occurrence of overfitting in several studies, thus demonstrating limitations of this technology on small-sized bases. This research also presented the convergence of evidence from different studies on the persistence of long-term effects of using neurofeedback in treating ADHD.

Adaptive Deep Cascade Broad Learning System and Its Application in Image Denoising

This article proposes a novel regularization deep cascade broad learning system (DCBLS) architecture, which includes one cascaded feature mapping nodes layer and one cascaded enhancement nodes layer. Then, the transformation feature representation is easily obtained by incorporating the enhancement nodes and the feature mapping nodes. Once such a representation is established, a final output layer is constructed by implementing a simple convex optimization model. Furthermore, a parallelization framework on the new method is designed to make it compatible with large-scale data. Simultaneously, an adaptive regularization parameter criterion is adopted under some conditions. Moreover, the stability and error estimate of this method are discussed and proved mathematically. The proposed method could extract sufficient available information from the raw data compared with the standard broad learning system and could achieve compellent successes in image denoising. The experiments results on benchmark datasets, including natural images as well as hyperspectral images, verify the effectiveness and superiority of the proposed method in comparison with the state-of-the-art approaches for image denoising.

Neuroimaging in Attention-Deficit/Hyperactivity Disorder: Recent Advances

Attention-deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental condition, leading to impaired attention and impulsive behaviors diagnosed in, but not limited to, children. ADHD can cause symptoms throughout life. This article summarizes structural (conventional, volumetric, and diffusion tensor imaging MRI) and functional [task-based functional MRI (fMRI), resting state fMRI, PET, and MR spectroscopy] brain findings in patients with ADHD. Consensus is lacking regarding altered anatomic or functional imaging findings of the brain in children with ADHD, likely because of the disorder's heterogeneity. Most anatomic studies report abnormalities in the frontal lobes, basal ganglia, and corpus callosum; decreased surface area in the left ventral frontal and right prefrontal cortex; thinner medial temporal lobes; and smaller caudate nuclei. Using fMRI, researchers have focused on the prefrontal and temporal regions, reflecting perception-action mapping alterations. Artificial intelligence models evaluating brain anatomy have highlighted changes in cortical thickness and shape of the inferior frontal cortex, bilateral sensorimotor cortex, left temporal lobe, and insula. Early intervention and/or normal brain maturation can alter imaging patterns and convert functional imaging studies to a normal pattern. While the imaging findings provide insight into the disease's neuropathophysiology, no definitive structural or functional pattern defines the disorder from a neuroradiologic perspective.

A Novel Knowledge Distillation-Based Feature Selection for the Classification of ADHD

Attention Deficit Hyperactivity Disorder (ADHD) is a brain disorder with characteristics such as lack of concentration, excessive fidgeting, outbursts of emotions, lack of patience, difficulty in organizing tasks, increased forgetfulness, and interrupting conversation, and it is affecting millions of people worldwide. There is, until now, not a gold standard test using which an ADHD expert can differentiate between an individual with ADHD and a healthy subject, making accurate diagnosis of ADHD a challenging task. We are proposing a Knowledge Distillation-based approach to search for discriminating features between the ADHD and healthy subjects. Learned embeddings from a large neural network, trained on the functional connectivity features, were fed to one hidden layer Autoencoder for reproduction of the embeddings using the same connectivity features. Finally, a forward feature selection algorithm was used to select a combination of most discriminating features between the ADHD and the Healthy Controls. We achieved promising classification results for each of the five individual sites. A combined accuracy of 81% in KKI, 60% Peking, 56% in NYU, 64% NI, and 56% OHSU and individual site wise accuracy of 72% in KKI, 60% Peking, 73% in NYU, 70% NI, and 71% OHSU were obtained using our extracted features. Our results also outperformed state-of-the-art methods in literature which validates the efficacy of our proposed approach.

Emulative, coherent, and causal dynamics between large-scale brain networks are neurobiomarkers of Accelerated Cognitive Ageing in epilepsy

Accelerated cognitive ageing (ACA) is an ageing co-morbidity in epilepsy that is diagnosed through the observation of an evident IQ decline of more than 1 standard deviation (15 points) around the age of 50 years old. To understand the mechanism of action of this pathology, we assessed brain dynamics with the use of resting-state fMRI data. In this paper, we present novel and promising methods to extract brain dynamics between large-scale resting-state networks: the emulative power, wavelet coherence, and granger causality between the networks were extracted in two resting-state sessions of 24 participants (10 ACA, 14 controls). We also calculated the widely used static functional connectivity to compare the methods. To find the best biomarkers of ACA, and have a better understanding of this epilepsy co-morbidity we compared the aforementioned between-network neurodynamics using classifiers and known machine learning algorithms; and assessed their performance. Results show that features based on the evolutionary game theory on networks approach, the emulative powers, are the best descriptors of the co-morbidity, using dynamics associated with the default mode and dorsal attention networks. With these dynamic markers, linear discriminant analysis could identify ACA patients at 82.9% accuracy. Using wavelet coherence features with decision-tree algorithm, and static functional connectivity features with support vector machine, ACA could be identified at 77.1% and 77.9% accuracy respectively. Granger causality fell short of being a relevant biomarker with best classifiers having an average accuracy of 67.9%. Combining the features based on the game theory, wavelet coherence, Granger-causality, and static functional connectivity- approaches increased the classification performance up to 90.0% average accuracy using support vector machine with a peak accuracy of 95.8%. The dynamics of the networks that lead to the best classifier performances are known to be challenged in elderly. Since our groups were age-matched, the results are in line with the idea of ACA patients having an accelerated cognitive decline. This classification pipeline is promising and could help to diagnose other neuropsychiatric disorders, and contribute to the field of psychoradiology.

ASD-SAENet: A Sparse Autoencoder, and Deep-Neural Network Model for Detecting Autism Spectrum Disorder (ASD) Using fMRI Data

Autism spectrum disorder (ASD) is a heterogenous neurodevelopmental disorder which is characterized by impaired communication, and limited social interactions. The shortcomings of current clinical approaches which are based exclusively on behavioral observation of symptomology, and poor understanding of the neurological mechanisms underlying ASD necessitates the identification of new biomarkers that can aid in study of brain development, and functioning, and can lead to accurate and early detection of ASD. In this paper, we developed a deep-learning model called ASD-SAENet for classifying patients with ASD from typical control subjects using fMRI data. We designed and implemented a sparse autoencoder (SAE) which results in optimized extraction of features that can be used for classification. These features are then fed into a deep neural network (DNN) which results in superior classification of fMRI brain scans more prone to ASD. Our proposed model is trained to optimize the classifier while improving extracted features based on both reconstructed data error and the classifier error. We evaluated our proposed deep-learning model using publicly available Autism Brain Imaging Data Exchange (ABIDE) dataset collected from 17 different research centers, and include more than 1,035 subjects. Our extensive experimentation demonstrate that ASD-SAENet exhibits comparable accuracy (70.8%), and superior specificity (79.1%) for the whole dataset as compared to other methods. Further, our experiments demonstrate superior results as compared to other state-of-the-art methods on 12 out of the 17 imaging centers exhibiting superior generalizability across different data acquisition sites and protocols. The implemented code is available on GitHub portal of our lab at: https://github.com/pcdslab/ASD-SAENet.

Exploring characteristic features of attention-deficit/hyperactivity disorder: findings from multi-modal MRI and candidate genetic data

The current study examined whether machine learning features best distinguishing attention-deficit/hyperactivity disorder (ADHD) from typically developing children (TDC) can explain clinical phenotypes using multi-modal neuroimaging and genetic data. Cortical morphology, diffusivity scalars, resting-state functional connectivity and polygenic risk score (PS) from norepinephrine, dopamine and glutamate genes were extracted from 47 ADHD and 47 matched TDC. Using random forests, classification accuracy was measured for each uni- and multi-modal model. The optimal model was used to explain symptom severity or task performance and its robustness was validated in the independent dataset including 18 ADHD and 18 TDC. The model consisting of cortical thickness and volume features achieved the best accuracy of 85.1%. Morphological changes across insula, sensory/motor, and inferior frontal cortex were also found as key predictors. Those explained 18.0% of ADHD rating scale, while dynamic regional homogeneity within default network explained 6.4% of the omission errors in continuous performance test. Ensemble of PS to optimal model showed minor effect on accuracy. Validation analysis achieved accuracy of 69.4%. Current findings suggest that structural deformities relevant to salience detection, sensory processing, and response inhibition may be robust classifiers and symptom predictors of ADHD. Altered local functional connectivity across default network predicted attentional lapse. However, further investigation is needed to clarify roles of genetic predisposition.

Deep learning applications for the classification of psychiatric disorders using neuroimaging data: Systematic review and meta-analysis

Deep learning (DL) methods have been increasingly applied to neuroimaging data to identify patients with psychiatric and neurological disorders. This review provides an overview of the different DL applications within psychiatry and compares DL model accuracy to standard machine learning (SML). Fifty-three articles were included for qualitative analysis, primarily investigating autism spectrum disorder (ASD; n = 22), schizophrenia (SZ; n = 22) and attention-deficit/hyperactivity disorder (ADHD; n = 9). Thirty-two of the thirty-five studies that directly compared DL to SML reported a higher accuracy for DL. Only sixteen studies could be included in a meta-regression to quantitatively compare DL and SML performance. This showed a higher odds ratio for DL models, though the comparison attained significance only for ASD. Our results suggest that deep learning of neuroimaging data is a promising tool for the classification of individual psychiatric patients. However, it is not yet used to its full potential: most studies use pre-engineered features, whereas one of the main advantages of DL is its ability to learn representations of minimally processed data. Our current evaluation is limited by minimal reporting of performance measures to enable quantitative comparisons, and the restriction to ADHD, SZ and ASD as current research focusses on large publicly available datasets. To truly uncover the added value of DL, we need carefully designed comparisons of SML and DL models which are yet rarely performed.

Promises and pitfalls of deep neural networks in neuroimaging-based psychiatric research

By promising more accurate diagnostics and individual treatment recommendations, deep neural networks and in particular convolutional neural networks have advanced to a powerful tool in medical imaging. Here, we first give an introduction into methodological key concepts and resulting methodological promises including representation and transfer learning, as well as modelling domain-specific priors. After reviewing recent applications within neuroimaging-based psychiatric research, such as the diagnosis of psychiatric diseases, delineation of disease subtypes, normative modeling, and the development of neuroimaging biomarkers, we discuss current challenges. This includes for example the difficulty of training models on small, heterogeneous and biased data sets, the lack of validity of clinical labels, algorithmic bias, and the influence of confounding variables.

Machine Learning Methods for Diagnosing Autism Spectrum Disorder and Attention- Deficit/Hyperactivity Disorder Using Functional and Structural MRI: A Survey

Here we summarize recent progress in machine learning model for diagnosis of Autism Spectrum Disorder (ASD) and Attention-deficit/Hyperactivity Disorder (ADHD). We outline and describe the machine-learning, especially deep-learning, techniques that are suitable for addressing research questions in this domain, pitfalls of the available methods, as well as future directions for the field. We envision a future where the diagnosis of ASD, ADHD, and other mental disorders is accomplished, and quantified using imaging techniques, such as MRI, and machine-learning models.

Supervised machine learning for diagnostic classification from large-scale neuroimaging datasets

There are growing concerns about the generalizability of machine learning classifiers in neuroimaging. In order to evaluate this aspect across relatively large heterogeneous populations, we investigated four disorders: Autism spectrum disorder (N = 988), Attention deficit hyperactivity disorder (N = 930), Post-traumatic stress disorder (N = 87) and Alzheimer's disease (N = 132). We applied 18 different machine learning classifiers (based on diverse principles) wherein the training/validation and the hold-out test data belonged to samples with the same diagnosis but differing in either the age range or the acquisition site. Our results indicate that overfitting can be a huge problem in heterogeneous datasets, especially with fewer samples, leading to inflated measures of accuracy that fail to generalize well to the general clinical population. Further, different classifiers tended to perform well on different datasets. In order to address this, we propose a consensus-classifier by combining the predictive power of all 18 classifiers. The consensus-classifier was less sensitive to unmatched training/validation and holdout test data. Finally, we combined feature importance scores obtained from all classifiers to infer the discriminative ability of connectivity features. The functional connectivity patterns thus identified were robust to the classification algorithm used, age and acquisition site differences, and had diagnostic predictive ability in addition to univariate statistically significant group differences between the groups. A MATLAB toolbox called Machine Learning in NeuroImaging (MALINI), which implements all the 18 different classifiers along with the consensus classifier is available from Lanka et al. (2019) The toolbox can also be found at the following URL: https://github.com/pradlanka/malini .

Density-based clustering of static and dynamic functional MRI connectivity features obtained from subjects with cognitive impairment

Various machine-learning classification techniques have been employed previously to classify brain states in healthy and disease populations using functional magnetic resonance imaging (fMRI). These methods generally use supervised classifiers that are sensitive to outliers and require labeling of training data to generate a predictive model. Density-based clustering, which overcomes these issues, is a popular unsupervised learning approach whose utility for high-dimensional neuroimaging data has not been previously evaluated. Its advantages include insensitivity to outliers and ability to work with unlabeled data. Unlike the popular k-means clustering, the number of clusters need not be specified. In this study, we compare the performance of two popular density-based clustering methods, DBSCAN and OPTICS, in accurately identifying individuals with three stages of cognitive impairment, including Alzheimer’s disease. We used static and dynamic functional connectivity features for clustering, which captures the strength and temporal variation of brain connectivity respectively. To assess the robustness of clustering to noise/outliers, we propose a novel method called recursive-clustering using additive-noise (R-CLAN). Results demonstrated that both clustering algorithms were effective, although OPTICS with dynamic connectivity features outperformed in terms of cluster purity (95.46%) and robustness to noise/outliers. This study demonstrates that density-based clustering can accurately and robustly identify diagnostic classes in an unsupervised way using brain connectivity.

Deep Representation Learning For Multimodal Brain Networks

Applying network science approaches to investigate the functions and anatomy of the human brain is prevalent in modern medical imaging analysis. Due to the complex network topology, for an individual brain, mining a discriminative network representation from the multimodal brain networks is non-trivial. The recent success of deep learning techniques on graph-structured data suggests a new way to model the non-linear cross-modality relationship. However, current deep brain network methods either ignore the intrinsic graph topology or require a network basis shared within a group. To address these challenges, we propose a novel end-to-end deep graph representation learning (Deep Multimodal Brain Networks - DMBN) to fuse multimodal brain networks. Specifically, we decipher the cross-modality relationship through a graph encoding and decoding process. The higher-order network mappings from brain structural networks to functional networks are learned in the node domain. The learned network representation is a set of node features that are informative to induce brain saliency maps in a supervised manner. We test our framework in both synthetic and real image data. The experimental results show the superiority of the proposed method over some other state-of-the-art deep brain network models.

Recent advances of deep learning in psychiatric disorders

Deep learning (DL) is a recently proposed subset of machine learning methods that has gained extensive attention in the academic world, breaking benchmark records in areas such as visual recognition and natural language processing. Different from conventional machine learning algorithm, DL is able to learn useful representations and features directly from raw data through hierarchical nonlinear transformations. Because of its ability to detect abstract and complex patterns, DL has been used in neuroimaging studies of psychiatric disorders, which are characterized by subtle and diffuse alterations. Here, we provide a brief review of recent advances and associated challenges in neuroimaging studies of DL applied to psychiatric disorders. The results of these studies indicate that DL could be a powerful tool in assisting the diagnosis of psychiatric diseases. We conclude our review by clarifying the main promises and challenges of DL application in psychiatric disorders, and possible directions for future research.

Towards a brain-based predictome of mental illness

Neuroimaging-based approaches have been extensively applied to study mental illness in recent years and have deepened our understanding of both cognitively healthy and disordered brain structure and function. Recent advancements in machine learning techniques have shown promising outcomes for individualized prediction and characterization of patients with psychiatric disorders. Studies have utilized features from a variety of neuroimaging modalities, including structural, functional, and diffusion magnetic resonance imaging data, as well as jointly estimated features from multiple modalities, to assess patients with heterogeneous mental disorders, such as schizophrenia and autism. We use the term "predictome" to describe the use of multivariate brain network features from one or more neuroimaging modalities to predict mental illness. In the predictome, multiple brain network-based features (either from the same modality or multiple modalities) are incorporated into a predictive model to jointly estimate features that are unique to a disorder and predict subjects accordingly. To date, more than 650 studies have been published on subject-level prediction focusing on psychiatric disorders. We have surveyed about 250 studies including schizophrenia, major depression, bipolar disorder, autism spectrum disorder, attention-deficit hyperactivity disorder, obsessive-compulsive disorder, social anxiety disorder, posttraumatic stress disorder, and substance dependence. In this review, we present a comprehensive review of recent neuroimaging-based predictomic approaches, current trends, and common shortcomings and share our vision for future directions.

Neuroimaging-Based Deep Learning in Autism Spectrum Disorder and Attention-Deficit/Hyperactivity Disorder

Deep learning (DL) is a kind of machine learning technique that uses artificial intelligence to identify the characteristics of given data and efficiently analyze large amounts of information to perform tasks such as classification and prediction. In the field of neuroimaging of neurodevelopmental disorders, various biomarkers for diagnosis, classification, prognosis prediction, and treatment response prediction have been examined; however, they have not been efficiently combined to produce meaningful results. DL can be applied to overcome these limitations and produce clinically helpful results. Here, we review studies that combine neurodevelopmental disorder neuroimaging and DL techniques to explore the strengths, limitations, and future directions of this research area.

Morbigenous brain region and gene detection with a genetically evolved random neural network cluster approach in late mild cognitive impairment

MOTIVATION

The multimodal data fusion analysis becomes another important field for brain disease detection and increasing researches concentrate on using neural network algorithms to solve a range of problems. However, most current neural network optimizing strategies focus on internal nodes or hidden layer numbers, while ignoring the advantages of external optimization. Additionally, in the multimodal data fusion analysis of brain science, the problems of small sample size and high-dimensional data are often encountered due to the difficulty of data collection and the specialization of brain science data, which may result in the lower generalization performance of neural network.

RESULTS

We propose a genetically evolved random neural network cluster (GERNNC) model. Specifically, the fusion characteristics are first constructed to be taken as the input and the best type of neural network is selected as the base classifier to form the initial random neural network cluster. Second, the cluster is adaptively genetically evolved. Based on the GERNNC model, we further construct a multi-tasking framework for the classification of patients with brain disease and the extraction of significant characteristics. In a study of genetic data and functional magnetic resonance imaging data from the Alzheimer's Disease Neuroimaging Initiative, the framework exhibits great classification performance and strong morbigenous factor detection ability. This work demonstrates that how to effectively detect pathogenic components of the brain disease on the high-dimensional medical data and small samples.

AVAILABILITY AND IMPLEMENTATION

The Matlab code is available at https://github.com/lizi1234560/GERNNC.git.

Deep Learning Convolutional Neural Networks Discriminate Adult ADHD From Healthy Individuals on the Basis of Event-Related Spectral EEG

Attention deficit hyperactivity disorder (ADHD) is a heterogeneous neurodevelopmental disorder that affects 5% of the pediatric and adult population worldwide. The diagnosis remains essentially clinical, based on history and exam, with no available biomarkers. In this paper, we describe a convolutional neural network (CNN) with a four-layer architecture combining filtering and pooling, which we train using stacked multi-channel EEG time-frequency decompositions (spectrograms) of electroencephalography data (EEG), particularly of event-related potentials (ERP) from ADHD patients (n = 20) and healthy controls (n = 20) collected during the Flanker Task, with 2800 samples for each group. We treat the data as in audio or image classification approaches, where deep networks have proven successful by exploiting invariances and compositional features in the data. The model reaches a classification accuracy of 88% ± 1.12%, outperforming the Recurrent Neural Network and the Shallow Neural Network used for comparison, and with the key advantage, compared with other machine learning approaches, of avoiding the need for manual selection of EEG spectral or channel features. The event-related spectrograms also provide greater accuracy compared to resting state EEG spectrograms. Finally, through the use of feature visualization techniques such as DeepDream, we show that the main features exciting the CNN nodes are a decreased power in the alpha band and an increased power in the delta-theta band around 100 ms for ADHD patients compared to healthy controls, suggestive of attentional and inhibition deficits, which have been previously suggested as pathophyisiological signatures of ADHD. While confirmation with larger clinical samples is necessary, these results suggest that deep networks may provide a useful tool for the analysis of EEG dynamics even from relatively small datasets, highlighting the potential of this methodology to develop biomarkers of practical clinical utility.

Multimodal neuroimaging-based prediction of adult outcomes in childhood-onset ADHD using ensemble learning techniques

Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent and heterogeneous neurodevelopmental disorder, which is diagnosed using subjective symptom reports. Machine learning classifiers have been utilized to assist in the development of neuroimaging-based biomarkers for objective diagnosis of ADHD. However, existing basic model-based studies in ADHD report suboptimal classification performances and inconclusive results, mainly due to the limited flexibility for each type of basic classifier to appropriately handle multi-dimensional source features with varying properties. This study applied ensemble learning techniques (ELTs), a meta-algorithm that combine several basic machine learning models into one predictive model in order to decrease variance, bias, or improve predictions, in multimodal neuroimaging data collected from 72 young adults, including 36 probands (18 remitters and 18 persisters of childhood ADHD) and 36 group-matched controls. All currently available optimization strategies for ELTs (i.e., voting, bagging, boosting and stacking techniques) were tested in a pool of semifinal classification results generated by seven basic classifiers. The high-dimensional neuroimaging features for classification included regional cortical gray matter (GM) thickness and surface area, GM volume of subcortical structures, volume and fractional anisotropy of major white matter fiber tracts, pair-wise regional connectivity and global/nodal topological properties of the functional brain network for cue-evoked attention process. As a result, the bagging-based ELT with the base model of support vector machine achieved the best results, with significant improvement of the area under the receiver of operating characteristic curve (0.89 for ADHD vs. controls and 0.9 for ADHD persisters vs. remitters). Features of nodal efficiency in right inferior frontal gyrus, right middle frontal (MFG)-inferior parietal (IPL) functional connectivity, and right amygdala volume significantly contributed to accurate discrimination between ADHD probands and controls; higher nodal efficiency of right MFG greatly contributed to inattentive and hyperactive/impulsive symptom remission, while higher right MFG-IPL functional connectivity strongly linked to symptom persistence in adults with childhood ADHD. Considering their improved robustness than the commonly implemented basic classifiers, findings suggest that ELTs may have the potential to identify more reliable neurobiological markers for neurodevelopmental disorders.

Accelerating Convolutional Neural Networks by Removing Interspatial and Interkernel Redundancies

Recently, the high computational resource demands of convolutional neural networks (CNNs) have hindered a wide range of their applications. To solve this problem, many previous works attempted to reduce the redundant calculations during the evaluation of CNNs. However, these works mainly focused on either interspatial or interkernel redundancy. In this paper, we further accelerate existing CNNs by removing both types of redundancies. First, we convert interspatial redundancy into interkernel redundancy by decomposing one convolutional layer to one block that we design. Then, we adopt rank-selection and pruning methods to remove the interkernel redundancy. The rank-selection method, which considerably reduces manpower, contributes to determining the number of kernels to be pruned in the pruning method. We apply a layer-wise training algorithm rather than the traditional end-to-end training to overcome the difficulty of convergence. Finally, we fine-tune the entire network to achieve better performance. Our method is applied on three widely used datasets of an image classification task. We achieve better results in terms of accuracy and compression rate compared with previous state-of-the-art methods.

MALINI (Machine Learning in NeuroImaging): A MATLAB toolbox for aiding clinical diagnostics using resting-state fMRI data

Resting-state functional Magnetic Resonance Imaging (rs-fMRI) has been extensively used for diagnostic classification because it does not require task compliance and is easier to pool data from multiple imaging sites, thereby increasing the sample size. A MATLAB-based toolbox called Machine Learning in NeuroImaging (MALINI) for feature extraction and disease classification is presented. The MALINI toolbox extracts functional and effective connectivity features from preprocessed rs-fMRI data and performs classification between healthy and disease groups using any of 18 popular and widely used machine learning algorithms that are based on diverse principles. A consensus classifier combining the power of multiple classifiers is also presented. The utility of the toolbox is illustrated by accompanying data consisting of resting-state functional connectivity features from healthy controls and subjects with various brain-based disorders: autism spectrum disorder from autism brain imaging data exchange (ABIDE), Alzheimer's disease and mild cognitive impairment from Alzheimer's disease neuroimaging initiative (ADNI), attention deficit hyperactivity disorder from ADHD-200, and post-traumatic stress disorder and post-concussion syndrome acquired in-house. Results of classification performed on the above datasets can be obtained from the main article titled “Supervised machine learning for diagnostic classification from large-scale neuroimaging datasets” [1]. The data was divided into homogeneous and heterogeneous splits, such that 80% could be used for training, model building and cross-validation, while the remaining 20% of the data could be used as a hold-out independent test data for replication of the classification performance, to ensure the robustness of the classifiers to population variance in image acquisition site and age of the sample.

Estimation of wave reflection in aorta from radial pulse waveform by artificial neural network: a numerical study

BACKGROUND AND OBJECTIVE

Wave reflection in aorta has been shown to have incremental value for predicting cardiovascular events. However, its estimation by wave separation analysis (WSA) is complex.

METHODS

In this study, a novel method was proposed based on a cascade artificial neural network (ANN) for wave reflection estimation by the frequency features of radial pressure waveform alone. The simulation database of 4000 samples was generated by a 55-segment transmission line model of human arterial tree and was used for evaluating the ANN with 10-fold cross validation for the estimation of reflection magnitude (RM_ANN) and reflection index (RI_ANN) of wave reflection in aorta. RM and RI also were estimated by the WSA with a triangle waveform of aortic flow (RM_WSA and RI_WSA) and with a real aortic flow waveform (RM_Ref and RI_Ref) as reference values.

RESULTS

The results showed the correlation coefficient and mean difference between RM_ANN and RM_Ref (R² = 0.92, mean ± standard deviation (SD) = 0.0 ± 0.02) and those between RI_ANN and RI_Ref (R² = 0.91, mean ± SD = 0.0 ± 0.01) were better than those between RM_WSA and RM_Ref (R² = 0.51, mean ± SD = 0.01 ± 0.07) and those between RI_WSA and RI_Ref (R² = 0.50, mean ± SD = 0.0 ± 0.02). As the sample diversity in the simulation database was increased but the total number of samples keeps constant, the advantage of the ANN, though decreased slightly, became more significant than those of WSA (RM_ANN VS. RM_Ref and RI_ANN VS. RI_Ref: R² = 0.88 and 0.88, mean ± SD = 0.0 ± 0.05 and 0.0 ± 0.05; RM_WSA VS. RM_Ref and RI_WSA VS. RI_Ref: R² = 0.24 and 0.24, mean ± SD = 0.07 ± 0.24 and 0.02 ± 0.08, respectively). In addition, the ANN can achieve better results than the traditional method WSA even only two hidden neurons are used.

CONCLUSIONS

ANN is a potential method for the estimation of wave reflection in aorta by a single radial pulse waveform, but further validation of this method in clinic trials is needed.

ASD-DiagNet: A Hybrid Learning Approach for Detection of Autism Spectrum Disorder Using fMRI Data

Heterogeneous mental disorders such as Autism Spectrum Disorder (ASD) are notoriously difficult to diagnose, especially in children. The current psychiatric diagnostic process is based purely on the behavioral observation of symptomology (DSM-5/ICD-10) and may be prone to misdiagnosis. In order to move the field toward more quantitative diagnosis, we need advanced and scalable machine learning infrastructure that will allow us to identify reliable biomarkers of mental health disorders. In this paper, we propose a framework called ASD-DiagNet for classifying subjects with ASD from healthy subjects by using only fMRI data. We designed and implemented a joint learning procedure using an autoencoder and a single layer perceptron (SLP) which results in improved quality of extracted features and optimized parameters for the model. Further, we designed and implemented a data augmentation strategy, based on linear interpolation on available feature vectors, that allows us to produce synthetic datasets needed for training of machine learning models. The proposed approach is evaluated on a public dataset provided by Autism Brain Imaging Data Exchange including 1, 035 subjects coming from 17 different brain imaging centers. Our machine learning model outperforms other state of the art methods from 10 imaging centers with increase in classification accuracy up to 28% with maximum accuracy of 82%. The machine learning technique presented in this paper, in addition to yielding better quality, gives enormous advantages in terms of execution time (40 min vs. 7 h on other methods). The implemented code is available as GPL license on GitHub portal of our lab (https://github.com/pcdslab/ASD-DiagNet).

Deep neural networks in psychiatry

Machine and deep learning methods, today's core of artificial intelligence, have been applied with increasing success and impact in many commercial and research settings. They are powerful tools for large scale data analysis, prediction and classification, especially in very data-rich environments ("big data"), and have started to find their way into medical applications. Here we will first give an overview of machine learning methods, with a focus on deep and recurrent neural networks, their relation to statistics, and the core principles behind them. We will then discuss and review directions along which (deep) neural networks can be, or already have been, applied in the context of psychiatry, and will try to delineate their future potential in this area. We will also comment on an emerging area that so far has been much less well explored: by embedding semantically interpretable computational models of brain dynamics or behavior into a statistical machine learning context, insights into dysfunction beyond mere prediction and classification may be gained. Especially this marriage of computational models with statistical inference may offer insights into neural and behavioral mechanisms that could open completely novel avenues for psychiatric treatment.

Novel relative relevance score for estimating brain connectivity from fMRI data using an explainable neural network approach

BACKGROUND

Functional integration or connectivity in brain is directional, non-linear as well as variable in time-lagged dependence. Deep neural networks (DNN) have become an indispensable tool everywhere, by learning higher levels of abstract and complex patterns from raw data. However, in neuroscientific community they generally work as black-boxes, leading to the explanation of results difficult and less intuitive. We aim to propose a brain-connectivity measure based on an explainable NN (xNN) approach.

NEW METHOD

We build a NN-based predictor for regression problem. Since we aim to determine the contribution/relevance of past data-point from one region i in the prediction of current data-point from another region j, i.e. the higher-order connectivity between two brain-regions, we employ layer-wise relevance propagation (Bach et al., 2015) (LRP, a method for explaining DNN predictions), which has not been done before to the best of our knowledge. Specifically, we propose a novel score depending on weights as a quantitative measure of connectivity, called as relative relevance score (xNN-RRS). The RRS is an intuitive and transparent score. We provide an interpretation of the trained NN weights with-respect-to the brain-connectivity.

RESULTS

Face validity of our approach is demonstrated with experiments on simulated data, over existing methods. We also demonstrate construct validity of xNN-RRS in a resting-state fMRI experiment.

COMPARISON

Our approach shows superior performance, in terms of accuracy and computational complexity, over existing state-of-the-art methods for brain-connectivity estimation.

CONCLUSION

The proposed method is promising to serve as a first post-hoc explainable NN-approach for brain-connectivity analysis in clinical applications.

Dynamics of Segregation and Integration in Directional Brain Networks: Illustration in Soldiers With PTSD and Neurotrauma

Brain functioning relies on various segregated/specialized neural regions functioning as an integrated-interconnected network (i.e., metastability). Various psychiatric and neurologic disorders are associated with aberrant functioning of these brain networks. In this study, we present a novel framework integrating the strength and temporal variability of metastability in brain networks. We demonstrate that this approach provides novel mechanistic insights which enables better imaging-based predictions. Using whole-brain resting-state fMRI and a graph-theoretic framework, we integrated strength and temporal-variability of complex-network properties derived from effective connectivity networks, obtained from 87 U.S. Army soldiers consisting of healthy combat controls (n = 28), posttraumatic stress disorder (PTSD; n = 17), and PTSD with comorbid mild-traumatic brain injury (mTBI; n = 42). We identified prefrontal dysregulation of key subcortical and visual regions in PTSD/mTBI, with all network properties exhibiting lower variability over time, indicative of poorer flexibility. Larger impairment in the prefrontal-subcortical pathway but not prefrontal-visual pathway differentiated comorbid PTSD/mTBI from the PTSD group. Network properties of the prefrontal-subcortical pathway also had significant association (R ² = 0.56) with symptom severity and neurocognitive performance; and were also found to possess high predictive ability (81.4% accuracy in classifying the disorders, explaining 66-72% variance in symptoms), identified through machine learning. Our framework explained 13% more variance in behaviors compared to the conventional framework. These novel insights and better predictions were made possible by our novel framework using static and time-varying network properties in our three-group scenario, advancing the mechanistic understanding of PTSD and comorbid mTBI. Our contribution has wide-ranging applications for network-level characterization of healthy brains as well as mental disorders.