Developing a breast cancer screening method is very important to facilitate early breast cancer detection and treatment. Building a screening method using medical imaging modality that does not cause body tissue damage (non-invasive) and does not involve physical touch is challenging. Thermography, a non-invasive and non-contact cancer screening method, can detect tumors at an early stage even under precancerous conditions by observing temperature distribution in both breasts. The thermograms obtained on thermography can be interpreted using deep learning models such as convolutional neural networks (CNNs). CNNs can automatically classify breast thermograms into categories such as normal and abnormal. Despite their demostrated utility, CNNs have not been widely used in breast thermogram classification. In this study, we aimed to summarize the current work and progress in breast cancer detection based on thermography and CNNs. We first discuss of breast thermography potential in early breast cancer detection, providing an overview of the availability of breast thermal datasets together with publicly accessible. We also discuss characteristics of breast thermograms and the differences between healthy and cancerous thermographic patterns. Breast thermogram classification using a CNN model is described step by step including a simulation example illustrating feature learning. We cover most research related to the implementation of deep neural networks for breast thermogram classification and propose future research directions for developing representative datasets, feeding the segmented image, assigning a good kernel, and building a lightweight CNN model to improve CNN performance.

In general, most of the existing convolutional neural network (CNN)-based deep-learning models suffer from spatial-information loss and inadequate feature-representation issues. This is due to their inability to capture multiscale-context information and the exclusion of semantic information throughout the pooling operations. In the early layers of a CNN, the network encodes simple semantic representations, such as edges and corners, while, in the latter part of the CNN, the network encodes more complex semantic features, such as complex geometric shapes. Theoretically, it is better for a CNN to extract features from different levels of semantic representation because tasks such as classification and segmentation work better when both simple and complex feature maps are utilized. Hence, it is also crucial to embed multiscale capability throughout the network so that the various scales of the features can be optimally captured to represent the intended task. Multiscale representation enables the network to fuse low-level and high-level features from a restricted receptive field to enhance the deep-model performance. The main novelty of this review is the comprehensive novel taxonomy of multiscale-deep-learning methods, which includes details of several architectures and their strengths that have been implemented in the existing works. Predominantly, multiscale approaches in deep-learning networks can be classed into two categories: multiscale feature learning and multiscale feature fusion. Multiscale feature learning refers to the method of deriving feature maps by examining kernels over several sizes to collect a larger range of relevant features and predict the input images’ spatial mapping. Multiscale feature fusion uses features with different resolutions to find patterns over short and long distances, without a deep network. Additionally, several examples of the techniques are also discussed according to their applications in satellite imagery, medical imaging, agriculture, and industrial and manufacturing systems.

Diabetes mellitus (DM) is one of the major diseases that cause death worldwide and lead to complications of diabetic foot ulcers (DFU). Improper and late handling of a diabetic foot patient can result in an amputation of the patient’s foot. Early detection of DFU symptoms can be observed using thermal imaging with a computer-assisted classifier. Previous study of DFU detection using thermal image only achieved 97% of accuracy, and it has to be improved. This article proposes a novel framework for DFU classification based on thermal imaging using deep neural networks and decision fusion. Here, decision fusion combines the classification result from a parallel classifier. We used the convolutional neural network (CNN) model of ShuffleNet and MobileNetV2 as the baseline classifier. In developing the classifier model, firstly, the MobileNetV2 and ShuffleNet were trained using plantar thermogram datasets. Then, the classification results of those two models were fused using a novel decision fusion method to increase the accuracy rate. The proposed framework achieved 100% accuracy in classifying the DFU thermal images in binary classes of positive and negative cases. The accuracy of the proposed Decision Fusion (DF) was increased by about 3.4% from baseline ShuffleNet and MobileNetV2. Overall, the proposed framework outperformed in classifying the images compared with the state-of-the-art deep learning and the traditional machine-learning-based classifier.

This paper investigates the optimal training duration and the optimal power allocation for the training and the data transmission that maximize the ergodic sum rate in single-cell uplink massive MIMO with MMSE receivers. Our channel model assumes that each user experiences the same spatial channel correlation. The expression for the ergodic sum rate is obtained in the large system regime where the number of antennas (N) at the base station and the number of users (K) tend to infinity with a fixed ratio. Interestingly, we show that the optimal training duration is equal to K and independent of the spatial correlation. We also derive the optimal power allocation that in facts depends on the spatial correlation, the channel coherence interval, and the uplink SNR. We show that more energy should be allocated for the training if the data transmission duration (td) is less than K, and vice versa. Moreover, equal power allocation is optimal when td = K. We also obtain an approximation for the optimal power allocation that depends on the mean of the correlation matrix eigenvalues. Numerical simulations show that our results based on the large system approximation are accurate and applicable for finite-size systems. The simulations also show that (1) the resulting ergodic sum rates obtained by employing the optimal power allocation and its approximation are indistinguishable, and (2) the optimal power allocation obtained from the uncorrelated channel model can be applied to the cases involving the correlated channels with indiscernible penalties on the ergodic sum rates.

Device-to-device (D2D) communication is one of the key enabling technologies for 5G to support many connections efficiently. In a cooperative out-band D2D communication network, relay nodes have an essential role in receiving and forwarding information signals from source (S) to destination (D), making the system more reliable and with higher energy efficiency. Therefore, the relay selection method is very important in determining the best relay among existing relays to achieve maximum performance. This paper proposes a new Adaptive Relay Selection (ARS) scheme for cooperative out-band D2D networks based on channel gain value and transmission link distance. Firstly, the best relay is adaptively selected among N available relays (R) based on the maximum channel gain values between S to R if the channel gain-based signal-to-noise ratio (SNR) is greater than the distance-based SNR. Otherwise, the best relay is selected based on the minimum distance between S to R. We also analyzed an exact closed-form throughput and the total energy consumption required for a cooperative out-band D2D communication system using the Quantization-and-Forward (QF) protocol. The numerical results show that the proposed ARS scheme has a higher throughput than the two previous schemes: maximum channel gain and minimum distance. Furthermore, the proposed ARS can reduce the total energy consumption, which indirectly impacts the resulting energy efficiency level. The proposed ARS scheme achieves higher energy efficiency than the previous schemes, either by link distance or power allocation. So, the proposed ARS scheme for an out-band D2D communication network is an appropriate solution for the next generation of cellular communications.

The cooperative device-to-device (D2D) system is becoming one of the most popular communication models in wireless 5G technology that combines the D2D system with cooperative communication to extend coverage and increase system performance. However, interference is one of the factors of decreasing service quality in cooperative D2D systems because of adjacent users/clusters and excessive power usage, which also causes interference to the receiver. Interference management, such as power control, can overcome interference problems in the cooperative D2D communication system. Therefore, this paper proposes a new form of adaptive power control (APC) based on the low-energy adaptive clustering hierarchy (LEACH) protocol for interference management systems in cooperative D2D systems that other studies have not considered. For a fair comparison, this paper also modifies the conventional fixed power control (FPC) method using the LEACH protocol and the modified fixed power control (MFPC) used in the previous study and then considers the system without power control. The simulation results show that the proposed APC can significantly reduce interference in cooperative D2D communication compared to the MFPC and without power control. Moreover, the simulation results also show that concerning the signal-to-interference and noise ratio (SINR) to the outage probability and throughput, the proposed APC performs better than the MFPC and without power control. Thus, the proposed APC can effectively reduce interference in cooperative D2D systems.