Application of artificial neural network in medical geochemistry.

For the evaluation of various adverse health effects of chemical elements occurring in the environment on humans, the comparison and linking of geochemical data (chemical composition of groundwater, soils, and dusts) with data on health status of population (so-called health indicators) play a key role. Geochemical and health data are predominantly nonlinear, and the use of standard statistical methods can lead to wrong conclusions. For linking such data, we find appropriate the use method of artificial neural networks (ANNs) which enable to eliminate data inhomogeneity and also potential data errors. Through method of ANNs, we are able to determine the order of influence of chemical elements on health indicators as well as to define limit values for the influential elements at which the health status of population is the most favourable (i.e. the lowest mortality, the highest life expectancy). For determination of dependence between the groundwater contents of chemical elements and health indicators, we recommend to create 200 ANNs. In further calculations performed for identification of order of influence of chemical elements as well as definition of limit values, we propose to work with median or mean values from calculated 200 ANNs. The ANN represents an appropriate method to be used for environmental and health data analysis in medical geochemistry.

Chemical composition of groundwater/drinking water and oncological disease mortality in Slovak Republic.

This study deals with the analysis of relationship between chemical composition of the groundwater/drinking water and the data on mortality from oncological diseases (MOD) in the Slovak Republic. Primary data consist of the Slovak national database of groundwater analyses (20,339 chemical analyses, 34 chemical elements/compounds) and data on MOD (17 health indicators) collected for the 10-year period (1994-2003). The chemical and health data were unified in the same form and expressed as the mean values for each of 2883 municipalities within the Slovak Republic. Pearson and Spearman correlation as well as artificial neural network (ANN) methods were used for analysis of the relationship between chemical composition of groundwater/drinking water and MOD. The most significant chemical elements having influence on MOD were identified together with their limit values (limit and optimal contents). Based on the results of calculations, made through the neural networks, the following eight chemical elements/parameters in the groundwater were defined as the most significant for MOD: Ca + Mg (mmol l-1), Ca, Mg, TDS, Cl, HCO3, SO4 and NO3. The results document the highest relationship between MOD and groundwater contents of Ca + Mg (mmol l-1), Ca and Mg. We observe increased MOD with low (deficit) contents of these three parameters of groundwater/drinking water. The following limit values were set for the most significant groundwater chemicals/parameters: Ca + Mg 1.73-5.85 mmol l-1, Ca 60.5-196.8 mg l-1 and Mg 25.6-35.8 mg l-1. At these concentration ranges, the mortality for oncological diseases in the Slovak Republic is at the lowest levels. These limit values are about twice higher in comparison with the current Slovak valid guideline values for the drinking water.

Chemical composition of groundwater and relative mortality for cardiovascular diseases in the Slovak Republic.

The study deals with the analysis of relationship between chemical composition of the groundwater/drinking water and the data on relative mortality for cardiovascular diseases (ReI) in the Slovak Republic. Primary data consist of the Slovak national database of groundwater analyses (20,339 chemical analyses, 34 chemical elements/compounds) and data on ReI collected for the 10-year period (1994-2003). The chemical and health data were unified in the same form and expressed as the mean values for each of 2883 municipalities within the Slovak Republic for further analysis. Artificial neural network was used as mathematic method for model data analysis. The most significant chemical elements having influence on ReI were identified together with their limit values (maximal acceptable, minimal necessary and optimal). Based on the results of calculations, made through the neural networks, the following ten chemical elements/parameters in the groundwater were defined as the most significant for ReI: Ca + Mg (mmol l(-1)), Ca, Mg, TDS, Cl, HCO3, SO4, NO3, SiO2 and PO4. The obtained results document the highest relationship between ReI and the groundwater contents of Ca + Mg (mmol l(-1)), Ca and Mg. Following limit values were set for the most significant groundwater chemicals/parameters: Ca + Mg 4.4-7.6 mmol l(-1), Ca > 89.4 mg l(-1) and Mg 42-78.1 mg l(-1). At these concentration ranges, the relative mortality for cardiovascular diseases in the Slovak Republic reaches the lowest levels. These limit values are about twice higher in comparison with the current Slovak valid guideline values for the drinking water.

Health risk estimates for groundwater and soil contamination in the Slovak Republic: a convenient tool for identification and mapping of risk areas.

We undertook a quantitative estimation of health risks to residents living in the Slovak Republic and exposed to contaminated groundwater (ingestion by adult population) and/or soils (ingestion by adult and child population). Potential risk areas were mapped to give a visual presentation at basic administrative units of the country (municipalities, districts, regions) for easy discussion with policy and decision-makers. The health risk estimates were calculated by US EPA methods, applying threshold values for chronic risk and non-threshold values for cancer risk. The potential health risk was evaluated for As, Ba, Cd, Cu, F, Hg, Mn, NO3 (-), Pb, Sb, Se and Zn for groundwater and As, B, Ba, Be, Cd, Cu, F, Hg, Mn, Mo, Ni, Pb, Sb, Se and Zn for soils. An increased health risk was identified mainly in historical mining areas highly contaminated by geogenic-anthropogenic sources (ore deposit occurrence, mining, metallurgy). Arsenic and antimony were the most significant elements in relation to health risks from groundwater and soil contamination in the Slovak Republic contributing a significant part of total chronic risk levels. Health risk estimation for soil contamination has highlighted the significance of exposure through soil ingestion in children. Increased cancer risks from groundwater and soil contamination by arsenic were noted in several municipalities and districts throughout the country in areas with significantly high arsenic levels in the environment. This approach to health risk estimations and visualization represents a fast, clear and convenient tool for delineation of risk areas at national and local levels.

The potential impact of geological environment on health status of residents of the Slovak Republic.

In order to assess the potential impact of the geological environment on the health of the population of the Slovak Republic, the geological environment was divided into eight major units: Paleozoic, Crystalline, Carbonatic Mesozoic and basal Paleogene, Carbonatic-silicate Mesozoic and Paleogene, Paleogene Flysch, Neovolcanics, Neogene and Quaternary sediments. Based on these geological units, the databases of environmental indicators (chemical elements/parameters in groundwater and soils) and health indicators (concerning health status and demographic development of the population) were compiled. The geological environment of the Neogene volcanics (andesites and basalts) has been clearly documented as having the least favourable impact on the health of Slovak population, while Paleogene Flysch geological environment (sandstones, shales, claystones) has the most favourable impact. The most significant differences between these two geological environments were observed, especially for the following health indicators: SMRI6364 (cerebral infarction and strokes) more than 70 %, SMRK (digestive system) 55 %, REI (circulatory system) and REE (endocrine and metabolic system) almost 40 % and REC (malignant neoplasms) more than 30 %. These results can likely be associated with deficit contents of Ca and Mg in groundwater from the Neogene volcanics that are only about half the level of Ca and Mg in groundwater of the Paleogene sediments.

Medical geochemistry research in Spissko-Gemerské rudohorie Mts., Slovakia.

This study presents an assessment of the potential impact of geological contamination of the environment on the health of the population in Spissko-Gemerské rudohorie Mts. (SGR Mts.). The concentration levels of potentially toxic elements (mainly As, Cd, Cu, Hg, Pb, Sb, and Zn) were determined in soils, groundwater, surface water, and stream sediments as well as in the food chain (locally grown vegetables). A medical study included some 30 health indicators for all 98 municipalities of the study area. The As and Sb contents in human fluids and tissues were analyzed in one municipality identified to be at the highest risk. Based on element content, environmental and health risks were calculated for respective municipalities. Out of 98 municipalities 14 were characterized with extremely high environmental risk and 10 were characterized with very high carcinogenic risk from arsenic (groundwater). Extensive statistical analysis of geochemical data (element contents in soils, groundwater, surface water, and stream sediments) and health indicators was performed. Significant correlations between element contents in the geological environment and health indicators, mainly cancer and cardiovascular diseases, were identified. Biological monitoring has confirmed the transfer of elements from the geological environment to human fluids and tissues as well as to the local food chain.