Fault type diagnosis is a very important tool to maintain the continuity of power transformer operation. Dissolved gas analysis (DGA) is one of the most effective and widely used techniques for predicting the power transformer fault types. In this paper, a convolutional neural network (CNN) model is proposed based on the DGA approach to accurately predict transformer fault types under different noise levels in measurements. The proposed model is applied with three categories of input ratios: conventional ratios (Rogers'4 ratios, IEC 60599 ratios, Duval triangle ratios), new ratios (five gas percentage ratios and new form six ratios), and hybrid ratios (conventional and new ratios together). The proposed model is trained and tested based on 589 dataset samples collected from electrical utilities and literature with varying noise levels up to ±20%. The results indicate that the CNN model with hybrid input ratios has superior prediction accuracy. The high accuracy of the proposed model is validated in comparison with conventional and recently published AI approaches. The proposed model is implemented based on MATLAB/toolbox 2020b.

Malaria is a contagious disease that affects millions of lives every year. Traditional diagnosis of malaria in laboratory requires an experienced person and careful inspection to discriminate healthy and infected red blood cells (RBCs). It is also very time-consuming and may produce inaccurate reports due to human errors. Cognitive computing and deep learning algorithms simulate human intelligence to make better human decisions in applications like sentiment analysis, speech recognition, face detection, disease detection, and prediction. Due to the advancement of cognitive computing and machine learning techniques, they are now widely used to detect and predict early disease symptoms in healthcare field. With the early prediction results, healthcare professionals can provide better decisions for patient diagnosis and treatment. Machine learning algorithms also aid the humans to process huge and complex medical datasets and then analyze them into clinical insights. This paper looks for leveraging deep learning algorithms for detecting a deadly disease, malaria, for mobile healthcare solution of patients building an effective mobile system. The objective of this paper is to show how deep learning architecture such as convolutional neural network (CNN) which can be useful in real-time malaria detection effectively and accurately from input images and to reduce manual labor with a mobile application. To this end, we evaluate the performance of a custom CNN model using a cyclical stochastic gradient descent (SGD) optimizer with an automatic learning rate finder and obtain an accuracy of 97.30% in classifying healthy and infected cell images with a high degree of precision and sensitivity. This outcome of the paper will facilitate microscopy diagnosis of malaria to a mobile application so that reliability of the treatment and lack of medical expertise can be solved.

Protecting patient privacy and medical records is a legal requirement. Traditional encryption methods fall short of handling the large volume of medical image data and their peculiar statistical properties. In this paper, we propose a generic medical image encryption framework based on a novel arrangement of two very efficient constructs, dynamic substitution boxes (S-boxes) and chaotic maps. The arrangement of S-box substitution before and after chaotic substitution is shown to successfully resist chosen plaintext and chosen ciphertext attacks. Special precautions are taken to fend off the reset attack against pseudorandom number generators. We show how to implement the generic framework using any key-dependent dynamic S-box construction method and any chaotic map. Experimental results show that the proposed framework successfully passes all security tests regardless of the chaotic map used for implementation. Based on speed analysis, we recommend the use of the classical Baker map or Henon map to achieve encryption throughput approaching 90 MB/s on a modern PC without hardware acceleration.

Image encryption schemes can be vulnerable to a variety of cryptanalysis attacks. The use of key-dependent dynamic S-boxes has been shown to improve security. Threats of chosen-plaintext and chosen-ciphertext attacks still linger. In this paper, we present an efficient algorithm for constructing secure dynamic S-boxes derived from Henon map. We use the proposed dynamic S-box to construct an image encryption scheme that includes a novel combination of security features to resist chosen-plaintext and chosen-ciphertext attacks. Namely, a hash verification step at the end of the decryption procedure effectively thwarts chosen-ciphertext attacks. The hash also serves as an image dependent initialization for the keystream, which together with using an image dependent S-box resist known-plaintext attacks. Furthermore, encryption keys are protected against cryptanalysis using elliptic curve cryptography (ECC).Therefore, the recovery of secret keys is as hard as the elliptic curve discrete logarithm problem even in the unlikely case of the recovery of the temporary S-box or keystream. Our evaluation of the proposed image encryption scheme reveals that it achieves a higher security standard than existing techniques. Moreover, the proposed scheme is computationally efficient with encryption throughput approaching 60 MB/s.

The development of a new dissolved gas analysis (DGA) method often requires a comparative study to assess the accuracy of the proposed technique. This is faced with the following challenges: (i) the time and effort required to implement and validate the implementation of existing DGA methods, adds to the comparative study cost; (ii) the output states of different DGA methods are not similar, which makes it difficult to put methods side by side in a comparative study; and (iii) the availability of test data is limited. In this study, a user‐friendly graphical user interface software package, DGALab, is developed to overcome these challenges. DGALab implements a unified DGA diagnosis framework to map the output states of DGA methods to uniform specifications. DGALab includes a library implementing most common DGA techniques, and includes a repository for input datasets available in the literature and collected directly from laboratories. DGALab simplifies the addition of new DGA techniques written in virtually any programming language. As a result, the process of developing a new DGA technique is greatly simplified using DGALab. To evaluate the software package results, the datasets and methods implemented therein were used to regenerate the results published in earlier research papers.