Diagnostic classification of autoantibody repertoires in endocrine ophthalmopathy using an artificial neural network.

PURPOSE: The aim of this study was to classify the human IgG autoantibody repertoire of sera from patients suffering from endocrine ophthalmopathy (EOP) and healthy subjects (CTRL) for diagnostic purposes using the recently developed Megablot technique. This technique allows for the simultaneous and quantitative screening of a large set of antigens and uses multivariate statistical techniques and an artificial neural network. METHODS: Sera were tested against Western blots (WBs) of SDS-PAGE preparations of proteins from human extraorbital eye muscle (EOP: n = 16; CTRL: n = 11). Digital image analysis was performed. The blots were subsequently analyzed by multivariate statistical techniques (analysis of discriminance) and an artificial neural network (probalistic neural network). RESULTS: The sera of both the EOP and CTRL groups showed a complex staining pattern against WBs of SDS-PAGEs from human eye muscle. Using the multivariate statistical technique for classification, all of the known samples and 85% of the unknown samples (not presented during calculation) were assigned to their correct clinical group. Using the artificial neural network as classifier, all of the samples presented during training and 96.3% of the unknown samples (not trained) were assigned correctly. CONCLUSIONS: The artificial neural network exceeds the ability of multivariate statistical techniques such as analysis of discriminance to assign unknown samples to their correct predefined group. Thus, the neural network exceeds other methods in generalizing some similarities of blots used for classification. This study reveals that our new technique and its evaluation using a neural network can be used as a helpful diagnostic tool in autoimmune diseases such as endocrine ophthalmopathy.

Analysis of tear-protein patterns as a diagnostic tool for the detection of dry eyes.

PURPOSE: To analyze the electrophoretic patterns of tear proteins from diabetic (DIDRY) and non-diabetic (DRY) patients with dry-eye disease in comparison with the patterns in healthy subjects (CTRL). The patterns were classified using multivariate statistical methods. METHODS: A total of 119 eyes were examined in this study (50 DIDRY, 39 DRY, 30 CTRL). The patients were primarily grouped according to the results of the basic secretory test (BST) and subjective symptoms (burning, foreign body sensations, tearing, "dryness" of the eyes). Patients with values < or = 10/5' plus subjective symptoms were classified as dry-eye patients. Tear proteins were separated by sodium-dodecyl-sulfate poly-acrylamide gel electrophoresis (SDS-PAGE). Digital image analysis was done using the ScanPacK (Biometra, Gottingen, Germany). A data set was created from each electrophoretic pattern. The data were analyzed by multivariate analysis of discriminance, k-means cluster analysis and with a factor analysis before k-means cluster analysis as a data reduction tool. RESULTS: The protein patterns of the three groups were significantly different (Wilks' lambda: 0.1425; P < 0.01): P < 0.05 (CTRL-DRY), P < 0.00005 (CTRL-DIDRY), and P < 0.0005 (DIDRY-DRY). There were more peaks/electrophoretic lane in the DRY and DIDRY groups (P < 0.05) than CTRL. Classification of all samples as DRY and CTRL gave the following results: known patterns 97% correct; unknown patterns 71.4% correct. Classification in DIDRY, DRY or CTRL was 92% correct for known patterns and 43% for unknown patterns. Using k-means cluster analysis, 72% of patients previously classified as "dry-eye" according to the BST value were classified as "dry-eye" based on their electrophoretic data too. CONCLUSIONS: Analysis of protein patterns and statistical evaluation are suitable for the detection of dry eyes. Tear-protein pattern analysis can provide more information on the pathogenesis of the disease. Thus, this new method might be more reliable than the BST value for both diagnostic and therapeutic purposes.

Iron status in a healthy population of Hungarian secondary school boys and girls.

Iron status was assessed in a sample of 103 male and 282 female students aged from 15 to 18 years attending secondary schools in Budapest. Using a ferritin model, in boys the prevalence of iron depletion, iron-deficient erythropoiesis and iron-deficient anemia was 3.9, 2.9 and 1.0%, respectively. In girls these values were 9.6, 8.2 and 2.8%, respectively. Males had better eating habits than females. Data for nutrients intake connected with iron status are discussed. The authors recommend the improvement of iron status by dietary means and oppose the indiscriminate iron supplementation.

Nutritional survey of pregnant women in Hungary.

A nutritional survey of a Hungarian group of pregnant women was carried out. One hundred and twenty nine women aged 25.9 years, entered the study, but only 70 completed all the protocol. Average body weight gain was 12.4 kg and the mean birth weight of the newborns was 3,386 g. Mean energy and nutrient intakes of pregnant women showed similar patterns as in Hungarian non-pregnant women of the same age. The mean energy intake was high (11 MJ), being 10% higher than for non-pregnant women. The mean protein and lipid intakes were also high, 91.9 g and 108.4 g, respectively, exceeding by 7% and 6% the intakes of the non-pregnant women. Dietary intakes of saturated fatty acids (SFAs) and monounsaturated fatty acids (MUFAs) were close to 12% of energy, and the intake of polyunsaturated fatty acids (PUFAs) was 7.6% of total energy. Palmitic acid (16:0), oleic acid (18:1 n-9) and linoleic acid (18:2 n-6) made the greatest contribution to the total peak area of SFAs, MUFAs, and PUFAs, respectively. The ratio of P/S (polyunsaturated/saturated fatty acids) was appropriate: 0.65; however, the ratio of linoleic acid (18:2 n-6), to linolenic acid (18:3 n-3) was high: 16. The cholesterol intake was somewhat lower in pregnant than in non-pregnant women. Complex carbohydrate intake was low (245 g) but it was still by 75% higher than in non-pregnant women. The excess sodium intake (6.3 g) was very similar to that of non-pregnant women. Mean values for retinol, tocopherol, ascorbic acid, cobalamin and copper intakes were higher than the Hungarian Recommended Dietary Allowances (RDA). Thiamine, riboflavin, pyridoxine, niacin, calcium, iron and zinc intakes were insufficient. Data showed an imbalance in the energy and nutrient intakes of Hungarian pregnant women, and this could be harmful for both the mother and pregnancy outcome.