István Lázár

Retinal Microaneurysm Detection Through Local Rotating Cross-Section Profile Analysis

A method for the automatic detection of microaneurysms (MAs) in color retinal images is proposed in this paper. The recognition of MAs is an essential step in the diagnosis and grading of diabetic retinopathy. The proposed method realizes MA detection through the analysis of directional cross-section profiles centered on the local maximum pixels of the preprocessed image. Peak detection is applied on each profile, and a set of attributes regarding the size, height, and shape of the peak are calculated subsequently. The statistical measures of these attribute values as the orientation of the cross-section changes constitute the feature set that is used in a naïve Bayes classification to exclude spurious candidates. We give a formula for the final score of the remaining candidates, which can be thresholded further for a binary output. The proposed method has been tested in the Retinopathy Online Challenge, where it proved to be competitive with the state-of-the-art approaches. We also present the experimental results for a private image set using the same classifier setup.

Minimal number of chromatographic test parameters for the characterisation of reversed-phase liquid chromatographic stationary phases

This paper focuses on the classification or differentiation of RP-HPLC columns based on measured chromatographic properties. A chemometric study has been conducted on a published data set consisting of 85 RP-HPLC columns and on a data set consisting of 47 self-tested columns. Principal component analysis enables determination of the number of parameters necessary for a rational differentiation. The results show that reducing the number of parameters for such differentiation still allows classification of the columns just as a higher number did. It is shown that three test parameters produce a classification similar to that obtained with five parameters.

Microaneurysm detection in retinal images using a rotating cross-section based model

Retinal image analysis is currently a very vivid field in biomedical image analysis. One of the most challenging tasks is the reliable automatic detection of microaneurysms (MAs). Computer systems that aid the automatic detection of diabetic retinopathy (DR) greatly rely on MA detection. In this paper, we present a method to construct an MA score map, from which the final MAs can be extracted by simple thresholding for a binary output, or by considering all the regional maxima to obtain probability scores. In contrary to most of the currently available MA detectors, the proposed one does not use any supervised training and classification. However, it is still competitive in the field, with a prominent performance in the detection of MAs close to the vasculature, regarding the state-of-the-art methods. The algorithm has been evaluated in a publicly available online challenge.

Formation and dissociation kinetics of the complexes Gd(DOTP)5- and Gd(DOTPMB)-

The monobutyl ester of H8DOTP, the ligand H4DOTPMB, was synthesized, and the protonation constants (KHi) and the stability constant of Gd(DOTPMB) were determined by pH-potentiometry (25 °C, 0.1 M Me4NCl): logKHi, (i = 1, 2, and 3) = 10.34(0.02), 7.72(0.025), and 2.42(0.030), respectively, and logKGdL = 12.19(0.05). The rates of formation of Gd(DOTPMB) and Gd(DOTP) were studied by 1H relaxometry in the pH range 5.4−7, and also by spectrophotometry in the case of Gd(DOTP) (7 < pH < 8). For both reactions, first-order rate constants were obtained at different concentration ratios of the reactants, which indicated the rapid formation of a reaction intermediate. The compositions of the intermediates are Gd(HiDOTPMB) (i = 1, 2) and HnGd(H2DOTP) (n = 0−4), respectively, where one or two protons are attached to the nitrogen atoms of the ligand. The rate of rearrangement (kr) of the intermediate Gd(HiDOTPMB) to the product Gd(DOTPMB) increases with increasing [OH−]: kr = kOH[OH−] + k2OH[OH−]2, where kOH = (1.3±0.25) ×103M−1s−1 and k2OH = (7.8±0.2) ×1011M−2s−1. For the formation reaction of Gd(DOTP), only the first term exists and kOH = (7.2±0.1) ×103M−1s−1. For the complexation reactions, similar mechanisms were proposed in which deprotonation of the species Gd(HDOTPMB) and HnGd(HDOTP) plays the rate-determining role. For the deprotonation, general base catalysis was found to be satisfactory. The rate of dissociation of Gd(DOTPMB) in 0.025−1.0 M HCl solution ([HCl] + [Me4NCl] = 1.0 M) was lower than that of Gd(DOTP) and the first-order rate constants exhibited saturation curves with increasing [H+]. Based on the assumption that the protonated species HGd(DOTPMB) and H5Gd(DOTP) dissociate, the rate constants (and protonation constants) were found to be (5.4±0.2) ×10−4M−1s−1 (KHGdL = 1.7±0.1) and (2.1±0.1) ×10−4M−1s−1 (KHH4GdL = 1.9±0.1), respectively.

Chelating Tendencies of Bioactive Aminophosphonates

The metal-binding abilities of a wide variety of bioactive aminophosphonates, from the simple aminoethanephosphonic acids to the rather large macrocyclic polyaza derivatives, are discussed with special emphasis on a comparison of the analogous carboxylic acid and phosphonic acid systems. Examples are given of the biological importance of metal ion – aminophosphonate interactions in living systems, and also of their actual and potential applicability in medicine.

NMR and potentiometric studies of 1,4,7-triazacyclononane-N,N′,N″-tris(methylenephosphonate monoethylester) and its complexes with metal ions

Abstract 1,4,7-Triazacyclononane- N , N ′, N ″-tris(methylene-phosphonate monoethylester) (NOTPME) and its complexes with the trivalent lanthanide cations and divalent calcium, magnesium, zinc and cadmium have been examined by potentiometry and NMR. The 31 P NMR spectra of the lanthanide-NOTPME complexes show single resonances characteristic of symmetrical chelated species at low concentrations (below 0.1 mM) but multiple resonances at higher concentrations. This tendency of the neutral complexes to form unsymmetrical, perhaps oligomeric, species is dependent upon the size of the Ln 3+ and the solution pH. The alkaline earth and divalent transition metal cation complexes do not show this same behavior. The stability constants of the Ln(NOTPME) complexes were found to be considerably smaller than their respective Ln(NOTA) complexes.

Automatic exudate detection using active contour model and regionwise classification

Diabetic retinopathy is one the most common cause of blindness in the world. Exudates are among the early signs of this disease, so its proper detection is a very important task to prevent consequent effects. In this paper, we propose a novel approach for exudate detection. First, we identify possible regions containing exudates using grayscale morphology. Then, we apply an active contour based method to minimize the Chan-Vese energy to extract accurate borders of the candidates. To remove those false candidates that have sufficient strong borders to pass the active contour method we use a regionwise classifier. Hence, we extract several shape features for each candidate and let a boosted Naïve Bayes classifier eliminate the false candidates. We considered the publicly available DiaretDB1 color fundus image set for testing, where the proposed method outperformed several state-of-the-art exudate detectors.

Integration of ground aerogel particles as chromatographic stationary phase into microchip

C16 modified and ground monolithic silica aerogel particles in submicrometer size, as a new type of stationary phase was prepared and integrated in polydimethylsiloxane (PDMS) microchip. The aerogel particles were packed into the microfluidic channel using a simple procedure, which does not require any special frit or fabrication step to retain the particles. The subnanoliter volume of samples can be transported through the porous, short length of packing with low pressure (< 3 bar). Food dyes as test components could be separated using low pressure within 6s. A 50-fold preconcentration could be achieved by retaining 100 nL volume of sample on the packing and elution with methanol.

Hybrid aerogel preparations as drug delivery matrices for low water-solubility drugs

A comprehensive study of 14 hybrid aerogels of different composition with applications in drug delivery has been carried out. The overall objective was to modulate the release behavior of drug-impregnated aerogels, from an almost instantaneous release to a semi-retarded delivery prolonged during several hours, through internal surface functionalization. The designed hybrid aerogels were composed of silica and gelatin and functionalized with either phenyl, long (16) hydrocarbon chain or methyl moiety. As model systems, three class II active agents (pKa<5.5), ibuprofen, ketoprofen and triflusal, were chosen to impregnate the aerogels. The work relied on the use of supercritical fluid technology for both the synthesis and functionalization of the hybrid aerogels, as well as for the impregnation with an active agent using supercritical CO2 as a solvent. For the impregnated aerogels, in vitro release profiles were recorded under gastric and intestinal pH-conditions using HPLC techniques. The release behavior observed for the three studied drugs was explained considering the measured dissolution profiles of the crystalline drugs, the aerogel composition and its functionalization. Such features are considered of great interest to tailor the bioavailability of drugs with low water solubility.

Segmentation of retinal vessels by means of directional response vector similarity and region growing

This paper presents a novel retinal vessel segmentation method. Opposed to the general approach in similar directional methods, where only the maximal or summed responses of a pixel are used, here, the directional responses of a pixel are considered as a vector. The segmentation method is a unique region growing procedure which combines a hysteresis thresholding scheme with the response vector similarity of adjacent pixels. A vessel score map is constructed as the combination of the statistical measures of the response vectors and its local maxima to provide the seeds for the region growing procedure. A nearest neighbor classifier based on a rotation invariant response vector similarity measure is used to filter the seed points. Many techniques in the literature that capture the Gaussian-like cross-section of vessels suffer from the drawback of giving false high responses to the steep intensity transitions at the boundary of the optic disc and bright lesions. To overcome this issue, we also propose a symmetry constrained multiscale matched filtering technique. The proposed vessel segmentation method has been tested on three publicly available image sets, where its performance proved to be competitive with the state-of-the-art and comparable to the accuracy of a human observer, as well.