Characterizing autism spectrum disorder by deep learning spontaneous brain activity from functional near-infrared spectroscopy

Deep learning approach to predict autism spectrum disorder: a systematic review and meta-analysis

BACKGROUND

The use of the deep learning (DL) approach has been suggested or applied to identify childhood autism spectrum disorder (ASD). The capacity to predict ASD, however, differs across investigations. Our study's objective was to conduct a meta-analysis to determine the DL for ASD in children's classification accuracy.

METHODS

Eligibility criteria were designed according to the purpose of the meta-analysis; PubMed, EMBASE, Cochrane Library, and Web of Science Database were searched for articles published up to April 16, 2023, on the accuracy of DL methods for ASD classification. Using the Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) to assess the quality of the included studies. Sensitivity, specificity, areas under the curve (AUC), summary receiver operating characteristic (SROC), and corresponding 95% confidence intervals (CIs) were compiled by using the bivariate random-effects models.

RESULTS

A total of 11 predictive trials based on DL models were included, involving 9495 ASD patients from 6 different databases. According to bivariate random-effects models' results, the overall sensitivity, specificity, and AUC of the DL technique for ASD were, 0.95 (95% CI = 0.88-0.98), 0.93 (95% CI = 0.85-0.97), and 0.98 (95%CI: 0.97-0.99), respectively. Subgroup analysis results found that different datasets did not cause heterogeneity (meta-regression P = 0.55). The Kaggle dataset's sensitivity and specificity were 0.94 (95%CI: 0.82-1.00) and 0.91 (95%CI: 0.76-1.00), and with 0.97 (95%CI: 0.92-1.00) and 0.97 (95%CI: 0.92-1.00) for ABIDE dataset.

CONCLUSIONS

DL techniques has satisfactory sensitivity, specificity, and AUC in ASD classification. However, the major heterogeneity of the included studies limited the effectiveness of this meta-analysis. Further trials need to be performed to demonstrate the clinical practicability of DL diagnosis.

Advances and trends in the application of functional near-infrared spectroscopy for pediatric assessments: a bibliometric analysis

Objective

The objective is to elucidate the collaboration and current research status in the pediatric field of fNIRS using bibliometric analysis, and to discuss future directions.

Method

Bibliometric analysis was conducted on publications related to pediatric fNIRS research published before June 2024 in the Web of Science Core Collection using VOSviewer software and R language.

Results

A total of 761 documents were retrieved, published by 2,686 authors from 893 institutions across 44 countries in 239 journals. The number of publications has significantly increased since 2012. The United States is the country with the highest number of publications, University College London is the institution with the most publications, Lloyd-Fox Sarah is the author with the most publications and significant influence, and "Neurophotonics" is the journal with the most publications. The current hotspots mainly involve using fNIRS to study executive functions and autism spectrum disorders in children.

Conclusion

The study provides useful reference information for researchers by analyzing publication numbers, collaborative networks, publishing journals, and research hotspots. In the future, there should be an emphasis on enhancing interdisciplinary and international collaboration to collectively dedicate efforts toward the advancement of fNIRS technology and the standardization of research.

Accurate prediction of hyaluronic acid concentration under temperature perturbations using near-infrared spectroscopy and deep learning

Accurate prediction of the concentration of a large number of hyaluronic acid (HA) samples under temperature perturbations can facilitate the rapid determination of HA's appropriate applications. Near-infrared (NIR) spectroscopy analysis combined with deep learning presents an effective solution to this challenge, with current research in this area being scarce. Initially, we introduced a novel feature fusion method based on an intersection strategy and used two-dimensional correlation spectroscopy (2DCOS) and Aquaphotomics to interpret the interaction information in HA solutions reflected by the fused features. Subsequently, we created an innovative, multi-strategy improved Walrus Optimization Algorithm (MIWaOA) for parameter optimization of the deep extreme learning machine (DELM). The final constructed MIWaOA-DELM model demonstrated superior performance compared to partial least squares (PLS), extreme learning machine (ELM), DELM, and WaOA-DELM models. The results of this study can provide a reference for the quantitative analysis of biomacromolecules in complex systems.

Diagnostic machine learning applications on clinical populations using functional near infrared spectroscopy: a review

Functional near-infrared spectroscopy (fNIRS) and its interaction with machine learning (ML) is a popular research topic for the diagnostic classification of clinical disorders due to the lack of robust and objective biomarkers. This review provides an overview of research on psychiatric diseases by using fNIRS and ML. Article search was carried out and 45 studies were evaluated by considering their sample sizes, used features, ML methodology, and reported accuracy. To our best knowledge, this is the first review that reports diagnostic ML applications using fNIRS. We found that there has been an increasing trend to perform ML applications on fNIRS-based biomarker research since 2010. The most studied populations are schizophrenia (n = 12), attention deficit and hyperactivity disorder (n = 7), and autism spectrum disorder (n = 6) are the most studied populations. There is a significant negative correlation between sample size (>21) and accuracy values. Support vector machine (SVM) and deep learning (DL) approaches were the most popular classifier approaches (SVM = 20) (DL = 10). Eight of these studies recruited a number of participants more than 100 for classification. Concentration changes in oxy-hemoglobin (ΔHbO) based features were used more than concentration changes in deoxy-hemoglobin (ΔHb) based ones and the most popular ΔHbO-based features were mean ΔHbO (n = 11) and ΔHbO-based functional connections (n = 11). Using ML on fNIRS data might be a promising approach to reveal specific biomarkers for diagnostic classification.

Classification of autism based on short-term spontaneous hemodynamic fluctuations using an adaptive graph neural network

BACKGROUND

Short-term spontaneous hemodynamic fluctuations were collected by the functional near-infrared spectroscopy (fNIRS) system to classify children with autism spectrum disorder (ASD) and typical development (TD), and to explore abnormalities in the left inferior frontal gyrus in ASD.

METHODS

Using the fNIRS data of 25 children with ASD and 22 children with TD, a graph neural network combined with the temporal convolution module and the graph convolution module was used, to extract the spatio-temporal features of the data and achieve accurate classification of ASD.

RESULTS

The graph neural network was used to obtain a good classification result in the left inferior frontal gyrus, with an accuracy of 97.1%, precision of 95.1%, and specificity of 93.4%. It was found that the 5th channel (which is located in BA 10) and the 8th channel (which is located in BA 47) in the left inferior frontal gyrus were closely correlated with ASD.

COMPARISON WITH PREVIOUSLY USED METHOD(S)

Compared with the previous deep learning model using the same input, the accuracy of our model has increased by up to 13%, and the correlation between channels in the left inferior frontal gyrus area with the best classification effect was explored through the graph neural network.

CONCLUSION

The adaptive graph neural network (AGNN) model may be able to mine more valuable information to distinguish ASD from TD and in addition, the left inferior frontal gyrus may have greater investigative value.

Diagnosis of Chronic Musculoskeletal Pain by Using Functional Near-Infrared Spectroscopy and Machine Learning

Chronic pain (CP) has been found to cause significant alternations of the brain's structure and function due to changes in pain processing and disrupted cognitive functions, including with respect to the prefrontal cortex (PFC). However, until now, no studies have used a wearable, low-cost neuroimaging tool capable of performing functional near-infrared spectroscopy (fNIRS) to explore the functional alternations of the PFC and thus automatically achieve a clinical diagnosis of CP. In this case-control study, the pain characteristics of 19 chronic pain patients and 32 healthy controls were measured using fNIRS. Functional connectivity (FC), FC in the PFC, and spontaneous brain activity of the PFC were examined in the CP patients and compared to those of healthy controls (HCs). Then, leave-one-out cross-validation and machine learning algorithms were used to automatically achieve a diagnosis corresponding to a CP patient or an HC. The current study found significantly weaker FC, notably higher small-worldness properties of FC, and increased spontaneous brain activity during resting state within the PFC. Additionally, the resting-state fNIRS measurements exhibited excellent performance in identifying the chronic pain patients via supervised machine learning, achieving F1 score of 0.8229 using only seven features. It is expected that potential FC features can be identified, which can thus serve as a neural marker for the detection of CP using machine learning algorithms. Therefore, the present study will open a new avenue for the diagnosis of chronic musculoskeletal pain by using fNIRS and machine learning techniques.

Automatic autism spectrum disorder detection using artificial intelligence methods with MRI neuroimaging: A review

Autism spectrum disorder (ASD) is a brain condition characterized by diverse signs and symptoms that appear in early childhood. ASD is also associated with communication deficits and repetitive behavior in affected individuals. Various ASD detection methods have been developed, including neuroimaging modalities and psychological tests. Among these methods, magnetic resonance imaging (MRI) imaging modalities are of paramount importance to physicians. Clinicians rely on MRI modalities to diagnose ASD accurately. The MRI modalities are non-invasive methods that include functional (fMRI) and structural (sMRI) neuroimaging methods. However, diagnosing ASD with fMRI and sMRI for specialists is often laborious and time-consuming; therefore, several computer-aided design systems (CADS) based on artificial intelligence (AI) have been developed to assist specialist physicians. Conventional machine learning (ML) and deep learning (DL) are the most popular schemes of AI used for diagnosing ASD. This study aims to review the automated detection of ASD using AI. We review several CADS that have been developed using ML techniques for the automated diagnosis of ASD using MRI modalities. There has been very limited work on the use of DL techniques to develop automated diagnostic models for ASD. A summary of the studies developed using DL is provided in the Supplementary Appendix. Then, the challenges encountered during the automated diagnosis of ASD using MRI and AI techniques are described in detail. Additionally, a graphical comparison of studies using ML and DL to diagnose ASD automatically is discussed. We suggest future approaches to detecting ASDs using AI techniques and MRI neuroimaging.

Deep learning in fNIRS: a review

Significance: Optical neuroimaging has become a well-established clinical and research tool to monitor cortical activations in the human brain. It is notable that outcomes of functional near-infrared spectroscopy (fNIRS) studies depend heavily on the data processing pipeline and classification model employed. Recently, deep learning (DL) methodologies have demonstrated fast and accurate performances in data processing and classification tasks across many biomedical fields. Aim: We aim to review the emerging DL applications in fNIRS studies. Approach: We first introduce some of the commonly used DL techniques. Then, the review summarizes current DL work in some of the most active areas of this field, including brain-computer interface, neuro-impairment diagnosis, and neuroscience discovery. Results: Of the 63 papers considered in this review, 32 report a comparative study of DL techniques to traditional machine learning techniques where 26 have been shown outperforming the latter in terms of the classification accuracy. In addition, eight studies also utilize DL to reduce the amount of preprocessing typically done with fNIRS data or increase the amount of data via data augmentation. Conclusions: The application of DL techniques to fNIRS studies has shown to mitigate many of the hurdles present in fNIRS studies such as lengthy data preprocessing or small sample sizes while achieving comparable or improved classification accuracy.

Deep-learning informed Kalman filtering for priori-free and real-time hemodynamics extraction in functional near-infrared spectroscopy

Separation of the physiological interferences and the neural hemodynamics has been a vitally important task in the realistic implementation of functional near-infrared spectroscopy (fNIRS). Although many efforts have been devoted, the established solutions to this issue additionally rely on priori information on the interferences and activation responses, such as time-frequency characteristics and spatial patterns, etc., also hindering the realization of real-time. To tackle the adversity, we herein propose a novel priori-free scheme for real-time physiological interference suppression. This method combines the robustness of deep-leaning-based interference characterization and adaptivity of Kalman filtering: a long short-term memory (LSTM) network is trained with the time-courses of the absorption perturbation baseline for interferences profiling, and successively, a Kalman filtering process is applied with reference to the noise prediction for real-time activation extraction. The proposed method is validated using both simulated dynamic data and in-vivo experiments, showing the comprehensively improved performance and promisingly appended superiority achieved in the purely data-driven way.

Detecting the content of the bright blue pigment in cream based on deep learning and near-infrared spectroscopy

The excessive content of additives in food is a radical problem that affects human health. However, traditional chemical methods are limited by a long cycle, low accuracy, and strong destructiveness, so a fast and accurate alternative is urgently needed. This paper proposes a prediction model introducing near-infrared spectroscopy and deep learning to perform fast and accurate non-destructive detection of artificial bright blue pigment in cream. The model results show that R² is 0.9638, RMSEP is 0.0157, and RPD is 4.4022. In the preprocessing part, this paper compares the traditional preprocessing methods (SNV, MSC, SG) horizontally and innovatively proposes the use of autoencoders to mitigate the dimensionality of data, which has immensely improved the follow-up prediction effect. In addition, it tries to perform regression prediction on spectral data and establish a fully connected convolutional neural network model through deep learning, whose result indicators prove better than those of traditional methods such as PLSR and MLR. When constructing the deep learning model, this paper applies knowledge evolution to compress the model to achieve a lower calculation cost and higher accuracy. Compared with the traditional methods, the model proposed in this paper has greater accuracy and higher speed with samples undamaged, which is worth popularizing.

The amplitude of fNIRS hemodynamic response in the visual cortex unmasks autistic traits in typically developing children

Autistic traits represent a continuum dimension across the population, with autism spectrum disorder (ASD) being the extreme end of the distribution. Accumulating evidence shows that neuroanatomical and neurofunctional profiles described in relatives of ASD individuals reflect an intermediate neurobiological pattern between the clinical population and healthy controls. This suggests that quantitative measures detecting autistic traits in the general population represent potential candidates for the development of biomarkers identifying early pathophysiological processes associated with ASD. Functional near-infrared spectroscopy (fNIRS) has been extensively employed to investigate neural development and function. In contrast, the potential of fNIRS to define reliable biomarkers of brain activity has been barely explored. Features of non-invasiveness, portability, ease of administration, and low-operating costs make fNIRS a suitable instrument to assess brain function for differential diagnosis, follow-up, analysis of treatment outcomes, and personalized medicine in several neurological conditions. Here, we introduce a novel standardized procedure with high entertaining value to measure hemodynamic responses (HDR) in the occipital cortex of adult subjects and children. We found that the variability of evoked HDR correlates with the autistic traits of children, assessed by the Autism-Spectrum Quotient. Interestingly, HDR amplitude was especially linked to social and communication features, representing the core symptoms of ASD. These findings establish a quick and easy strategy for measuring visually-evoked cortical activity with fNIRS that optimize the compliance of young subjects, setting the background for testing the diagnostic value of fNIRS visual measurements in the ASD clinical population.

Looking Back at the Next 40 Years of ASD Neuroscience Research

During the last 40 years, neuroscience has become one of the most central and most productive approaches to investigating autism. In this commentary, we assemble a group of established investigators and trainees to review key advances and anticipated developments in neuroscience research across five modalities most commonly employed in autism research: magnetic resonance imaging, functional near infrared spectroscopy, positron emission tomography, electroencephalography, and transcranial magnetic stimulation. Broadly, neuroscience research has provided important insights into brain systems involved in autism but not yet mechanistic understanding. Methodological advancements are expected to proffer deeper understanding of neural circuitry associated with function and dysfunction during the next 40 years.

Brain exosomes as minuscule information hub for Autism Spectrum Disorder

INTRODUCTION

Autism spectrum disorder (ASD) is a neurodevelopmental disorder initiating in the first three years of life. Early initiation of management therapies can significantly improve the health and quality of life of ASD subjects. Thus, indicating the need for suitable biomarkers for the early identification of ASD. Various biological domains were investigated in the quest for reliable biomarkers. However, most biomarkers are in the preliminary stage, and clinical validation is yet to be defined. Exosome based research gained momentum in various Central Nervous System disorders for biomarker identification. However, the utility and prospect of exosomes in ASD is still underexplored.

AREAS COVERED

In the present review, we summarized the biomarker discovery current status and the future of brain-specific exosomes in understanding pathophysiology and its potential as a biomarker. The studies reviewed herein were identified via systematic search (dated: June 2021) of PubMed using variations related to autism (ASD OR autism OR Autism spectrum disorder) AND exosomes AND/OR biomarkers.

EXPERT OPINION

As exosomess are highly relevant in brain disorders like ASD, direct access to brain tissue for molecular assessment is ethically impossible. Thus investigating the brain-derived exosomes would undoubtedly answer many unsolved aspects of the pathogenesis and provide reliable biomarkers.

Deep learning for neuroimaging-based diagnosis and rehabilitation of Autism Spectrum Disorder: A review

Accurate diagnosis of Autism Spectrum Disorder (ASD) followed by effective rehabilitation is essential for the management of this disorder. Artificial intelligence (AI) techniques can aid physicians to apply automatic diagnosis and rehabilitation procedures. AI techniques comprise traditional machine learning (ML) approaches and deep learning (DL) techniques. Conventional ML methods employ various feature extraction and classification techniques, but in DL, the process of feature extraction and classification is accomplished intelligently and integrally. DL methods for diagnosis of ASD have been focused on neuroimaging-based approaches. Neuroimaging techniques are non-invasive disease markers potentially useful for ASD diagnosis. Structural and functional neuroimaging techniques provide physicians substantial information about the structure (anatomy and structural connectivity) and function (activity and functional connectivity) of the brain. Due to the intricate structure and function of the brain, proposing optimum procedures for ASD diagnosis with neuroimaging data without exploiting powerful AI techniques like DL may be challenging. In this paper, studies conducted with the aid of DL networks to distinguish ASD are investigated. Rehabilitation tools provided for supporting ASD patients utilizing DL networks are also assessed. Finally, we will present important challenges in the automated detection and rehabilitation of ASD and propose some future works.

Narrowband Resting-State fNIRS Functional Connectivity in Autism Spectrum Disorder

Hemispheric asymmetry in the power spectrum of low-frequency spontaneous hemodynamic fluctuations has been previously observed in autism spectrum disorder (ASD). This observation may imply a specific narrow-frequency band in which individuals with ASD could show more significant alteration in resting-state functional connectivity (RSFC). To test this assumption, we evaluated narrowband RSFC at several frequencies for functional near-infrared spectroscopy signals recorded from the bilateral temporal lobes on 25 children with ASD and 22 typically developing (TD) children. In several narrow-frequency bands, we observed altered interhemispheric RSFC in ASD. However, in the band of 0.01–0.02 Hz, more mirrored channel pairs (or cortical sites) showed significantly weaker RSFC in the ASD group. Receiver operating characteristic analysis further demonstrated that RSFC in the narrowband of 0.01–0.02 Hz might have better differentiation ability between the ASD and TD groups. This may indicate that the narrowband RSFC could serve as a characteristic for the prediction of ASD.

Automated ASD detection using hybrid deep lightweight features extracted from EEG signals

BACKGROUND

Autism spectrum disorder is a common group of conditions affecting about one in 54 children. Electroencephalogram (EEG) signals from children with autism have a common morphological pattern which makes them distinguishable from normal EEG. We have used this type of signal to design and implement an automated autism detection model.

MATERIALS AND METHOD

We propose a hybrid lightweight deep feature extractor to obtain high classification performance. The system was designed and tested with a big EEG dataset that contained signals from autism patients and normal controls. (i) A new signal to image conversion model is presented in this paper. In this work, features are extracted from EEG signal using one-dimensional local binary pattern (1D_LBP) and the generated features are utilized as input of the short time Fourier transform (STFT) to generate spectrogram images. (ii) The deep features of the generated spectrogram images are extracted using a combination of pre-trained MobileNetV2, ShuffleNet, and SqueezeNet models. This method is named hybrid deep lightweight feature generator. (iii) A two-layered ReliefF algorithm is used for feature ranking and feature selection. (iv) The most discriminative features are fed to various shallow classifiers, developed using a 10-fold cross-validation strategy for automated autism detection.

RESULTS

A support vector machine (SVM) classifier reached 96.44% accuracy based on features from the proposed model.

CONCLUSIONS

The results strongly indicate that the proposed hybrid deep lightweight feature extractor is suitable for autism detection using EEG signals. The model is ready to serve as part of an adjunct tool that aids neurologists during autism diagnosis in medical centers.

Shedding light on neuroscience: Two decades of functional near-infrared spectroscopy applications and advances from a bibliometric perspective

Functional near-infrared spectroscopy (fNIRS) is a noninvasive optical brain-imaging technique that detects changes in hemoglobin concentration in the cerebral cortex. fNIRS devices are safe, silent, portable, robust against motion artifacts, and have good temporal resolution. fNIRS is reliable and trustworthy, as well as an alternative and a complement to other brain-imaging modalities, such as electroencephalography or functional magnetic resonance imaging. Given these advantages, fNIRS has become a well-established tool for neuroscience research, used not only for healthy cortical activity but also as a biomarker during clinical assessment in individuals with schizophrenia, major depressive disorder, bipolar disease, epilepsy, Alzheimer's disease, vascular dementia, and cancer screening. Owing to its wide applicability, studies on fNIRS have increased exponentially over the last two decades. In this study, scientific publications indexed in the Web of Science databases were collected and a bibliometric-type methodology was developed. For this purpose, a comprehensive science mapping analysis, including top-ranked authors, journals, institutions, countries, and co-occurring keywords network, was conducted. From a total of 2310 eligible documents, 6028 authors and 531 journals published fNIRS-related papers, Fallgatter published the highest number of articles and was the most cited author. University of Tübingen in Germany has produced the most trending papers since 2000. USA was the most prolific country with the most active institutions, followed by China, Japan, Germany, and South Korea. The results also revealed global trends in emerging areas of research, such as neurodevelopment, aging, and cognitive and emotional assessment.

Trends in artificial intelligence, machine learning, and chemometrics applied to chemical data

Artificial intelligence-based methods such as chemometrics, machine learning, and deep learning are promising tools that lead to a clearer and better understanding of data. Only with these tools, data can be used to its full extent, and the gained knowledge on processes, interactions, and characteristics of the sample is maximized. Therefore, scientists are developing data science tools mentioned above to automatically and accurately extract information from data and increase the application possibilities of the respective data in various fields. Accordingly, AI-based techniques were utilized for chemical data since the 1970s and this review paper focuses on the recent trends of chemometrics, machine learning, and deep learning for chemical and spectroscopic data in 2020. In this regard, inverse modeling, preprocessing methods, and data modeling applied to spectra and image data for various measurement techniques are discussed.

Deep learning model for ultrafast quantification of blood flow in diffuse correlation spectroscopy

Diffuse correlation spectroscopy (DCS) is increasingly used in the optical imaging field to assess blood flow in humans due to its non-invasive, real-time characteristics and its ability to provide label-free, bedside monitoring of blood flow changes. Previous DCS studies have utilized a traditional curve fitting of the analytical or Monte Carlo models to extract the blood flow changes, which are computationally demanding and less accurate when the signal to noise ratio decreases. Here, we present a deep learning model that eliminates this bottleneck by solving the inverse problem more than 2300% faster, with equivalent or improved accuracy compared to the nonlinear fitting with an analytical method. The proposed deep learning inverse model will enable real-time and accurate tissue blood flow quantification with the DCS technique.

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.