Application of artificial intelligence to optimize the diagnosis and treatment of acute coronary syndrome: a real experience

BACKGROUND/INTRODUCTION: cardiovascular disease (CVD) is the leading cause of death in the world, with acute myocardial infarction (AMI) being the main responsible for this leadership. One of the determining factors in the outcome of Acute Coronary Syndrome (ACS) is the time to start treatment. An extremely promising path for obtaining earlier diagnosis and treatment has been the use of technological innovations in emergency care units. PRUPOSE: this study was carried out to assess the impact of applying a technology hub in the chest pain scenario in the emergency room, regarding the feasibility and potential reduction of time for diagnosis and treatment of ACS. METHODS: data obtained from 10 hospitals in the public health system in Brazil, which implemented the technology hub in the last 7 months (May to October 2021), were analyzed. This technology hub uses Artificial Intelligence (AI) to identify electrocardiograms (ECGs) with a high probability of alterations, which must be reported within 5 minutes by the cardiologist on shift (24/7) on the platform. RESULTS: 5,506 ECGs were entered into the platform, of which 53.77% (2,961) were considered abnormal; of these, 9.92% (294) had alterations compatible with ischemic events (currents of injury or myocardial ischemia). The median time for the ECG report made by the specialist was 2 minutes and 51 seconds. CONCLUSION: the implementation of a technology hub in the chest pain scenario in the emergency room proved to be feasible and has great potential for reducing the distance between symptoms and the treatment of patients with ACS.

Pyrethrum extract encapsulated in nanoparticles: Toxicity studies based on genotoxic and hematological effects in bullfrog tadpoles.

The environment receives about 2.7 kg.ha-1 annually of pesticides, used in crop production. Pesticides may have a negative impact on environmental biodiversity and potentially induce physiological effects on non-target species. Advances in technology and nanocarrier systems for agrochemicals led to new alternatives to minimize these impacts, such as nanopesticides, considered more efficient, safe and sustainable. However, it is important to evaluate the risk potential, action and toxicity of nanopesticides in aquatic and terrestrial organisms. This study aims to evaluate genotoxic and hematological biomarkers in bullfrog tadpoles (Lithobates catesbeianus) submitted to acute exposure (48 h) to pyrethrum extract (PYR) and solid lipid nanoparticles loaded with PYR. Results showed increased number of leukocytes during acute exposure, specifically eosinophils in nanoparticle-exposed groups, and basophil in PYR-exposed group. Hematological analysis showed that PYR encapsulated in nanoparticles significantly increased the erythrocyte number compared to the other exposed groups. Data from the comet assay indicated an increase in frequency of the classes that correspond to more severe DNA damages in exposed groups, being that the PYR-exposed group showed a high frequency of class-4 DNA damage. Moreover, erythrocyte nuclear abnormalities were triggered by short-time exposure in all treatments, which showed effects significantly higher than the control group. These results showed genotoxic responses in tadpoles, which could trigger cell death pathways. Concluding, these analyses are important for applications in assessment of contaminated aquatic environments and their biomonitoring, which will evaluate the potential toxicity of xenobiotics, for example, the nanoparticles and pyrethrum extract in frog species. However, further studies are needed to better understand the effects of nanopesticides and botanical insecticides on non-target organisms, in order to contribute to regulatory aspects of future uses for these systems.

Living and Health Conditions Associated with Overweight and Obesity among Elderly.

BACKGROUND: The epidemiological and nutritional transition processes in the last decades underlie the rising trend of obesity in the elderly and is related to increased risk of chronic non-communicable diseases and decreased functional status. OBJECTIVE: To analyze the association of demographic, socioeconomic, lifestyle and health-related factors with overweight and obesity in elderly. DESIGN: Cross-sectional study. SETTING: Carried out in Campinas-São Paulo, Brazil, in 2011. PARTICIPANTS: 452 non-institutionalized elderly (aged ≥60 years), half were users of a government-run soup kitchen and the other half were neighbors of the same sex. RESULTS: Overweight frequency (BMI ≥25 and <30 kg/m2) was 44.5% and obesity (BMI ≥30 kg/m2) was 21.7%. In the multiple multinomial logistic regression model adjusted for sex, age group and economic class, there was greater chance of overweight among those that reported dyslipidemia; those that reported arthritis/ arthrosis/rheumatism and that once or more per week replaced supper by a snack were more likely to be obese. Elderly who did not leave home daily and reported diabetes had higher chance of overweight and obesity. CONCLUSIONS: Overweight and obesity are associated with worse living and health-related conditions, such as physical inactivity, changes in eating behaviors, and chronic diseases. Public health policies should encourage regular physical activity and healthy eating behaviors, focusing on traditional diet, through nutritional education, in order to reduce the prevalence of overweight and obesity and chronic diseases.

Gene selection based on multi-class support vector machines and genetic algorithms

Microarrays are a new technology that allows biologists to better understand the interactions between diverse pathologic state at the gene level. However, the amount of data generated by these tools becomes problematic, even though data are supposed to be automatically analyzed (e.g., for diagnostic purposes). The issue becomes more complex when the expression data involve multiple states. We present a novel approach to the gene selection problem in multi-class gene expression-based cancer classification, which combines support vector machines and genetic algorithms. This new method is able to select small subsets and still improve the classification accuracy.