A method to extract the retina characteristic points for the purpose of medical diagnosis of the human eye is presented in this research. The proposed method helps to make the primary decision about the illness faster and can be used on mobile devices. The algorithm is mostly based on the characteristic points (the so-called minutiae). These structures are commonly used in the biometric applications for fingerprint-based people recognition. In the case of the conducted research, this trait was used to differentiate healthy eyes from unhealthy ones. The methods were evaluated by appropriately implemented algorithms, showing promising results. Each solution was created with object-oriented programming language. The accuracy of the classification (healthy versus samples with pathological changes) was evaluated using four algorithms: k-Nearest Neighbors, k-Means and Support Vector Machines (SVM) with linear and third-degree polynomial as well as our own approach based on counting the minutiae number. Performance requirements were also checked, and it was verified that the computing power of modern mobile devices is sufficient to implement the proposed solution. The highest accuracy result was equal to 96,45% and was obtained with the third-degree polynomial SVM. This was a novel approach. For comparative purposes, we also implemented currently used solutions for image analysis - deep learning (DL) and Convolution Neural Networks (CNNs). Both medical and computer science backgrounds are presented in the work with the main methodology components to include image segmentation using the Gaussian Matched Filter, binarization by Local Entropy Thresholding and classification with the previously mentioned approaches.

One of the most important modules of computer systems is the one that is responsible for user safety. It was proven that simple passwords and logins cannot guarantee high efficiency and are easy to obtain by the hackers. The well-known alternative is identity recognition based on biometrics. In recent years, more interest was observed in iris as a biometrics trait. It was caused due to high efficiency and accuracy guaranteed by this measurable feature. The consequences of such interest are observable in the literature. There are multiple, diversified approaches proposed by different authors. However, neither of them uses discrete fast Fourier transform (DFFT) components to describe iris sample. In this work, the authors present their own approach to iris-based human identity recognition with DFFT components selected with principal component analysis algorithm. For classification, three algorithms were used—k-nearest neighbors, support vector machines and artificial neural networks. Performed tests have shown that satisfactory results can be obtained with the proposed method.

The K3M thinning algorithm is a general method for image data reduction by skeletonization. It had proved its feasibility in most cases as a reliable and robust solution in typical applications of thinning, particularly in preprocessing for optical character recognition. However, the algorithm had still some weak points. Since then K3M has been revised, addressing the best known drawbacks. This paper presents a modified version of the algorithm. A comparison is made with the original one and two other thinning approaches. The proposed modification, among other things, solves the main drawback of K3M, namely, the results of thinning an image after rotation with various angles.

Hard exudates are one of the most characteristic and dangerous signs of diabetic retinopathy. They can be marked during the routine ophthalmological examination and seen in color fundus photographs (i.e., using a fundus camera). The purpose of this paper is to introduce an algorithm that can extract pathological changes (i.e., hard exudates) in diabetic retinopathy. This was a retrospective, nonrandomized study. A total of 100 photos were included in the analysis—50 sick and 50 normal eyes. Small lesions in diabetic retinopathy could be automatically diagnosed by the system with an accuracy of 98%. During the experiments, the authors used classical image processing methods such as binarization or median filtration, and data was read from the d-Eye sensor. Sixty-seven patients (39 females and 28 males with ages ranging between 50 and 64) were examined. The results have shown that the proposed solution accuracy level equals 98%. Moreover, the algorithm returns correct classification decisions for high quality images and low quality samples. Furthermore, we consider taking retina photos using mobile phones rather than fundus cameras, which is more practical. The paper presents an innovative approach. The results are introduced and the algorithm is described.

The aim of this paper is to investigate the impact of selected data augmentation techniques on the learning performance of neural networks for dynamic signature verification. The paper investigates selected data augmentation techniques in deep learning for verification purpose of dynamic signature. Two neural networks were used as classifiers: MLP and LSTM-FCN. Investigation of five selected augmentation methods and experiments were performed on the open source signature database SVC2004. The authors tested both classifiers without augmentation and then with data augmentation for three extensions of the learning set and three sizes of the user database. They presented the results of the experiments in tabular form for each augmentation method. The results were compared with the existing dynamic signature verification methods and given in the paper.

Since November 2019, the COVID-19 Pandemic produced by Severe Acute Respiratory Syndrome Severe Coronavirus 2 (hereafter COVID-19) has caused approximately seven million deaths globally. Several studies have been conducted using technological tools to prevent infection, to prevent spread, to detect, to vaccinate, and to treat patients with COVID-19. This work focuses on identifying and analyzing machine learning (ML) algorithms used for detection (prediction and diagnosis), monitoring (treatment, hospitalization), and control (vaccination, medical prescription) of COVID-19 and its variants. This study is based on PRISMA methodology and combined bibliometric analysis through VOSviewer with a sample of 925 articles between 2019 and 2022 derived in the prioritization of 32 papers for analysis. Finally, this paper discusses the study’s findings, which are directions for applying ML to address COVID-19 and its variants.