Machine Learning to predict tuberculosis in cattle from the state of Sao Paulo, Brazil

Tuberculosis is a well-known and worldwide spread zoonosis. In Brazil 1.594.787 cases were confirmed cases since 2001, where, in Sao Paulo state, 8.226 deaths were reported. This study aims to present steps related to the use of machine learning algorithms for predictive analysis for bovine tuberculosis. For this, an application was made based on data from farms in state of São Paulo, Brazil, of an epidemiological survey, using a specific questionnaire, carried out on farms (n = 1,743). Response variable was presented by apparent prevalence of positive properties for disease, and predictors by (k = 77) predictors related to type of farm, type of lactation, number of animals on property. Application was organized according to following steps: division of data in training (75%) and testing (25%), pre-processing of predictors, learning and model evaluation. In the learning step, algorithm for adjusting gradient boosted trees models was used. The hyperparameters of algorithms were optimized by 10-fold cross-validation, to select those corresponding to best models. Models showed an accuracy of 88.07%, with an error in learning process equal to 3%. In the test / model validation procedure (n = 436), an error in 12% estimate was observed. Five important predictors were daily milk production, number of cows, type of farm, bovine breed and slaughter of adult animals. Proportion of false positives among all individuals whose response of interest was observed was 2.06%, and proportion of false negatives among those with a response of absent interest was 9.86%. It is hoped that, with increase in trained surveillance to detect the disease and availability of data, it will be possible to develop predictive models of machine learning with potential to efficiently assist professionals in disease control and assist in education program in animal health

Key messages

Predictive analyzes in health: application for tuberculosis in cattle from the state of Sao Paulo, Brazil. An infectious disease and zoonosis important to the world that needs support to develop means to control and consequently eradicate it.

Knowledge about human visceral leishmaniosis in an endemic area in Pernambuco, Brazil

The timely detection and adequate attention of cases of Human Visceral Leishmaniasis can provide a reduction in the morbidity and mortality of this disease. This study evaluated the degree of knowledge about Human Visceral Leishmaniasis among professionals from basic health units in the municipality of Caruaru, an endemic area for Visceral Leishmaniasis in the state of Pernambuco-Brazil. From June to August 2017, questionnaires were applied to doctors and nurses of the units that presented cases and / or deaths due to LVH between the years 2006 and 2015. Regarding the results, it was possible to verify, from the analysis of multiple correspondence, important interdependence relations such as the lack of knowledge of the endemicity of the Municipality in question with the misinformation about treatment, in addition to the association between not knowing the etiologic agent of the disease and the lack of knowledge about the treatment and the municipal health unit, suggesting an impediment to the control of LVH. The results also demonstrated a general lack of knowledge regarding the etiological, epidemiological, assistance and surveillance aspects of this condition in the city and the need to train these professionals on the subject, considering that knowledge about visceral leishmaniasis is considered a protective factor and capable to reduce the risk of this disease occurring. Therefore, we emphasize the need to maintain, review and restructure the control and prevention measures already known and implemented. However, it is necessary to strengthen a valuable and fundamental area in the fight against the disease, that of health education, focusing mainly on primary care professionals, who are still the main source of welcome for sick individuals. These are fundamental to reduce the morbidity and mortality of human VL, as a link in the transmission chain, but also as perpetuators of information about the disease in order to prevent it through attitudes.

Key messages

One of the main impacts of this work is the applicability and utility in health services in Brazil and other countries that face neglected diseases such as visceral leishmaniasis. Another great contribution of this work is to demonstrate, in a practical example, how education is fundamental in prevention and health promotion strategies.

Experimental toxoplasmosis in pregnant bitches

Three Toxoplasma gondii free bitches (No. 3-5) were fed 15000 sporulated T. gondii oocysts at 56, 40 and 32 day of gestation and the outcome of the pregnancy was monitored. Two of the three dogs infected during pregnancy showed evidence of congenital infection and one aborted. Two control bitches not fed oocysts delivered eight uninfected healthy pups. This study demonstrated that T. gondii can be congenitally transmitted in dogs when bitches are infected during pregnancy.