Daniel Zaborski

Examples of the Use of Data Mining Methods in Animal Breeding

Data mining techniques involve mainly searching for various relationships in large data sets. However, they can also be used in a much narrower range, sometimes as an alternative to classical statistics. The characteristic feature of these models is the use of a specific strategy, usually requiring the division of data into training set, sometimes also verification set, which enable the evaluation of the model quality as well as a test set for checking its prognostic or classification abilities. Among many different methods belonging to data mining, the following can be distinguished: the general models of classification and regression trees (G_Trees), general CHAID (Chi-square Automatic Interaction Detection) models, interactive classification and regression trees (also with boosting – Boosted Trees), random forest, MARS (Multivariate Adaptive Regression Splines), artificial neural networks (ANN), other machine learning methods such as: naive Bayes classifier (NBC), support vector machines (SVM), k-nearest neighbors (k-NN) and other regarded (or not) by different authors as data mining techniques. These methods are more and more frequently applied to various issues associated with animal breeding and husbandry.

Refractive status and ocular axial length in preterm infants without retinopathy of prematurity with regard to birth weight and gestational age.

Abstract Purpose: To obtain ultrasonographic measurements of ocular axial length (AL) in preterm infants without retinopathy of prematurity (ROP) but with different refractive power in regard to birth weight (BW) and gestational age (GA). Methods: Refraction was measured after cycloplegia (at 6 months of life) in 350 eyes of 180 preterm (non-astigmatic) infants without ROP. Subjects were grouped according to the refractive error: A [above –6.0 dioptres (D)]; B (–3.1 to –6.0 D); C (0 to –3.0 D); D (0.1 to +3.0 D); E (+3.1 to +6.0 D); F (above +6.0 D). The AL measurement was performed by ocular A-scan ultrasound biometry (10 MHz probe). Results: The longest AL was found in group B (20.62 mm) compared to group D and E (19.35, 19.28 mm; P≤0.01) and group F and A (19.63, 19.39 mm; P≤0.05). Only regressive correction for BW was statistically significant. Correlations between AL and BW (Rs=0.23) or GA (Rs=0.17) were found only in group E. Conclusions: AL of myopic eyes was significantly longer. In general, hyperopia was positively correlated with BW, whereas correlation between myopia and BW or GA was not found.

Quality of life in patients with venous stasis ulcers and others with advanced venous insufficiency.

The quality of life (QoL) in patients with advanced venous insufficiency (including venous stasis ulcers, skin discoloration, stasis eczema, and lipodermatosclerosis) assessed using the Clinical Etiological Anatomical Pathophysiological (CEAP) and Venous Clinical Severity Score (VCSS) classifications is presented. Also, disease features such as: intensity of pain, edema and inflammatory response that exerted the most profound effect on different domains of QoL are reported. The global QoL in patients with lower leg venous ulcerations was relatively similar to that observed in other patients with chronic venous insufficiency. The presence of venous ulcerations was associated with lower QoL in a Physical domain. Significant correlations were found between pain intensity and the values of Physical, Physiological, Level of Independence and Environmental domains, between edema intensity and Social domain as well as between the intensity of inflammatory response and Physical and Spiritual domains.

Effect of GO-Fe3O4 and rotating magnetic field on cellular metabolic activity of mammalian cells

The effect of hybrid material—graphene flakes with Fe3O4 nanospheres (GO-Fe3O4), graphene oxide (GO) and magnetite nanospheres (Fe3O4) in rotating magnetic field on mammalian cells metabolism has been studied. Several reports shown that exposure to magnetic field may have influence on cellular membrane permeability. Thus, the aim of presented study was to determine the cellular response of L929 fibroblast cells to nanomaterials and rotating magnetic field for 8-h exposure experiment. The GO had tendency to adsorb proteins, thus cell metabolism was decreased and the effect of that mechanism was enhanced by impact of nanospheres and rotating magnetic field. The highest reduction of cellular metabolism was recorded for WST-1 and NR assays at concentration 100 µg/mL of all tested nanomaterials and magnetic induction value 10.06 mT. The lactate dehydrogenase leakage assay has shown significant changes in membrane permeability. Further studies need to be carried out to precisely determine the mechanism of that process.

Detection of difficult calvings in the Polish Holstein-Friesian Black-and-White heifers

The aim of this study was to check the effectiveness of classification and regression trees (CART), support vector machines (SVM) and generalized linear model (GLZ) in the detection of difficult calvings in the Polish Holstein-Friesian Black-and-White heifers and to identify the most influential predictors of such calvings with special regard to oestrogen receptor alpha (ERα) and cytochrome P450 (CYP19) aromatase genotypes. A total of 529 calving records containing four continuous and four categorical predictors were analysed. Calving category (difficult vs. easy) was a dependent variable. The most significant predictors were as follows: gestation length, body condition score index and calving age (for CART), as well as ERα genotype (for GLZ). Sensitivity (percentage of correctly diagnosed difficult calvings) and specificity (percentage of correctly diagnosed easy calvings) on a test set were 70.0% and 94.7% (for CART) as well as 50.0% and 99.0% (for SVM and GLZ), respectively. The studied models were characterized by good detection performance and high specificity and reliability of predictions.

Detection of Difficult Calvings in Dairy Cows Using Boosted Classification Trees

The aim of this study was to apply boosted classification trees to dystocia detection in dairy cattle and to indicate its most important predictors. A total of 1742 calving records were used. Proportions of correctly detected difficult and easy calvings in a test set were 75.0%, 92.0% and 75.0%, 77.3%, for heifers and multiparous cows, respectively. The key predictors of calving difficulty were: pregnancy length, body condition score index, calving age, proportion of Holstein-Friesian genes (for heifers) as well as calving and calving-to-conception intervals, mean daily milk yield, fat and protein content in milk and 4%-fat corrected milk yield (for cows). The data mining method used in this study allowed us to obtain the models of good quality, confirmed by their relatively high effectiveness of detecting dystotic animals.

The PRNP ins/del and octapeptide repeat polymorphisms in Jersey cattle and their associations with production traits

So far the studies on the prion protein (PRNP) gene have mainly been focused on the search for an association between detected polymorphisms in this gene and the susceptibility of cattle to bovine spongiform encephalopathy. The PRNP gene location on chromosome 13 suggests that the polymorphisms within this gene can be potential quantitative trait locus (QTL) markers for milk production traits. The aim of this study was the analysis of the potential effect of the insertion/deletion polymorphisms within the promoter region (23 bp) and intron 1 (12 bp) of the PRNP gene on milk production traits in Jersey cattle. The study material consisted of 171 Jersey cows. The following milk production traits were analysed: milk yield, milk fat and protein yield, milk fat and protein content. Statistically significant differences were found in milk fat content in the second lactation between animal groups representing the ins/del and del/del genotypes for the 23 bp polymorphism (P≤0.05). The ins/del 12 bp polymorphism was found to have significant (P≤0.05) and highly significant (P≤0.01) effects on milk yield.

Recognition of culling reasons in Polish dairy cows using data mining methods

Data mining methods were used to recognize culling reasons in dairy cows.It is impossible to accurately recognize culling reasons using routinely recorded data.More specific data are required to improve the recognition ability of the models. Cow longevity and reasons for culling are one of the most important research problems in the contemporary cattle breeding. Therefore, the analysis of the relationship between cow performance and involuntary disposal contributes to taking more informed decisions in herd management. The aim of the present study was to compare the efficiency of artificial neural networks (ANN) and boosted classification trees (BT) with that of linear discriminant analysis (LDA) and classification functions (CF) in recognizing culling reasons of dairy cows in Poland, based on the lifetime performance data, routinely monitored in a herd. The analyses carried out in the present study showed that the accurate recognition of different culling reasons based on predictors included in the above-mentioned models is, in general, impossible. Only BT had limited discrimination abilities, but the results obtained using this method were not much improved compared with ANN and LDA with CF. In order to predict precisely various culling reasons, more specific data are required. They could be obtained from the increasingly popular, technologically advanced, systems of real-time monitoring of animal health status (physical activity, rumination rate, etc.), dependent also on environmental conditions (e.g. temperature-humidity index).

The role of lipid dysregulation and vascular risk factors in glaucomatous retrobulbar circulation.

The aim of this study was to evaluate selected lipid-related and vascular factors and their effect on retrobulbar hemodynamics in glaucoma. Fifty-six patients with primary open angle glaucoma (POAG) [POAG group; mean age 68.32 years (SD±0.21)] and 54 patients in control group [CG, mean age 68.1 years (SD±5.34)] were examined. Peak systolic velocity, end-diastolic velocity, mean velocity, pulsatility index, and resistive index of the ophthalmic artery, the central retinal artery and the posterior ciliary arteries were measured by Color Doppler Imaging. Selected lipid-related, systemic and local vascular parameters were evaluated. Statistical methods included Shapiro-Wilk, Student-t and Mann-Whitney U tests, and Spearman rank correlations. In POAG group systolic arterial blood pressure, diastolic arterial blood pressure, total cholesterol, low density lipoprotein cholesterol (LDL-ch), and intraocular pressure were significantly higher; while ocular perfusion pressure, high density lipoprotein cholesterol (HDL-ch) and diastolic ocular perfusion pressure were significantly lower (p≤0.05). Color Doppler Imaging confirmed blood flow abnormalities in all investigated arteries. In addition, significant correlations of HDL-ch, LDL-ch and triglycerides (TG) with peak systolic velocity, end-diastolic velocity and mean velocity were found in individual arteries (p≤0.05). Also, significant associations of systolic arterial blood pressure, ocular perfusion pressure, systolic oclular perfusion pressure and diastolic ocular perfusion pressure with peak systolic velocity, end-diastolic velocity, mean velocity and resistive index were revealed in the posterior ciliary arteries (p≤0.05). Dysregulation of lipid-related and vascular factors, as well as statistical correlation between the above and retrobulbar blood flow indices, might imply their role in vasoconstrictive processes during glaucomatous endotheliopathy.

Temperature changes in selected areas of body surface induced by systemic cryostimulation.

BACKGROUND The aim of this study was to assess the distribution and dynamics of temperature changes on the surface of selected body parts after systemic cryostimulation. The changes that occurred as a reaction to the 1st, 5th, and 10th session of a series of 10 sessions were also analyzed. METHODS The study group consisted of 24 students (12 women and 12 men, ∼21 yr of age) from the University School of Physical Education in Krakow. They were treated in a cryogenic chamber at the Rehabilitation Center in Krakow once daily for 10 d. The mean temperature in the chamber was -130°C ± 10°C and the session duration was 3 min. Thermovisual examination of temperature distribution in the selected parts of the upper and lower extremities was conducted before and immediately after a session on the 1(st) (S1), 5(th) (S2), and 10(th) (S3) day of treatment. All thermograms were digitally recorded using a Flir Therma CAM TM Sc500 camera. RESULTS On the 5th day of treatment, the examined group demonstrated the smallest mean temperature changes (4.57°C-17.31°C for the anterior part of the upper extremities in men and the posterior part of the lower extremities in women, respectively) before and after cryostimulation. The most significant temperature changes were observed in the group of women (6.80°C-20.08°C for the posterior parts of the upper extremities on S2 and the lower extremities on S3, respectively). CONCLUSION There is an important difference in response to cryogenic temperature between men and women in a series.