In recent years, Convolutional Neural Networks (ConvNets) have rapidly emerged as a widespread machine learning technique in a number of applications especially in the area of medical image classification and segmentation. In this paper, we propose a novel approach that uses ConvNet for classifying brain medical images into healthy and unhealthy brain images. The unhealthy images of brain tumors are categorized also into low grades and high grades. In particular, we use the modified version of the Alex Krizhevsky network (AlexNet) deep learning architecture on magnetic resonance images as a potential tumor classification technique. The classification is performed on the whole image where the labels in the training set are at the image level rather than the pixel level. The results showed a reasonable performance in characterizing the brain medical images with an accuracy of 91.16%.

Biomedical telemetry is an emerging field of research which enables the formation of a transmission link from inside a living body to an external device. Implantable medical devices are now one of such valuable advancement in the field of biomedical telemetry. Implantable patch antennas are gaining attention and are becoming more of a choice for implantable medical devices that uses mostly RF telemetry. In this work, a state‐of‐the‐art design for a rectangular implantable flexible patch antenna is proposed. The operation band for the antenna is chosen in the Industrial, Scientific and Medical (ISM) band (2.4–2.4835 GHz). The tiny dimension of the antenna, including the 9.45 μm thickness of the patch itself allows the antenna to be highly flexible and provides excellent results even at extreme bent conditions. For simulation environment, a three‐layer human tissue model was used, where the antenna was encapsulated between the fat and skin layer. CST Microwave Studio was chosen to design and simulate the antenna. Several performance parameters were simulated, such as the operating resonant frequency, the return loss, radiation pattern, specific absorption rate and also sensitivity of the antenna when introduced to bending.

Taxonomic classification has a wide-range of applications such as finding out more about evolutionary history. Compared to the estimated number of organisms that nature harbors, humanity does not have a thorough comprehension of to which specific classes they belong. The classification of living organisms can be done in many machine learning techniques. However, in this study, this is performed using convolutional neural networks. Moreover, a DNA encoding technique is incorporated in the algorithm to increase performance and avoid misclassifications. The algorithm proposed outperformed the state of the art algorithms in terms of accuracy and sensitivity, which illustrates a high potential for using it in many other applications in genome analysis.

Voice over Internet Protocol (VoIP) networks have recently emerged as a promising telecommunication medium for transmitting voice signal. One of the essential aspects that interests researchers is how to estimate the quality of transmitted voice over VoIP for several purposes such as design and technical issues. Two methodologies are used to evaluate the voice, which are subjective and objective methods. In this study, the authors propose a non‐intrusive machine learning‐based (NIML) objective method to estimate the quality of voice. In particular, they build a training set of parameters – from the network and the voice itself – along with the quality of voices as labels. The voice quality is estimated using the perceptual evaluation of speech quality (PESQ) method as an intrusive algorithm. Then, the authors use a set of classifiers to build models for estimating the quality of the transmitted voice from the training set. The experimental results show that the classifier models have a valuable performance where Random Forest model has superior results compared to other models of precision 94.1%, recall 94.2%, and receiver operating characteristic area 99.2% as evaluation metrics.

Positron Emission Tomography scan images are extensively used in radiotherapy planning, clinical diagnosis, assessment of growth and treatment of a tumor. These all rely on fidelity and speed of detection and delineation algorithm. Despite intensive research, segmentation has remained a challenging problem due to the diverse image content, resolution, shape, and noise. This paper presents a fast positron emission tomography tumor segmentation method using superpixels. Principal component analysis is applied on the superpixels and their average value. The distance vector of each superpixel from the average is computed in the principal components coordinate system. Finally, k-means clustering is applied on the distance vector to recognize tumor and non-tumor superpixels. The proposed approach is implemented in MATLAB 2016A, and promising accuracy with execution time of 2.35 ± 0.26 s is achieved. Fast execution time is achieved since the number of superpixels, and the size of distance vector on which clustering was done are low compared to the number of pixels in the image.