Prediction and Analysis of Autism Spectrum Disorder Using Machine Learning Techniques.

Autism spectrum disorder is a severe, life-prolonged neurodevelopmental disease typified by disabilities that are chronic or limited in the development of socio-communication skills, thinking abilities, activities, and behavior. In children aged two to three years, the symptoms of autism are more evident and easier to recognize. The major part of the existing literature on autism spectrum disorder is covered by a prediction system based on traditional machine learning algorithms such as support vector machine, random forest, multiple layer perceptron, naive Bayes, convolution neural network, and deep neural network. The proposed models are validated by using performance measurement parameters such as accuracy, precision, and recall. In this research, autism spectrum disorder prediction has been investigated and compared using common parameters such as application type, simulation method, comparison methodology, and input data. The key purpose of this study is to give a centralized framework to use for researchers working on autism spectrum disorder prediction. The best results were obtained by using the random forest algorithm as it performs better than other traditional machine learning algorithms. The achieved accuracy is 89.23%. The workflow representations of the investigated frameworks assist readers in comprehending the fundamental workings and architectures of these frameworks.

Effects of Mutual Coupling on Gain and Beam Width of a Linear Array of a Dielectric Resonator Antenna for Main Beam Scanning Applications.

The effects of mutual coupling on the scanning characteristics of a four-element linear rectangular dielectric resonator antenna array (RDRA) are investigated for different inter-element spacing in this work. In particular, the gain and half-power beam width (HPBW) of an RDRA are studied for various scan angles in the E- and H-plane configurations. It is shown that for both the E and H planes, mutual coupling has an adverse effect on the performance of both phased array configurations. The H-plane array, however, is more stable than the E-plane array in terms of a gain and beam width performance comparison. The HPBW increases and gain decreases more in the E plane than the H plane when the scan angle is increased.

Robotics in neurosurgery: Current prevalence and future directions.

Background: The first instance of a robotic-assisted surgery occurred in neurosurgery; however, it is now more common in other fields such as urology and gynecology. This study aims to characterize the prevalence of robotic surgery among current neurosurgery programs as well as identify trends in clinical trials pertaining to robotic neurosurgery. Methods: Each institution's website was analyzed for the mention of a robotic neurosurgery program and procedures. The future potential of robotics in neurosurgery was assessed by searching for current clinical trials pertaining to neurosurgical robotic surgery. Results: Of the top 100 programs, 30 offer robotic cranial and 40 offer robotic spinal surgery. No significant differences were observed with robotic surgical offerings between geographic regions in the US. Larger programs (faculty size 16 or over) had 20 of the 30 robotic cranial programs (66.6%), whereas 21 of the 40 robotic spinal programs (52.5%) were at larger programs. An initial search of clinical trials revealed 223 studies, of which only 13 pertained to robotic neurosurgery. Spinal fixation was the most common intervention (six studies), followed by Deep Brain Stimulation (DBS, two studies), Cochlear implants (two studies), laser ablation (LITT, one study), and endovascular embolization (one study). Most studies had industry sponsors (9/13 studies), while only five studies had hospital sponsors. Conclusion: Robotic neurosurgery is still in its infancy with less than half of the top programs offering robotic procedures. Future directions for robotics in neurosurgery appear to be focused on increased automation of stereotactic procedures such as DBS and LITT and robot-assisted spinal surgery.

A CPW fed quad-port MIMO DRA for sub-6 GHz 5G applications.

The present work investigates a novel four-port, multiple-input multiple-output (MIMO), single element dielectric resonator antenna (DRA) for sub-6 GHz band. The DRA is designed and fabricated into a symmetric cross shape and fed using a coplanar waveguide (CPW) feed. A single radiator with four ports is rarely found in the literature. The -10 dB impedance bandwidth covered by the antenna is from 5.52 GHz to 6.2 GHz (11.6%) which covers fifth generation (5G) new radio (NR) bands N47 and wireless local area network (WLAN) IEEE 802.11a band. The isolation between orthogonal ports is about 15 dB while the isolation between opposite ports is 12 dB. The radiation pattern of the proposed antenna is bidirectional due to the absence of a ground plane below the DRA. The orthogonal modes excited in the DRA are [Formula: see text] and [Formula: see text] through the four symmetrical CPW feeds. The simulated and measured results of the proposed design show that MIMO characteristics are achieved by pattern diversity between the ports. Due to the perfect symmetry of the design, the proposed work could be extended to MIMO array applications as well.

Brain Decoding Using fMRI Images for Multiple Subjects through Deep Learning.

Substantial information related to human cerebral conditions can be decoded through various noninvasive evaluating techniques like fMRI. Exploration of the neuronal activity of the human brain can divulge the thoughts of a person like what the subject is perceiving, thinking, or visualizing. Furthermore, deep learning techniques can be used to decode the multifaceted patterns of the brain in response to external stimuli. Existing techniques are capable of exploring and classifying the thoughts of the human subject acquired by the fMRI imaging data. fMRI images are the volumetric imaging scans which are highly dimensional as well as require a lot of time for training when fed as an input in the deep learning network. However, the hassle for more efficient learning of highly dimensional high-level features in less training time and accurate interpretation of the brain voxels with less misclassification error is needed. In this research, we propose an improved CNN technique where features will be functionally aligned. The optimal features will be selected after dimensionality reduction. The highly dimensional feature vector will be transformed into low dimensional space for dimensionality reduction through autoadjusted weights and combination of best activation functions. Furthermore, we solve the problem of increased training time by using Swish activation function, making it denser and increasing efficiency of the model in less training time. Finally, the experimental results are evaluated and compared with other classifiers which demonstrated the supremacy of the proposed model in terms of accuracy.

Synthesis and evaluation of Ca-doped ferrihydrite as a novel adsorbent for the efficient removal of fluoride.

Ferric hydrate has been extensively applied for the removal of various types of pollutants from wastewater because of its low cost and high efficiency. However, its wide-scale application has been greatly restricted by high-dose and low-adsorption capacity. Therefore, a novel Ca-doped ferrihydrite adsorbent has been synthesized and used for the enhanced removal of fluoride from wastewater in the presence of other co-existing ions. At 5 mg/L initial fluoride concentration and pH 5, the removal efficiency of fluoride approached to 97.5% and remained stable. Similarly, with the increase of dose from 100 to 300 mg/L, the fluoride removal linearly increased to 98% and remained plateau at neutral pH. Also, the presence of co-existing ions such as NO3-, SO42-, Cl-, and natural organic matter has not significantly influenced the removal performance of the adsorbent. Fluoride removal best fit the pseudo-second-order reaction kinetics and Langmuir isotherm model. The prepared adsorbent exhibited a maximum adsorption capacity of 53.21 mg/g for fluoride uptake from water. The SEM-EDX confirmed the doping of Ca onto the ferrihydrite where the elemental peaks of Ca and Fe emerged at the energy value of about 3.6 Kev and 7.1 Kev respectively in EDX analysis. In addition, SEM results of Ca-doped ferrihydrite adsorbent illustrated that a large microplates type of products was acquired after synthesis. The regeneration results confirmed that adsorbent could retain their original adsorption capacity after five regeneration cycles. The current study suggested that Ca-doped ferrihydrite has the application potential for the enhanced adsorption of fluoride from the water phase.

An Improved Brain MRI Classification Methodology Based on Statistical Features and Machine Learning Algorithms.

In this paper, we have proposed a novel methodology based on statistical features and different machine learning algorithms. The proposed model can be divided into three main stages, namely, preprocessing, feature extraction, and classification. In the preprocessing stage, the median filter has been used in order to remove salt-and-pepper noise because MRI images are normally affected by this type of noise, the grayscale images are also converted to RGB images in this stage. In the preprocessing stage, the histogram equalization has also been used to enhance the quality of each RGB channel. In the feature extraction stage, the three channels, namely, red, green, and blue, are extracted from the RGB images and statistical measures, namely, mean, variance, skewness, kurtosis, entropy, energy, contrast, homogeneity, and correlation, are calculated for each channel; hence, a total of 27 features, 9 for each channel, are extracted from an RGB image. After the feature extraction stage, different machine learning algorithms, such as artificial neural network, k-nearest neighbors' algorithm, decision tree, and Naïve Bayes classifiers, have been applied in the classification stage on the features extracted in the feature extraction stage. We recorded the results with all these algorithms and found that the decision tree results are better as compared to the other classification algorithms which are applied on these features. Hence, we have considered decision tree for further processing. We have also compared the results of the proposed method with some well-known algorithms in terms of simplicity and accuracy; it was noted that the proposed method outshines the existing methods.

An Efficient Methodology for Brain MRI Classification Based on DWT and Convolutional Neural Network.

In this paper, a model based on discrete wavelet transform and convolutional neural network for brain MR image classification has been proposed. The proposed model is comprised of three main stages, namely preprocessing, feature extraction, and classification. In the preprocessing, the median filter has been applied to remove salt-and-pepper noise from the brain MRI images. In the discrete wavelet transform, discrete Harr wavelet transform has been used. In the proposed model, 3-level Harr wavelet decomposition has been applied on the images to remove low-level detail and reduce the size of the images. Next, the convolutional neural network has been used for classifying the brain MR images into normal and abnormal. The convolutional neural network is also a prevalent classification method and has been widely used in different areas. In this study, the convolutional neural network has been used for brain MRI classification. The proposed methodology has been applied to the standard dataset, and for performance evaluation, we have used different performance evaluation measures. The results indicate that the proposed method provides good results with 99% accuracy. The proposed method results are then presented for comparison with some state-of-the-art algorithms where simply the proposed method outperforms the counterpart algorithms. The proposed model has been developed to be used for practical applications.

Investigation of 9000 hours multi-stress aging effects on High-Temperature Vulcanized Silicone Rubber with silica (nano/micro) filler hybrid composite insulator.

Degradation in the polymeric insulators is caused due to the environmental stresses. The main aim of this paper is to explore the improved aging characteristics of hybrid samples by adding nano/micro silica in High Temperature Vulcanized Silicone Rubber (HTV-SiR) under long term accelerated aging conditions for 9000 hours. As HTV-SiR is unable to sustain environmental stresses for a long time, thus a long term accelerated aging behavior is an important phenomenon to be considered for field application. The aging characteristics of nano/micro filled HTV-SiR are analyzed by using techniques such as Scanning Electron Microscopy (SEM), Leakage Current (LC), Fourier Transform Infrared Microscopy (FTIR), Hydrophobicity Classification (HC), and breakdown strength for the aging time of 9000 hours. FTIR and leakage currents are measured after every cycle. All the co-filled samples revealed escalated aging characteristics as compared to the neat sample except the SN8 sample (8% nano-silica+20% micro-silica) after 9000 hours of aging. The highest loading of 6% and 8% nano-silica with 20% micro-silica do not contribute to the improved performance when compared with the neat and hybrid samples. However, from the critical experimental analysis, it is deduced that SN2 sample (2% nano-silica+20% micro-silica) is highly resistant to the long term accelerated aging conditions. SN2 has no cracks, lower loss percentages in the important FTIR absorption peaks, higher breakdown strength and superior HC after aging as compared to the unfilled and hybrid samples.

A Cost-Effective Electric Vehicle Intelligent Charge Scheduling Method for Commercial Smart Parking Lots Using a Simplified Convex Relaxation Technique.

Deployment of efficient and cost-effective parking lots is a known bottleneck for the electric vehicles (EVs) sector. A comprehensive solution incorporating the requirements of all key stakeholders is required. Taking up the challenge, we propose a real-time EV smart parking lot model to attain the following objectives: (a) maximize the smart parking lot revenue by accommodating maximum number of EVs and (b) minimize the cost of power consumption by participating in a demand response (DR) program offered by the utility since it is a tool to answer and handle the electric power usage requirements for charging the EV in the smart parking lot. With a view to achieving these objectives, a linear programming-based binary/cyclic (0/1) optimization technique is developed for the EV charge scheduling process. It is difficult to solve the problems of binary optimization in real-time given that the complexity of the problem increases with the increase in number of EV. We deploy a simplified convex relaxation technique integrated with the linear programming solution to overcome this problem. The algorithm achieves: minimum power consumption cost of the EV smart parking lot; efficient utilization of available power; maximization of the number of the EV to be charged; and minimum impact on the EV battery lifecycle. DR participation provide benefits by offering time-based and incentive-based hourly intelligent charging schedules for the EV. A thorough comparison is drawn with existing variable charging rate-based techniques in order to demonstrate the comparative validity of our proposed technique. The simulation results show that even under no DR event, the proposed scheme results in 2.9% decrease in overall power consumption cost for a 500 EV scenario when compared to variable charging rate method. Moreover, in similar conditions, such as no DR event and for 500 EV arrived per day, there is a 2.8% increase in number of EV charged per day, 3.2% improvement in the average state-of-charge (SoC) of the EV, 12.47% reduction in the average time intervals required to achieve final SoC.

Adaptive Filtering on GPS-Aided MEMS-IMU for Optimal Estimation of Ground Vehicle Trajectory.

Fusion of the Global Positioning System (GPS) and Inertial Navigation System (INS) for navigation of ground vehicles is an extensively researched topic for military and civilian applications. Micro-electro-mechanical-systems-based inertial measurement units (MEMS-IMU) are being widely used in numerous commercial applications due to their low cost; however, they are characterized by relatively poor accuracy when compared with more expensive counterparts. With a sudden boom in research and development of autonomous navigation technology for consumer vehicles, the need to enhance estimation accuracy and reliability has become critical, while aiming to deliver a cost-effective solution. Optimal fusion of commercially available, low-cost MEMS-IMU and the GPS may provide one such solution. Different variants of the Kalman filter have been proposed and implemented for integration of the GPS and the INS. This paper proposes a framework for the fusion of adaptive Kalman filters, based on Sage-Husa and strong tracking filtering algorithms, implemented on MEMS-IMU and the GPS for the case of a ground vehicle. The error models of the inertial sensors have also been implemented to achieve reliable and accurate estimations. Simulations have been carried out on actual navigation data from a test vehicle. Measurements were obtained using commercially available GPS receiver and MEMS-IMU. The solution was shown to enhance navigation accuracy when compared to conventional Kalman filter.

Multisensor-Based Target-Tracking Algorithm with Out-of-Sequence-Measurements in Cluttered Environments.

A localization and tracking algorithm for an early-warning tracking system based on the information fusion of Infrared (IR) sensor and Laser Detection and Ranging (LADAR) is proposed. The proposed Kalman filter scheme incorporates Out-of-Sequence Measurements (OOSMs) to address long-range, high-speed incoming targets to be tracked by networked Remote Observation Sites (ROS) in cluttered environments. The Rauch⁻Tung⁻Striebel (RTS) fixed lag smoothing algorithm is employed in the proposed technique to further improve tracking accuracy, which, in turn, is used for target profiling and efficient filter initialization at the targeted platform. This efficient initialization increases the probability of target engagement by increasing the distance at which it can be effectively engaged. The increased target engagement range also reduces risk of any damage from debris of the engaged target. Performance of the proposed target localization algorithm with OOSM and RTS smoothing is evaluated in terms of root mean square error (RMSE) for both position and velocity, which accurately depicts the improved performance of the proposed algorithm in comparison with existing retrodiction-based OOSM filtering algorithms. The effects of assisted target state initialization at the targeted platform are also evaluated in terms of Time to Impact (TTI) and true track retention, which also depict the advantage of the proposed strategy.