Laszlo Kovacs

Bio: Laszlo Kovacs is an academic researcher from University of Debrecen. The author has contributed to research in topics: Majority rule & Fundus (eye). The author has an hindex of 6, co-authored 24 publications receiving 257 citations.

Generalizing the Majority Voting Scheme to Spatially Constrained Voting

Abstract: Generating ensembles from multiple individual classifiers is a popular approach to raise the accuracy of the decision. As a rule for decision making, majority voting is a usually applied model. In this paper, we generalize classical majority voting by incorporating probability terms pn,k to constrain the basic framework. These terms control whether a correct or false decision is made if k correct votes are present among the total number of n. This generalization is motivated by object detection problems, where the members of the ensemble are image processing algorithms giving their votes as pixels in the image domain. In this scenario, the terms pn,k can be specialized by a geometric constraint. Namely, the votes should fall inside a region matching the size and shape of the object to vote together. We give several theoretical results in this new model for both dependent and independent classifiers, whose individual accuracies may also differ. As a real world example, we present our ensemble-based system developed for the detection of the optic disc in retinal images. For this problem, experimental results are shown to demonstrate the characterization capability of this system. We also investigate how the generalized model can help us to improve an ensemble with extending it by adding a new algorithm.

Crystallographic Studies on α- and β-D-glucopyranosyl Formamide Analogues, Inhibitors of Glycogen Phosphorylase

Abstract: The catalytic site of glycogen phosphorylase (GP) is currently under investigation as a target for inhibition of hepatic glycogenolysis under high glucose conditions. Three D-glucopyranosyl analogues, C-(1-azido-α-D-glucopyranosyl) formamide, C-(1-acetamido-α-D-glucopyranosyl) formamide, and C-(1-hydroxy-β-D-glucopyranosyl) formamide, were recognised as moderate competitive inhibitors of muscle glycogen phosphorylase b (GPb) [with respect to α-D-glucose 1-phosphate (Glc-1-P)] with Ki values of 1.80 (±0.2) mM, 0.31 (±0.01) mM, and 0.88 (±0.04) mM, respectively. In order to elucidate the structural basis of inhibition, we determined the structure of muscle GPb complexed with the three compounds at 2.1, 2.06 and 2.0 A resolution, respectively. The complex structures revealed that the inhibitors can be accommodated in the catalytic site of T-state GPb with very little change of the tertiary structure, and provide a rationalisation for understanding potency of the inhibitors. The glucopyranose moiety m...

Graph based detection of optic disc and fovea in retinal images

Abstract: Diabetic retinopathy (DR) is the damage to the eye's retina that occurs with long-term diabetes, which can eventually lead to blindness. Screening programs for DR are being introduced, however, an important prerequisite for automation is the accurate localization of the main anatomical features in the image, notably the optic disc (OD) and the macula. A series of interesting algorithms have been proposed in the recent past and the performance is generally good, but each method has situations, where it fails. This paper presents a novel framework for automatic detection of optic disc and macula in retinal fundus images using a combination of different optic disc and macula detectors represented by a weighted complete graph. A node pruning procedure removes the worst vertices of the graph by satisfying predefined geometric constraints and get best possible detector outputs to be combined using a weighted average. The extensive tests have shown that combining the predictions of multiple detectors is more accurate than any of the individual detectors making up the ensemble.

Generalized weighted majority voting with an application to algorithms having spatial output

Abstract: In this paper we propose a method using a generalization of the weighted majority voting scheme to locate the optic disc (OD) in retinal images automatically. The location with the maximal sum of the weights of OD center candidates falling into a disc of radius predefined in the clinical protocol is chosen for optic disc. We have worked out a weighted voting scheme, where besides the weights, an additional (e.g. geometrical) condition has to be taken into account in making the final decision. We can achieve better overall performance with this generalized weighted voting system than with the weighted majority voting and each individual algorithm.

An ensemble-based system for automatic screening of diabetic retinopathy

Abstract: In this paper, an ensemble-based method for the screening of diabetic retinopathy (DR) is proposed. This approach is based on features extracted from the output of several retinal image processing algorithms, such as image-level (quality assessment, pre-screening, AM/FM), lesion-specific (microaneurysms, exudates) and anatomical (macula, optic disk) components. The actual decision about the presence of the disease is then made by an ensemble of machine learning classifiers. We have tested our approach on the publicly available Messidor database, where 90% sensitivity, 91% specificity and 90% accuracy and 0.989 AUC are achieved in a disease/no-disease setting. These results are highly competitive in this field and suggest that retinal image processing is a valid approach for automatic DR screening.

Automatic Detection of Optic Disc Based on PCA and Mathematical Morphology

Abstract: The algorithm proposed in this paper allows to automatically segment the optic disc from a fundus image. The goal is to facilitate the early detection of certain pathologies and to fully automate the process so as to avoid specialist intervention. The method proposed for the extraction of the optic disc contour is mainly based on mathematical morphology along with principal component analysis (PCA). It makes use of different operations such as generalized distance function (GDF), a variant of the watershed transformation, the stochastic watershed, and geodesic transformations. The input of the segmentation method is obtained through PCA. The purpose of using PCA is to achieve the grey-scale image that better represents the original RGB image. The implemented algorithm has been validated on five public databases obtaining promising results. The average values obtained (a Jaccard's and Dice's coefficients of 0.8200 and 0.8932, respectively, an accuracy of 0.9947, and a true positive and false positive fractions of 0.9275 and 0.0036) demonstrate that this method is a robust tool for the automatic segmentation of the optic disc. Moreover, it is fairly reliable since it works properly on databases with a large degree of variability and improves the results of other state-of-the-art methods.

Stock market prediction using machine learning classifiers and social media, news

Abstract: Accurate stock market prediction is of great interest to investors; however, stock markets are driven by volatile factors such as microblogs and news that make it hard to predict stock market index based on merely the historical data. The enormous stock market volatility emphasizes the need to effectively assess the role of external factors in stock prediction. Stock markets can be predicted using machine learning algorithms on information contained in social media and financial news, as this data can change investors’ behavior. In this paper, we use algorithms on social media and financial news data to discover the impact of this data on stock market prediction accuracy for ten subsequent days. For improving performance and quality of predictions, feature selection and spam tweets reduction are performed on the data sets. Moreover, we perform experiments to find such stock markets that are difficult to predict and those that are more influenced by social media and financial news. We compare results of different algorithms to find a consistent classifier. Finally, for achieving maximum prediction accuracy, deep learning is used and some classifiers are ensembled. Our experimental results show that highest prediction accuracies of 80.53% and 75.16% are achieved using social media and financial news, respectively. We also show that New York and Red Hat stock markets are hard to predict, New York and IBM stocks are more influenced by social media, while London and Microsoft stocks by financial news. Random forest classifier is found to be consistent and highest accuracy of 83.22% is achieved by its ensemble.