Autism Spectrum Disorder, also known as ASD, is a neurodevelopmental disease that impairs speech, social interaction, and behavior. Machine learning is a field of artificial intelligence that focuses on creating algorithms that can learn patterns and make ASD classification based on input data. The results of using machine learning algorithms to categorize ASD have been inconsistent. More research is needed to improve the accuracy of the classification of ASD. To address this, deep learning such as 1D CNN has been proposed as an alternative for the classification of ASD detection. The proposed techniques are evaluated on publicly available three different ASD datasets (children, Adults, and adolescents). Results strongly suggest that 1D CNNs have shown improved accuracy in the classification of ASD compared to traditional machine learning algorithms, on all these datasets with higher accuracy of 99.45%, 98.66%, and 90% for Autistic Spectrum Disorder Screening in Data for Adults, Children, and Adolescents respectively as they are better suited for the analysis of time series data commonly used in the diagnosis of this disorder
A biconical antenna has been developed for ultra-wideband sensing. A wide impedance bandwidth of around 115% at bandwidth 3.73-14 GHz is achieved which shows that the proposed antenna exhibits a fairly sensitive sensor for microwave medical imaging applications. The sensor and instrumentation is used together with an improved version of delay and sum image reconstruction algorithm on both fatty and glandular breast phantoms. The relatively new imaging set-up provides robust reconstruction of complex permittivity profiles especially in glandular phantoms, producing results that are well matched to the geometries and composition of the tissues. Respectively, the signal-to-clutter and the signal-to-mean ratios of the improved method are consis
... Show MorePaintless colouring metals using lasers has gained more attention in aesthetic and functional applications. Nevertheless, the correlation between the processing parameters and the resulting colour remains complicated, nonlinear, and challenging to model. In this study data-driven machine learning (ML) technique was developed to predict the created colour by the laser colouring process of stainless-steel grade 304, in CIE L*, a*, b*, based on key processing parameters. The laser power (P), scanning speed (v), and number of passes (N) were employed in the colouring process. Four machine learning algorithms were evaluated, with the Random Forest (RF) model exhibiting the best performance. The assessment of model accuracy was conducted by the s
... Show MoreEmotion could be expressed through unimodal social behaviour’s or bimodal or it could be expressed through multimodal. This survey describes the background of facial emotion recognition and surveys the emotion recognition using visual modality. Some publicly available datasets are covered for performance evaluation. A summary of some of the research efforts to classify emotion using visual modality for the last five years from 2013 to 2018 is given in a tabular form.
Traffic‐induced ground vibrations cause significant problems for residents and nearby structures. Reducing the effect of these vibrations on the neighboring environment is a key challenge, particularly in urban areas. This study presents both numerical and experimental investigations of the performance of mass scatters for screening ground vibrations. A three‐dimensional numerical model is validated and extended to conduct a comparative study on the efficiency of three geotechnical methods of isolation. These methods include trench barriers, wave‐impeding blocks (WIBs), and mass scatters. The results showed that mass scatters represent an efficient way of scattering ground vi
The electrocardiogram (ECG) is the recording of the electrical potential of the heart versus time. The analysis of ECG signals has been widely used in cardiac pathology to detect heart disease. The ECGs are non-stationary signals which are often contaminated by different types of noises from different sources. In this study, simulated noise models were proposed for the power-line interference (PLI), electromyogram (EMG) noise, base line wander (BW), white Gaussian noise (WGN) and composite noise. For suppressing noises and extracting the efficient morphology of an ECG signal, various processing techniques have been recently proposed. In this paper, wavelet transform (WT) is performed for noisy ECG signals. The graphical user interface (GUI)
... Show More