An optimized Kernel Extreme Learning Machine for the classification of the autism spectrum disorder by using gaze tracking images

Identification of autism spectrum disorder based on electroencephalography: A systematic review

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by difficulties in social communication and repetitive and stereotyped behaviors. According to the World Health Organization, about 1 in 100 children worldwide has autism. With the global prevalence of ASD, timely and accurate diagnosis has been essential in enhancing the intervention effectiveness for ASD children. Traditional ASD diagnostic methods rely on clinical observations and behavioral assessment, with the disadvantages of time-consuming and lack of objective biological indicators. Therefore, automated diagnostic methods based on machine learning and deep learning technologies have emerged and become significant since they can achieve more objective, efficient, and accurate ASD diagnosis. Electroencephalography (EEG) is an electrophysiological monitoring method that records changes in brain spontaneous potential activity, which is of great significance for identifying ASD children. By analyzing EEG data, it is possible to detect abnormal synchronous neuronal activity of ASD children. This paper gives a comprehensive review of the EEG-based ASD identification using traditional machine learning methods and deep learning approaches, including their merits and potential pitfalls. Additionally, it highlights the challenges and the opportunities ahead in search of more effective and efficient methods to automatically diagnose autism based on EEG signals, which aims to facilitate automated ASD identification.

Using visual attention estimation on videos for automated prediction of autism spectrum disorder and symptom severity in preschool children

Atypical visual attention in individuals with autism spectrum disorders (ASD) has been utilised as a unique diagnosis criterion in previous research. This paper presents a novel approach to the automatic and quantitative screening of ASD as well as symptom severity prediction in preschool children. We develop a novel computational pipeline that extracts learned features from a dynamic visual stimulus to classify ASD children and predict the level of ASD-related symptoms. Experimental results demonstrate promising performance that is superior to using handcrafted features and machine learning algorithms, in terms of evaluation metrics used in diagnostic tests. Using a leave-one-out cross-validation approach, we obtained an accuracy of 94.59%, a sensitivity of 100%, a specificity of 76.47% and an area under the receiver operating characteristic curve (AUC) of 96% for ASD classification. In addition, we obtained an accuracy of 94.74%, a sensitivity of 87.50%, a specificity of 100% and an AUC of 99% for ASD symptom severity prediction.

Oculomotor randomness is higher in autistic children and increases with the severity of symptoms

A variety of studies have suggested that at least some children with autism spectrum disorder (ASD) view the world differently. Differences in gaze patterns as measured by eye tracking have been demonstrated during visual exploration of images and natural viewing of movies with social content. Here we analyzed the temporal randomness of saccades and blinks during natural viewing of movies, inspired by a recent measure of "randomness" applied to micro-movements of the hand and head in ASD (Torres et al., 2013; Torres & Denisova, 2016). We analyzed a large eye-tracking dataset of 189 ASD and 41 typically developing (TD) children (1-11 years old) who watched three movie clips with social content, each repeated twice. We found that oculomotor measures of randomness, obtained from gamma parameters of inter-saccade intervals (ISI) and blink duration distributions, were significantly higher in the ASD group compared with the TD group and were correlated with the ADOS comparison score, reflecting increased "randomness" in more severe cases. Moreover, these measures of randomness decreased with age, as well as with higher cognitive scores in both groups and were consistent across repeated viewing of each movie clip. Highly "random" eye movements in ASD children could be associated with high "neural variability" or noise, poor sensory-motor control, or weak engagement with the movies. These findings could contribute to the future development of oculomotor biomarkers as part of an integrative diagnostic tool for ASD.

Hybrid Techniques of Facial Feature Image Analysis for Early Detection of Autism Spectrum Disorder Based on Combined CNN Features

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by difficulties in social communication and repetitive behaviors. The exact causes of ASD remain elusive and likely involve a combination of genetic, environmental, and neurobiological factors. Doctors often face challenges in accurately identifying ASD early due to its complex and diverse presentation. Early detection and intervention are crucial for improving outcomes for individuals with ASD. Early diagnosis allows for timely access to appropriate interventions, leading to better social and communication skills development. Artificial intelligence techniques, particularly facial feature extraction using machine learning algorithms, display promise in aiding the early detection of ASD. By analyzing facial expressions and subtle cues, AI models identify patterns associated with ASD features. This study developed various hybrid systems to diagnose facial feature images for an ASD dataset by combining convolutional neural network (CNN) features. The first approach utilized pre-trained VGG16, ResNet101, and MobileNet models. The second approach employed a hybrid technique that combined CNN models (VGG16, ResNet101, and MobileNet) with XGBoost and RF algorithms. The third strategy involved diagnosing ASD using XGBoost and an RF based on features of VGG-16-ResNet101, ResNet101-MobileNet, and VGG16-MobileNet models. Notably, the hybrid RF algorithm that utilized features from the VGG16-MobileNet models demonstrated superior performance, reached an AUC of 99.25%, an accuracy of 98.8%, a precision of 98.9%, a sensitivity of 99%, and a specificity of 99.1%.

Has Machine Learning Enhanced the Diagnosis of Autism Spectrum Disorder?

Autism spectrum disorder (ASD) is a complex neurological condition that limits an individual's capacity for communication and learning throughout their life. Although symptoms of Autism can be diagnosed in individuals of different ages, it is labeled as a developmental disorder because symptoms typically start to show up in the initial 2 years of childhood. Autism has no single known cause but multiple factors contribute to its etiology in children. Because symptoms and severity of ASD vary in every individual, there could be many causes. Detection of ASD in the early stages is crucial for providing a path for rehabilitation that enhances the quality of life and integrates the ASD person into the social, family, and professional spheres. Assessment of ASD includes experienced observers in neutral environments, which brings constraints and biases to a lack of credibility and fails to accurately reflect performance in terms of real-world scenarios. To get around these limitations, the conducted review offers a thorough analysis of the impact on the individual and the ones living around them and most recent research on how these techniques are implemented in the diagnosis of ASD. As a result of improvements in technology, assessments now include processing unconventional data than can be collected from measurements arising out of laboratory chemistry or of electrophysiological origin. Examples of these technologies include virtual reality and sensors including eye-tracking imaging. Studies have been conducted towards recognition of emotion and brain networks to identify functional connectivity and discriminate between people with ASD and people who are thought to be typically developing. Diagnosis of Autism has recently made substantial use of long short term memory (LSTM), convolutional neural network (CNN) and its variants, the random forest (RF) and naive Bayes (NB) machine learning techniques. It is hoped that researchers will develop methodologies that increase the probability of identification of ASD in its varied forms and contribute towards improved lifestyle for patients with ASD and those affected by the pathology.

The intelligent Traffic Management System for Emergency Medical Service Station Location and Allocation of Ambulances

In the present study, the optimization of medical services considering the role of intelligent traffic management is of concern. In this regard, a two-objective mathematical model of a medical emergency system is assessed in order to determine the location of emergency stations and determine the required number of ambulances to be allocated to the station. The objective functions are the maximization of covering the emergency demands and minimization of total costs. Moreover, the use of an intelligent traffic management system to speed up the ambulance is addressed. In this regard, the proposed two-objective mathematical model has been formulated, and a robust counterpart formulation under uncertainty is applied. In the proposed method, the values of the objective function increase as the problem becomes wider and, with a slight difference in large dimensions, converge in terms of the solution. The numerical results indicate that, as the problem's complexity increases, the robust optimization method is still effective because, with the increasing complexity of the problem, it can still solve large-scale problems in a reasonable time. Moreover, the difference between the value of the objective function in the proposed method and the presence of uncertainty parameters is very small and, in large dimensions, is quite logical and negligible. The sensitivity analysis shows that, with increasing demand, both the number of ambulances required and the amount of objective function have increased significantly.

Brent Crude Oil Price Forecast Utilizing Deep Neural Network Architectures

Brent crude oil is considered as one of the most important sources of crude oil pricing in the worldwide market, and it is used to set the price of two-thirds of the traded crude oil supplies in the world. To predict the price of Brent crude oil, LSTM and Bi-LSTM methods are applied, which are the architecture of the recursive neural network. Initially, the database creates the appropriate data for the period January 2015 to March 2021 from Brent crude oil price signals and daily data from a financial market, and then, the modeling process is performed via the use of MATLAB software. Also, about 90% of the data are for training and the remaining for validation and comparison. Using LSTM and Bi-LSTM neural networks, the network architecture has been worked on, and by adding the number of layers and changing the solvers (SGDM, RMSProp, and Adam), the errors of different models are compared with each other. Nonlinear techniques of artificial neural networks and deep learning were used for modeling. Then, the network architecture was worked on and the model error rate was evaluated by comparing different layers and solvents such as SGDM, RMSProp, and Adam. The superiority of SGDM solvent over others was shown, and finally, it can be mentioned as the superior method of modeling of price forecasting in Brent crude oil field. The results show that the model with two layers of LSTM and SGDM solver has less error and better accuracy.