Search

Summary: Cognitive radio (CR) has become an interesting research field that attracts researchers due to its increasing spectrum efficiency. Therefore, spectrum sensing (SS) is the essential function of cognitive radio systems. This paper presents an efficient SS model based on convolutional neural networks (CNNs). We use the spectrogram images of the received signals as the input to the CNN and use various images for signal and noise at different low primary‐user (PU) signal‐to‐noise ratios (SNRs) to train the network model. The model extracts the main features from the spectrogram images to represent signals and noise. Hence, this model can efficiently discriminate between signal and noise at different SNRs. The detection performance of the suggested model is compared with those of the traditional one‐stage, two‐stage SS methods, and different previous CNN models. The obtained outcomes demonstrate that the suggested model increases the detection accuracy more than those of the previous one‐stage SS methods by 17% at low SNRs of −20 dB and more than the previous two‐stage SS methods by 8% at low SNRs of −20 dB. In addition, it is demonstrated that the suggested model offers shorter sensing times than those of the two‐stage and one‐stage SS methods in the orders of 16.3, 16.6, 1.1, and 1.5 ms at SNRs of −20, −15, −10 and −5 dB, respectively. Furthermore, the proposed model improves the detection accuracy better than the different previously compared CNN models by 28% and 19% at low SNRs of −20 and −15 dB, respectively. [ABSTRACT FROM AUTHOR]

Deep learning is one of the most promising machine learning techniques that revolutionalized the artificial intelligence field. The known traditional and convolutional neural networks (CNNs) have been utilized in medical pattern recognition applications that depend on deep learning concepts. This is attributed to the importance of anomaly detection (AD) in automatic diagnosis systems. In this paper, the AD is performed on medical electroencephalography (EEG) signal spectrograms and medical corneal images for Internet of medical things (IoMT) systems. Deep learning based on the CNN models is employed for this task with training and testing phases. Each input image passes through a series of convolution layers with different kernel filters. For the classification task, pooling and fully‐connected layers are utilized. Computer simulation experiments reveal the success and superiority of the proposed models for automated medical diagnosis in IoMT systems. [ABSTRACT FROM AUTHOR]

Smart health surveillance technology has attracted wide attention between patients and professionals or specialists to provide early detection of critical abnormal situations without the need to be in direct contact with the patient. This paper presents a secure smart monitoring portable multivital signal system based on Internet-of-Things (IoT) technology. The implemented system is designed to measure the key health parameters: heart rate (HR), blood oxygen saturation (SpO2), and body temperature, simultaneously. The captured physiological signals are processed and encrypted using the Advanced Encryption Standard (AES) algorithm before sending them to the cloud. An ESP8266 integrated unit is used for processing, encryption, and providing connectivity to the cloud over Wi-Fi. On the other side, trusted medical organization servers receive and decrypt the measurements and display the values on the monitoring dashboard for the authorized specialists. The proposed system measurements are compared with a number of commercial medical devices. Results demonstrate that the measurements of the proposed system are within the 95% confidence interval. Moreover, Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Mean Relative Error (MRE) for the proposed system are calculated as 1.44, 1.12, and 0.012, respectively, for HR, 1.13, 0.92, and 0.009, respectively, for SpO2, and 0.13, 0.11, and 0.003, respectively, for body temperature. These results demonstrate the high accuracy and reliability of the proposed system. [ABSTRACT FROM AUTHOR]

Brain tumor is a mass of anomalous cells in the brain. Medical imagining techniques have a vital role in the diagnosis of brain tumors. Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) techniques are the most popular techniques to localize the tumor area. Brain tumor segmentation is very important for the diagnosis of tumors. In this paper, we introduce a framework to perform brain tumor segmentation, and then localize the region of the tumor, accurately. The proposed framework begins with the fusion of MR and CT images by the Non‐Sub‐Sampled Shearlet Transform (NSST) with the aid of the Modified Central Force Optimization (MCFO) to get the optimum fusion result from the quality metrics perspective. After that, image interpolation is applied to obtain a High‐Resolution (HR) image from the Low‐Resolution (LR) ones. The objective of the interpolation process is to enrich the details of the fusion result prior to segmentation. Finally, the threshold and the watershed segmentation are applied sequentially to localize the tumor region, clearly. The proposed framework enhances the efficiency of segmentation to help the specialists diagnose brain tumors. [ABSTRACT FROM AUTHOR]

Human identification is considered as a serious challenge for several applications such as cybersecurity and access control. Recently, the trend of human identification has been directed to human biometrics, which can be used to recognize persons based on some physiological or behavioral characteristics that they own, such as fingerprint, iris, and biosignals. There are several types of human biosignals including electroencephalography (EEG), electrocardiography (ECG), and photoplethysmography (PPG) signals. This paper presents a human identification system based on PPG signals. The proposed system consists of three main phases: signal acquisition, signal pre‐processing, and feature extraction/classification. The pre‐processing phase involves denoising of the acquired signal, transformation of the 1D signal sequence into a 2D image, and computation of the spectrogram. Feature extraction is carried out on the images obtained from the pre‐processing phase. Features are extracted from the images based on convolutional neural networks (CNNs). The proposed CNN model consists of a sequence of convolutional (CNV) and pooling layers. Finally, the obtained feature maps are fed to the classifier to discriminate human identities. The proposed identification algorithm is applied on signals with and without an additive white Gaussian noise (AWGN). The simulation results reveal that the proposed algorithm achieves an accuracy of 99.5% with the spectrogram representation and 89.8% with the 2D image representation, in the absence of noise. In addition, the paper gives a discussion of the efficiency of denoising techniques such as wavelet denoising, Savitzky–Golay and Kalman filtering, when involved with the proposed algorithm. The simulation results prove that the wavelet dencoising technique has a best performance among the discussed noise reduction techniques. [ABSTRACT FROM AUTHOR]

Automatic detection of maculopathy disease is a very important step to achieve high‐accuracy results for the early discovery of the disease to help ophthalmologists to treat patients. Manual detection of diabetic maculopathy needs much effort and time from ophthalmologists. Detection of exudates from retinal images is applied for the maculopathy disease diagnosis. The first proposed framework in this paper for retinal image classification begins with fuzzy preprocessing in order to improve the original image to enhance the contrast between the objects and the background. After that, image segmentation is performed through binarization of the image to extract both blood vessels and the optic disc and then remove them from the original image. A gradient process is performed on the retinal image after this removal process for discrimination between normal and abnormal cases. Histogram of the gradients is estimated, and consequently the cumulative histogram of gradients is obtained and compared with a threshold cumulative histogram at certain bins. To determine the threshold cumulative histogram, cumulative histograms of images with exudates and images without exudates are obtained and averaged for each type, and the threshold cumulative histogram is set as the average of both cumulative histograms. Certain histogram bins are selected and thresholded according to the estimated threshold cumulative histogram, and the results are used for retinal image classification. In the second framework in this paper, a Convolutional Neural Network (CNN) is utilized to classify normal and abnormal cases. [ABSTRACT FROM AUTHOR]