Differentiable neural architecture search for optimal spatial/temporal brain function network decomposition

2D MoTe2 memristors for energy-efficient artificial synapses and neuromorphic applications

The potential of two-dimensional (2D) transition metal dichalcogenides (TMDs), especially molybdenum telluride (MoTe2), in sophisticated electrical and low-energy neuromorphic applications, has attracted a lot of interest. The creation, characteristics, and uses of MoTe2-based memristive devices are summarized in this review paper, with an emphasis on their potential as artificial synapses for neuromorphic computing. We thoroughly examine the special properties of MoTe2, such as its remarkable resistance switching response, excellent linearity in synaptic potentiation, and customizable phase states. These characteristics make it possible to implement basic computational functions with minimal energy consumption, including decimal arithmetic operations and the commutative principles of addition and multiplication. In addition to simulating intricate synaptic processes such as long-term potentiation (LTP), long-term depression (LTD), and spike-timing-dependent plasticity (STDP), the article emphasizes the experimental performances of MoTe2 memristors, which include their capacity to execute exact decimal arithmetic operations. The demonstration of centimeter-scale 2D MoTe2 film-based memristor arrays attaining over 90% recognition accuracy in handwritten digit identification tests further demonstrates the devices' great scalability, stability, and incorporation capabilities. Notwithstanding these developments, issues such as poor environmental robustness, phase transition sensitivity, and low thermal stability still exist. The creation of hybrid or composite materials, doping, and structural alteration are some of the methods to get beyond these obstacles that are covered in the paper. The need for scalable, economical synthesis techniques and a better comprehension of the material's mechanical, optical, and electrical properties through modeling and experiments are emphasized.

Advances in Ga2O3-based memristor devices, modeling, properties, and applications for low power neuromorphic computing

About a decade ago, gallium oxide (Ga2O3) was found to be a very attractive ultrawide-bandgap (4.6-4.9 eV) semiconductor for next-generation low-power devices. Ga2O3 materials have attracted a lot of scientific and technical interest because of their outstanding properties and numerous application opportunities in the field of semiconductor based memristor technology. This review is focused on Ga2O3 thin-film memristors for smart technologies. The capacitance behavior of memristors is very important for adapting nonlinear memristor responses. Also, this comprehensive review explores in depth the ideas, device construction, and manufacturing procedures for Ga2O3-based memristor devices. To improve the device's behavior and performance improvement, a detailed analysis of many modeling and simulation techniques is given. Also, advanced characterization techniques, such as electrical, structural, and thermal evaluations, for studying artificial optoelectronic synaptic characteristics, which are important for use in computational neuroscience, are discussed in detail. The synaptic activities revealed that learning and memory processes were aided by potentiation and depression similar to those found in biological synapses. The most notable accomplishment is the realization of quaternary memory storage in a single device. This idea is supported by empirical evidence and simulations, which demonstrate the possibility of storing and maintaining multiple memory states. This study establishes oxide semiconductor memristors as a doorway to quaternary memory storage and improved synaptic functioning, paving the way for optoelectronic synaptic devices with greater memory capacity.

A review of AutoML optimization techniques for medical image applications

Automatic analysis of medical images using machine learning techniques has gained significant importance over the years. A large number of approaches have been proposed for solving different medical image analysis tasks using machine learning and deep learning approaches. These approaches are quite effective thanks to their ability to analyze large volume of medical imaging data. Moreover, they can also identify patterns that may be difficult for human experts to detect. Manually designing and tuning the parameters of these algorithms is a challenging and time-consuming task. Furthermore, designing a generalized model that can handle different imaging modalities is difficult, as each modality has specific characteristics. To solve these problems and automate the whole pipeline of different medical image analysis tasks, numerous Automatic Machine Learning (AutoML) techniques have been proposed. These techniques include Hyper-parameter Optimization (HPO), Neural Architecture Search (NAS), and Automatic Data Augmentation (ADA). This study provides an overview of several AutoML-based approaches for different medical imaging tasks in terms of optimization search strategies. The usage of optimization techniques (evolutionary, gradient-based, Bayesian optimization, etc.) is of significant importance for these AutoML approaches. We comprehensively reviewed existing AutoML approaches, categorized them, and performed a detailed analysis of different proposed approaches. Furthermore, current challenges and possible future research directions are also discussed.

STANet: A Novel Spatio-Temporal Aggregation Network for Depression Classification with Small and Unbalanced FMRI Data

Background: Early diagnosis of depression is crucial for effective treatment and suicide prevention. Traditional methods rely on self-report questionnaires and clinical assessments, lacking objective biomarkers. Combining functional magnetic resonance imaging (fMRI) with artificial intelligence can enhance depression diagnosis using neuroimaging indicators, but depression-specific fMRI datasets are often small and imbalanced, posing challenges for classification models. New Method: We propose the Spatio-Temporal Aggregation Network (STANet) for diagnosing depression by integrating convolutional neural networks (CNN) and recurrent neural networks (RNN) to capture both temporal and spatial features of brain activity. STANet comprises the following steps: (1) Aggregate spatio-temporal information via independent component analysis (ICA). (2) Utilize multi-scale deep convolution to capture detailed features. (3) Balance data using the synthetic minority over-sampling technique (SMOTE) to generate new samples for minority classes. (4) Employ the attention-Fourier gate recurrent unit (AFGRU) classifier to capture long-term dependencies, with an adaptive weight assignment mechanism to enhance model generalization. Results: STANet achieves superior depression diagnostic performance, with 82.38% accuracy and a 90.72% AUC. The Spatio-Temporal Feature Aggregation module enhances classification by capturing deeper features at multiple scales. The AFGRU classifier, with adaptive weights and a stacked Gated Recurrent Unit (GRU), attains higher accuracy and AUC. SMOTE outperforms other oversampling methods. Additionally, spatio-temporal aggregated features achieve better performance compared to using only temporal or spatial features. Comparison with existing methods: STANet significantly outperforms traditional classifiers, deep learning classifiers, and functional connectivity-based classifiers. Conclusions: The successful performance of STANet contributes to enhancing the diagnosis and treatment assessment of depression in clinical settings on imbalanced and small fMRI.

Modeling Functional Brain Networks for ADHD via Spatial Preservation-Based Neural Architecture Search

Modeling functional brain networks (FBNs) for attention deficit hyperactivity disorder (ADHD) has sparked significant interest since the abnormal functional connectivity is discovered in certain functional magnetic resonance imaging (fMRI)-based brain regions compared to typical developmental control (TC) individuals. However, existing models for modeling FBNs generally use dimensionality reduction techniques to process the high dimensional input data, which results in confusion and an inaccurate representation of voxel interactions between spatially close brain regions, causing misdiagnosis of the disease. To address these issues, we propose a spatial preservation-based neural architecture search (SP-NAS) for FBNs modeling in ADHD. The main work includes three-fold: 1) A spatial preservation module is designed to embed original spatial information into dimensionality reduction data, addressing the challenge of a large number of parameters in the original data and mitigating disease misdiagnosis resulting from voxel confusion between different brain regions caused by dimensionality reduction. 2) A search space using more suitable search operations is constructed to efficiently extract spatial-temporal interaction characteristics of fMRI data in ADHD while narrowing the search space. 3) Cross-regional association differences between ADHD and TC groups are explored for ADHD auxiliary diagnosis since the abnormal activation regions of ADHD relative to TC on the brain regions and the abnormal connectivity between the lesion brain regions are identified. Model validation results on the ADHD-200 dataset show that the FBNs obtained from SP-NAS not only achieve competitive results in ADHD diagnosis but also reveal abnormal connections in the lesion regions of ADHD consistent with clinical diagnosis.

ADHD diagnosis guided by functional brain networks combined with domain knowledge

Utilizing functional magnetic resonance imaging (fMRI) to model functional brain networks (FBNs) is increasingly prominent in attention-deficit/hyperactivity disorder (ADHD) research, revealing neural impact and mechanisms through the exploration of activated brain regions. However, current FBNs-based methods face two major challenges. The primary challenge stems from the limitations of existing modeling methods in accurately capturing both regional correlations and long-distance dependencies (LDDs) within the dynamic brain, thereby affecting the diagnostic accuracy of FBNs as biomarkers. Additionally, limited sample size and class imbalance also pose a challenge to the learning performance of the model. To address the issues, we propose an automated diagnostic framework, which integrates modeling, multimodal fusion, and classification into a unified process. It aims to extract representative FBNs and efficiently incorporate domain knowledge to guide ADHD classification. Our work mainly includes three-fold: 1) A multi-head attention-based region-enhancement module (MAREM) is designed to simultaneously capture regional correlations and LDDs across the entire sequence of brain activity, which facilitates the construction of representative FBNs. 2) The multimodal supplementary learning module (MSLM) is proposed to integrate domain knowledge from phenotype data with FBNs from neuroimaging data, achieving information complementarity and alleviating the problems of insufficient medical data and unbalanced sample categories. 3) An ADHD automatic diagnosis framework guided by FBNs and domain knowledge (ADF-FAD) is proposed to help doctors make more accurate decisions, which is applied to the ADHD-200 dataset to confirm its effectiveness. The results indicate that the FBNs extracted by MAREM perform well in modeling and classification. After with MSLM, the model achieves accuracy of 92.4%, 74.4%, and 80% at NYU, PU, and KKI, respectively, demonstrating its ability to effectively capture crucial information related to ADHD diagnosis. Codes are available at https://github.com/zhuimengxuebao/ADF-FAD.

Task sub-type states decoding via group deep bidirectional recurrent neural network

Decoding brain states under different cognitive tasks from functional magnetic resonance imaging (fMRI) data has attracted great attention in the neuroimaging filed. However, the well-known temporal dependency in fMRI sequences has not been fully exploited in existing studies, due to the limited temporal-modeling capacity of the backbone machine learning algorithms and rigid training sample organization strategies upon which the brain decoding methods are built. To address these limitations, we propose a novel method for fine-grain brain state decoding, namely, group deep bidirectional recurrent neural network (Group-DBRNN) model. We first propose a training sample organization strategy that consists of a group-task sample generation module and a multiple-scale random fragment strategy (MRFS) module to collect training samples that contain rich task-relevant brain activity contrast (i.e., the comparison of neural activity patterns between different tasks) and maintain the temporal dependency. We then develop a novel decoding model by replacing the unidirectional RNNs that are widely used in existing brain state decoding studies with bidirectional stacked RNNs to better capture the temporal dependency, and by introducing a multi-task interaction layer (MTIL) module to effectively model the task-relevant brain activity contrast. Our experimental results on the Human Connectome Project task fMRI dataset (7 tasks consisting of 23 task sub-type states) show that the proposed model achieves an average decoding accuracy of 94.7% over the 23 fine-grain sub-type states. Meanwhile, our extensive interpretations of the intermediate features learned in the proposed model via visualizations and quantitative assessments of their discriminability and inter-subject alignment evidence that the proposed model can effectively capture the temporal dependency and task-relevant contrast.

Memristor-Based Neuromorphic Chips

In the era of information, characterized by an exponential growth in data volume and an escalating level of data abstraction, there has been a substantial focus on brain-like chips, which are known for their robust processing power and energy-efficient operation. Memristors are widely acknowledged as the optimal electronic devices for the realization of neuromorphic computing, due to their innate ability to emulate the interconnection and information transfer processes witnessed among neurons. This review paper focuses on memristor-based neuromorphic chips, which provide an extensive description of the working principle and characteristic features of memristors, along with their applications in the realm of neuromorphic chips. Subsequently, a thorough discussion of the memristor array, which serves as the pivotal component of the neuromorphic chip, as well as an examination of the present mainstream neural networks, is delved. Furthermore, the design of the neuromorphic chip is categorized into three crucial sections, including synapse-neuron cores, networks on chip (NoC), and neural network design. Finally, the key performance metrics of the chip is highlighted, as well as the key metrics related to the memristor devices are employed to realize both the synaptic and neuronal components.

Modeling default mode network patterns via a universal spatio-temporal brain attention skip network

Designing a comprehensive four-dimensional resting-state functional magnetic resonance imaging (4D Rs-fMRI) based default mode network (DMN) modeling methodology to reveal the spatio-temporal patterns of individual DMN, is crucial for understanding the cognitive mechanisms of the brain and the pathogenesis of psychiatric disorders. However, there are still two limitations of existing approaches for DMN modeling. The approaches either (1) simply split the spatio-temporal components and ignore the overall character of the spatio-temporal patterns or (2) are biased in the process of feature extraction for DMN modeling, and their spatio-temporal accuracy is thus not warranted. To this end, we propose a novel Spatio-Temporal Brain Attention Skip Network (STBAS-Net) to model the personalized spatio-temporal patterns of the DMN. STBAS-Net consists of spatial and temporal components, where the multi-head attention skip connection block in the spatial component achieves detailed feature extraction and enhancement in the shallow stage. Under the guidance of spatial information, we technically fuse multiple spatio-temporal information in the temporal component, which dexterously exploits the overall spatio-temporal features and achieves mutual constraints of spatio-temporal patterns to characterize the spatio-temporal patterns of the DMN. We verify the proposed STBAS-Net on a publicly released 4D Rs-fMRI dataset and an EMCI dataset. The experimental results show that compared with existing advanced methods, the proposed network can more accurately model the personalized spatio-temporal patterns of the human brain DMN and successfully identify abnormal spatio-temporal patterns in EMCI patients. This study provides a potential tool for revealing the spatio-temporal patterns of the human brain DMN and is expected to provide an effective methodological framework for future exploration of abnormal brain spatio-temporal patterns and modeling of other functional brain networks.

A comprehensive survey of complex brain network representation

Recent years have shown great merits in utilizing neuroimaging data to understand brain structural and functional changes, as well as its relationship to different neurodegenerative diseases and other clinical phenotypes. Brain networks, derived from different neuroimaging modalities, have attracted increasing attention due to their potential to gain system-level insights to characterize brain dynamics and abnormalities in neurological conditions. Traditional methods aim to pre-define multiple topological features of brain networks and relate these features to different clinical measures or demographical variables. With the enormous successes in deep learning techniques, graph learning methods have played significant roles in brain network analysis. In this survey, we first provide a brief overview of neuroimaging-derived brain networks. Then, we focus on presenting a comprehensive overview of both traditional methods and state-of-the-art deep-learning methods for brain network mining. Major models, and objectives of these methods are reviewed within this paper. Finally, we discuss several promising research directions in this field.

Extracting transition features among brain states based on coarse-grained similarity measurement for autism spectrum disorder analysis

BACKGROUND

The abnormal brain functional connectivity (FC) of patients with mental diseases is closely linked to the transition features among brain states. However, the current research on state transition will produce certain division deviations in the measurement method of state division, and also ignore the transition features among multiple states that contain more abundant information for analyzing brain diseases.

PURPOSE

To investigate the potential of the proposed method based on coarse-grained similarity measurement to solve the problem of state division, and consider the transition features among multiple states to analyze the FC abnormalities of autism spectrum disorder (ASD) patients.

METHODS

We used resting-state functional magnetic resonance imaging to examine 45 ASD and 47 healthy controls (HC). The FC between brain regions was calculated by the sliding window and correlation algorithm, and a novel coarse-grained similarity measure method was used to cluster the FC networks into five states, and then extract the features both of the state itself and the transition features among multiple states for analysis and diagnosis.

RESULTS

(1) The state as divided by the coarse-grained measurement method improves the diagnostic performance of individuals with ASD compared with previous methods. (2) The transition features among multiple states can provide complementary information to the features of the state itself in the ASD analysis and diagnosis. (3) ASD individuals have different brain state transitions than HC. Specifically, the abnormalities in intra- and inter-network connectivity of ASD patients mainly occur in the default mode network, the visual network, and the cerebellum.

CONCLUSIONS

Such results demonstrate that our approach with new measurements and new features is effective and promising in brain state analysis and ASD diagnosis.

Computing personalized brain functional networks from fMRI using self-supervised deep learning

A novel self-supervised deep learning (DL) method is developed to compute personalized brain functional networks (FNs) for characterizing brain functional neuroanatomy based on functional MRI (fMRI). Specifically, a DL model of convolutional neural networks with an encoder-decoder architecture is developed to compute personalized FNs directly from fMRI data. The DL model is trained to optimize functional homogeneity of personalized FNs without utilizing any external supervision in an end-to-end fashion. We demonstrate that a DL model trained on fMRI scans from the Human Connectome Project can identify personalized FNs and generalizes well across four different datasets. We further demonstrate that the identified personalized FNs are informative for predicting individual differences in behavior, brain development, and schizophrenia status. Taken together, the self-supervised DL allows for rapid, generalizable computation of personalized FNs.

Unsupervised cross-domain functional MRI adaptation for automated major depressive disorder identification

Resting-state functional magnetic resonance imaging (rs-fMRI) data have been widely used for automated diagnosis of brain disorders such as major depressive disorder (MDD) to assist in timely intervention. Multi-site fMRI data have been increasingly employed to augment sample size and improve statistical power for investigating MDD. However, previous studies usually suffer from significant inter-site heterogeneity caused for instance by differences in scanners and/or scanning protocols. To address this issue, we develop a novel discrepancy-based unsupervised cross-domain fMRI adaptation framework (called UFA-Net) for automated MDD identification. The proposed UFA-Net is designed to model spatio-temporal fMRI patterns of labeled source and unlabeled target samples via an attention-guided graph convolution module, and also leverage a maximum mean discrepancy constrained module for unsupervised cross-site feature alignment between two domains. To the best of our knowledge, this is one of the first attempts to explore unsupervised rs-fMRI adaptation for cross-site MDD identification. Extensive evaluation on 681 subjects from two imaging sites shows that the proposed method outperforms several state-of-the-art methods. Our method helps localize disease-associated functional connectivity abnormalities and is therefore well interpretable and can facilitate fMRI-based analysis of MDD in clinical practice.

Hierarchical Individual Naturalistic Functional Brain Networks with Group Consistency uncovered by a Two-Stage NAS-Volumetric Sparse DBN Framework

The functional magnetic resonance imaging under naturalistic paradigm (NfMRI) showed great advantages in identifying complex and interactive functional brain networks due to its dynamics and multimodal information. In recent years, various deep learning models, such as deep convolutional autoencoder (DCAE), deep belief network (DBN) and volumetric sparse deep belief network (vsDBN), can obtain hierarchical functional brain networks (FBN) and temporal features from fMRI data. Among them, the vsDBN model revealed a good capability in identifying hierarchical FBNs by modelling fMRI volume images. However, due to the high dimensionality of fMRI volumes and the diverse training parameters of deep learning methods, especially the network architecture that is the most critical parameter for uncovering the hierarchical organization of human brain function, researchers still face challenges in designing an appropriate deep learning framework with automatic network architecture optimization to model volumetric NfMRI. In addition, most of the existing deep learning models ignore the group-wise consistency and inter-subject variation properties embedded in NfMRI volumes. To solve these problems, we proposed a two-stage neural architecture search and vs DBN model (two-stage NAS-vsDBN model) to identify the hierarchical human brain spatio-temporal features possessing both group-consistency and individual-uniqueness under naturalistic condition. Moreover, our model defined reliable network structure for modelling volumetric NfMRI data via NAS framework, and the group-level and individual-level FBNs and associated temporal features exhibited great consistency. In general, our method well identified the hierarchical temporal and spatial features of the brain function and revealed the crucial properties of neural processes under natural viewing condition.Significance StatementIn this paper, we proposed and applied a novel analytical strategy - a two-stage NAS-vsDBN model to identify both group-level and individual-level spatio-temporal features at multi-scales from volumetric NfMRI data. The proposed PSO-based NAS framework can find optimal neural structure for both group-wise and individual-level vs-DBN models. Furthermore, with well-established correspondence between two stages of vsDBN models, our model can effectively detect group-level FBNs that reveal the consistency in neural processes across subjects and individual-level FBNs that maintain the subject specific variability, verifying the inherent property of brain function under naturalistic condition.

Deep Learning and Medical Image Analysis for COVID-19 Diagnosis and Prediction

The coronavirus disease 2019 (COVID-19) pandemic has imposed dramatic challenges to health-care organizations worldwide. To combat the global crisis, the use of thoracic imaging has played a major role in diagnosis, prediction, and management for COVID-19 patients with moderate to severe symptoms or with evidence of worsening respiratory status. In response, the medical image analysis community acted quickly to develop and disseminate deep learning models and tools to meet the urgent need of managing and interpreting large amounts of COVID-19 imaging data. This review aims to not only summarize existing deep learning and medical image analysis methods but also offer in-depth discussions and recommendations for future investigations. We believe that the wide availability of high-quality, curated, and benchmarked COVID-19 imaging data sets offers the great promise of a transformative test bed to develop, validate, and disseminate novel deep learning methods in the frontiers of data science and artificial intelligence. Expected final online publication date for the Annual Review of Biomedical Engineering, Volume 24 is June 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Hierarchical Spatio-Temporal Modeling of Naturalistic Functional Magnetic Resonance Imaging Signals via Two-Stage Deep Belief Network With Neural Architecture Search

Naturalistic functional magnetic resonance imaging (NfMRI) has become an effective tool to study brain functional activities in real-life context, which reduces the anxiety or boredom due to difficult or repetitive tasks and avoids the problem of unreliable collection of brain activity caused by the subjects’ microsleeps during resting state. Recent studies have made efforts on characterizing the brain’s hierarchical organizations from fMRI data by various deep learning models. However, most of those models have ignored the properties of group-wise consistency and inter-subject difference in brain function under naturalistic paradigm. Another critical issue is how to determine the optimal neural architecture of deep learning models, as manual design of neural architecture is time-consuming and less reliable. To tackle these problems, we proposed a two-stage deep belief network (DBN) with neural architecture search (NAS) combined framework (two-stage NAS-DBN) to model both the group-consistent and individual-specific naturalistic functional brain networks (FBNs), which reflected the hierarchical organization of brain function and the nature of brain functional activities under naturalistic paradigm. Moreover, the test-retest reliability and spatial overlap rate of the FBNs identified by our model reveal better performance than that of widely used traditional methods. In general, our model provides a promising method for characterizing hierarchical spatiotemporal features under the natural paradigm.

An O-shape Neural Network With Attention Modules to Detect Junctions in Biomedical Images Without Segmentation

Junction plays an important role in biomedical research such as retinal biometric identification, retinal image registration, eye-related disease diagnosis and neuron reconstruction. However, junction detection in original biomedical images is extremely challenging. For example, retinal images contain many tiny blood vessels with complicated structures and low contrast, which makes it challenging to detect junctions. In this paper, we propose an O-shape Network architecture with Attention modules (Attention O-Net), which includes Junction Detection Branch (JDB) and Local Enhancement Branch (LEB) to detect junctions in biomedical images without segmentation. In JDB, the heatmap indicating the probabilities of junctions is estimated and followed by choosing the positions with the local highest value as the junctions, whereas it is challenging to detect junctions when the images contain weak filament signals. Therefore, LEB is constructed to enhance the thin branch foreground and make the network pay more attention to the regions with low contrast, which is helpful to alleviate the imbalance of the foreground between thin and thick branches and to detect the junctions of the thin branch. Furthermore, attention modules are utilized to introduce the feature maps from LEB to JDB, which can establish a complementary relationship and further integrate local features and contextual information between the two branches. The proposed method achieves the highest average F1-scores of 0.82, 0.73 and 0.94 in two retinal datasets and one neuron dataset, respectively. The experimental results confirm that Attention O-Net outperforms other state-of-the-art detection methods, and is helpful for retinal biometric identification.

Evolving fuzzy k-nearest neighbors using an enhanced sine cosine algorithm: Case study of lupus nephritis

Because of its simplicity and effectiveness, fuzzy K-nearest neighbors (FKNN) is widely used in literature. The parameters have an essential impact on the performance of FKNN. Hence, the parameters need to be attuned to suit different problems. Also, choosing more representative features can enhance the performance of FKNN. This research proposes an improved optimization technique based on the sine cosine algorithm (LSCA), which introduces a linear population size reduction mechanism for enhancing the original algorithm's performance. Moreover, we developed an FKNN model based on the LSCA, it simultaneously performs feature selection and parameter optimization. Firstly, the search performance of LSCA is verified on the IEEE CEC2017 benchmark test function compared to the classical and improved algorithms. Secondly, the validity of the LSCA-FKNN model is verified on three medical datasets. Finally, we used the proposed LSCA-FKNN to predict lupus nephritis classes, and the model showed competitive results. The paper will be supported by an online web service for any question at https://aliasgharheidari.com.