Using Artificial Intelligence to Develop a Multivariate Model with a Machine Learning Model to Predict Complications in Mexican Diabetic Patients without Arterial Hypertension (National Nested Case-Control Study): Metformin and Elevated Normal Blood Pressure Are Risk Factors, and Obesity Is Protective.

Diabetes mellitus is a disease with no cure that can cause complications and even death. Moreover, over time, it will lead to chronic complications. Predictive models have been used to identify people with a tendency to develop diabetes mellitus. At the same time, there is limited information regarding the chronic complications of patients with diabetes. Our study is aimed at creating a machine-learning model that will be able to identify the risk factors of a diabetic patient developing chronic complications such as amputations, myocardial infarction, stroke, nephropathy, and retinopathy. The design is a national nested case-control study with 63,776 patients and 215 predictors with four years of data. Using an XGBoost model, the prediction of chronic complications has an AUC of 84%, and the model has identified the risk factors for chronic complications in patients with diabetes. According to the analysis, the most crucial risk factors based on SHAP values (Shapley additive explanations) are continued management, metformin treatment, age between 68 and 104 years, nutrition consultation, and treatment adherence. But we highlight two exciting findings. The first is a reaffirmation that high blood pressure figures across patients with diabetes without hypertension become a significant risk factor at diastolic > 70 mmHg (OR: 1.095, 95% CI: 1.078-1.113) or systolic > 120 mmHg (OR: 1.147, 95% CI: 1.124-1.171). Furthermore, people with diabetes with a BMI > 32 (overall obesity) (OR: 0.816, 95% CI: 0.8-0.833) have a statistically significant protective factor, which the paradox of obesity may explain. In conclusion, the results we have obtained show that artificial intelligence is a powerful and feasible tool to use for this type of study. However, we suggest that more studies be conducted to verify and elaborate upon our findings.

[Obesity, CKD and gender on the 30-day in-hospital mortality risk in adults with COVID-19]. / Obesidad, ERC y género en el riesgo de mortalidad hospitalaria a 30 días en adultos con COVID-19.

Background: The coronavirus disease 2019 (COVID-19) pandemic is a serious health problem. The Mexican adult population has a high prevalence of obesity and chronic diseases that increase the risk of dying from this disease. Objective: To identify comorbidities predicting the risk of mortality at 30 days in hospitalized adult subjects with positive laboratory COVID-19 test and to evaluate the interaction between chronic diseases and gender. Material and methods: A retrospective cohort study was conducted in 2020, in a western region of the Mexican Pacific. Data from 51,135 hospitalized patients with positive COVID-19 test were analyzed and were retrieved from a normative system for the epidemiological surveillance of viral respiratory diseases (SINOLAVE, according to its initials in Spanish). Death within the first 30 days from hospital admission was the main outcome and risk ratios (RR) with 95% confidence intervals (95% CI) were calculated. Results: The overall mortality rate was 49.6% and most of the comorbidities analyzed were associated with a higher risk of death. There were significant interactions between gender and obesity (p = 0.003) and chronic kidney disease (p = 0.019). The effect of obesity on the risk of a fatal outcome varied by gender: female, RR = 1.04 (95% CI 1.03-1.07); male, RR = 1.07 (95% CI: 1.06-1.09). Conclusions: A high mortality was observed among the hospitalized patients analyzed and statistically significant factors associated with their risk were identified (gender, obesity, and kidney disease).

Introducción: la pandemia de la enfermedad por coronavirus 2019 (COVID-19) es un problema serio de salud. La población adulta mexicana tiene una alta prevalencia de obesidad y de enfermedades crónicas que incrementan el riesgo de morir por esta enfermedad. Objetivo: identificar comorbilidades predictoras del riesgo de mortalidad a 30 días en sujetos adultos hospitalizados con COVID-19 demostrado por laboratorio y evaluar la interacción entre enfermedades crónicas y el género del paciente. Material y métodos: se hizo un estudio de cohorte retrospectivo en el 2020, en una región del occidente del pacífico mexicano. Se analizaron los datos de 51,135 pacientes hospitalizados con COVID-19, los cuales fueron extraídos de un sistema normativo para la vigilancia epidemiológica de enfermedades respiratorias virales (SINOLAVE). La muerte dentro de los primeros 30 días desde la admisión hospitalaria fue el evento principal y fueron estimadas razones de riesgo (RR) con intervalos de confianza del 95% (IC 95%). Resultados: la mortalidad global fue del 49.6% y la mayoría de las comorbilidades analizadas se asociaron con un mayor riesgo de muerte. Hubo interacciones significativas entre el género y la obesidad (p = 0.003) y la enfermedad renal crónica (p = 0.019). El efecto de la obesidad sobre el riesgo de un desenlace fatal varió en función del género: mujeres, RR = 1.04 (IC 95% 1.03-1.07); hombres, RR = 1.07 (IC 95% 1.06-1.09). Conclusiones: se observó una alta mortalidad entre los pacientes hospitalizados analizados y se identificaron factores asociados a su riesgo (género, obesidad y enfermedad renal).

New Deep Learning Genomic-Based Prediction Model for Multiple Traits with Binary, Ordinal, and Continuous Phenotypes.

Multiple-trait experiments with mixed phenotypes (binary, ordinal and continuous) are not rare in animal and plant breeding programs. However, there is a lack of statistical models that can exploit the correlation between traits with mixed phenotypes in order to improve prediction accuracy in the context of genomic selection (GS). For this reason, when breeders have mixed phenotypes, they usually analyze them using univariate models, and thus are not able to exploit the correlation between traits, which many times helps improve prediction accuracy. In this paper we propose applying deep learning for analyzing multiple traits with mixed phenotype data in terms of prediction accuracy. The prediction performance of multiple-trait deep learning with mixed phenotypes (MTDLMP) models was compared to the performance of univariate deep learning (UDL) models. Both models were evaluated using predictors with and without the genotype × environment (G×E) interaction term (I and WI, respectively). The metric used for evaluating prediction accuracy was Pearson's correlation for continuous traits and the percentage of cases correctly classified (PCCC) for binary and ordinal traits. We found that a modest gain in prediction accuracy was obtained only in the continuous trait under the MTDLMP model compared to the UDL model, whereas for the other traits (1 binary and 2 ordinal) we did not find any difference between the two models. In both models we observed that the prediction performance was better for WI than for I. The MTDLMP model is a good alternative for performing simultaneous predictions of mixed phenotypes (binary, ordinal and continuous) in the context of GS.

A Benchmarking Between Deep Learning, Support Vector Machine and Bayesian Threshold Best Linear Unbiased Prediction for Predicting Ordinal Traits in Plant Breeding.

Genomic selection is revolutionizing plant breeding. However, still lacking are better statistical models for ordinal phenotypes to improve the accuracy of the selection of candidate genotypes. For this reason, in this paper we explore the genomic based prediction performance of two popular machine learning methods: the Multi Layer Perceptron (MLP) and support vector machine (SVM) methods vs. the Bayesian threshold genomic best linear unbiased prediction (TGBLUP) model. We used the percentage of cases correctly classified (PCCC) as a metric to measure the prediction performance, and seven real data sets to evaluate the prediction accuracy, and found that the best predictions (in four out of the seven data sets) in terms of PCCC occurred under the TGLBUP model, while the worst occurred under the SVM method. Also, in general we found no statistical differences between using 1, 2 and 3 layers under the MLP models, which means that many times the conventional neuronal network model with only one layer is enough. However, although even that the TGBLUP model was better, we found that the predictions of MLP and SVM were very competitive with the advantage that the SVM was the most efficient in terms of the computational time required.

Multi-environment Genomic Prediction of Plant Traits Using Deep Learners With Dense Architecture.

Genomic selection is revolutionizing plant breeding and therefore methods that improve prediction accuracy are useful. For this reason, active research is being conducted to build and test methods from other areas and adapt them to the context of genomic selection. In this paper we explore the novel deep learning (DL) methodology in the context of genomic selection. We compared DL methods with densely connected network architecture to one of the most often used genome-enabled prediction models: Genomic Best Linear Unbiased Prediction (GBLUP). We used nine published real genomic data sets to compare a fraction of all possible deep learning models to obtain a "meta picture" of the performance of DL methods with densely connected network architecture. In general, the best predictions were obtained with the GBLUP model when genotype×environment interaction (G×E) was taken into account (8 out of 9 data sets); when the interactions were ignored, the DL method was better than the GBLUP in terms of prediction accuracy in 6 out of the 9 data sets. For this reason, we believe that DL should be added to the data science toolkit of scientists working on animal and plant breeding. This study corroborates the view that there are no universally best prediction machines.

Multi-trait, Multi-environment Deep Learning Modeling for Genomic-Enabled Prediction of Plant Traits.

Multi-trait and multi-environment data are common in animal and plant breeding programs. However, what is lacking are more powerful statistical models that can exploit the correlation between traits to improve prediction accuracy in the context of genomic selection (GS). Multi-trait models are more complex than univariate models and usually require more computational resources, but they are preferred because they can exploit the correlation between traits, which many times helps improve prediction accuracy. For this reason, in this paper we explore the power of multi-trait deep learning (MTDL) models in terms of prediction accuracy. The prediction performance of MTDL models was compared to the performance of the Bayesian multi-trait and multi-environment (BMTME) model proposed by Montesinos-López et al. (2016), which is a multi-trait version of the genomic best linear unbiased prediction (GBLUP) univariate model. Both models were evaluated with predictors with and without the genotype×environment interaction term. The prediction performance of both models was evaluated in terms of Pearson's correlation using cross-validation. We found that the best predictions in two of the three data sets were found under the BMTME model, but in general the predictions of both models, BTMTE and MTDL, were similar. Among models without the genotype×environment interaction, the MTDL model was the best, while among models with genotype×environment interaction, the BMTME model was superior. These results indicate that the MTDL model is very competitive for performing predictions in the context of GS, with the important practical advantage that it requires less computational resources than the BMTME model.

Aggregation is the key to succeed in random walks.

In a random walk (RW) in Z an individual starts at 0 and moves at discrete unitary steps to the right or left with respective probabilities p and 1-p. Assuming p > 1/2 and finite a, a > 1, the probability that state a will be reached before -a is Q(a, p) where Q(a, p) > p. Here we introduce the cooperative random walk (CRW) involving two individuals that move independently according to a RW each but dedicate a fraction of time θ to approach the other one unit. This simple strategy seems to be effective in increasing the expected number of individuals arriving to a first. We conjecture that this is a possible underlying mechanism for efficient animal migration under noisy conditions.

Empirical evidence of the effect of school gathering on the dynamics of dengue epidemics.

INTRODUCTION: Dengue fever is an important vector-transmitted disease that affects more than 100 countries worldwide. Locations where individuals tend to gather may play an important role in disease transmission in the presence of the vector. By controlling mosquitoes' breeding places, this study aims to analyze the effect of reducing transmission in elementary schools (grades 1-9) on the dynamics of the epidemic at a regional level. MATERIALS AND METHODS: In 2007, we implemented a massive campaign in a region of México (Colima state, 5,191 km(2), population 568,000) focused on training janitors to locate and avoid mosquitoes' breeding places, the objective being to maintain elementary schools free of mosquitoes. RESULTS: We observed 45% reduction in dengue incidence compared to the previous year. In contrast, the rest of Mexico observed an 81% increase in incidence on average. DISCUSSION: Costs associated with campaigns focusing on cleaning schools are very low and results seem to be promising. Nevertheless, more controlled studies are needed.

Biological Control of the Chagas Disease Vector Triatoma infestans with the Entomopathogenic Fungus Beauveria bassiana Combined with an Aggregation Cue: Field, Laboratory and Mathematical Modeling Assessment.

BACKGROUND: Current Chagas disease vector control strategies, based on chemical insecticide spraying, are growingly threatened by the emergence of pyrethroid-resistant Triatoma infestans populations in the Gran Chaco region of South America. METHODOLOGY AND FINDINGS: We have already shown that the entomopathogenic fungus Beauveria bassiana has the ability to breach the insect cuticle and is effective both against pyrethroid-susceptible and pyrethroid-resistant T. infestans, in laboratory as well as field assays. It is also known that T. infestans cuticle lipids play a major role as contact aggregation pheromones. We estimated the effectiveness of pheromone-based infection boxes containing B. bassiana spores to kill indoor bugs, and its effect on the vector population dynamics. Laboratory assays were performed to estimate the effect of fungal infection on female reproductive parameters. The effect of insect exuviae as an aggregation signal in the performance of the infection boxes was estimated both in the laboratory and in the field. We developed a stage-specific matrix model of T. infestans to describe the fungal infection effects on insect population dynamics, and to analyze the performance of the biopesticide device in vector biological control. CONCLUSIONS: The pheromone-containing infective box is a promising new tool against indoor populations of this Chagas disease vector, with the number of boxes per house being the main driver of the reduction of the total domestic bug population. This ecologically safe approach is the first proven alternative to chemical insecticides in the control of T. infestans. The advantageous reduction in vector population by delayed-action fungal biopesticides in a contained environment is here shown supported by mathematical modeling.

An application of queuing theory to SIS and SEIS epidemic models.

In this work we consider every individual of a population to be a server whose state can be either busy (infected) or idle (susceptible). This server approach allows to consider a general distribution for the duration of the infectious state, instead of being restricted to exponential distributions. In order to achieve this we first derive new approximations to quasistationary distribution (QSD) of SIS (Susceptible- Infected- Susceptible) and SEIS (Susceptible- Latent- Infected- Susceptible) stochastic epidemic models. We give an expression that relates the basic reproductive number, R0 and the server utilization,p.

Applications of occupancy urn models to epidemiology.

This paper shows how occupancy urn models can be used to derive useful results in epidemiology. First we show how simple epidemic models can be re-interpreted in terms of occupancy problems. We use this reformulation to derive an expression for the expected epidemic size, that is, the total number of infected at the end of an outbreak. We also use this approach to derive point and interval estimates of the Basic Reproduction Ratio, R0 . We show that this construction does not require that the underlying SIR model be a homogeneous Poisson process, leading to a geometric distribution for the number of contacts before removal, but instead it supports a general distribution. The urn model construction is easy to handle and represents a rich field for further exploitation.

The effect of the hiv/aids epidemic on africa's truck drivers.

The AIDS epidemic is having a growing impact on the transport sector of the economy of sub-Saharan Africa, where long-distance truck drivers are at an increased risk of infection due to their frequent contacts with commercial sex workers. The spread of AIDS in the transport industry is especially significant to the economy, as truck drivers are largely responsible for transporting crops and supplies needed for daily subsistence. In this paper we analyze these effects via two models, one employing a switch and the other a Verhulst saturation function, to describe the rate at which new drivers are recruited in terms of the supply and demand for them in the general population. Results provide an estimate of the epidemic's economic impact on the transportation sector through the loss of truck drivers (an estimated 10% per year, with endemic levels near 90%).

A markov chain approach to calculate r(0) in stochastic epidemic models.

R(0) has been defined as "The expected number of secondary infections originated by a "typical" infective individual when introduced into a population of susceptibles", and it is perhaps the single most important parameter in epidemic models. A general framework to calculate R(0) that can be applied to complicated stochastic epidemic models that may include demography, several strains, latent or carrier-like states, with or without density-dependent parameters is introduced. This framework helps us to understand the concept of a "typical" infective individual used in the deterministic definition of R(0). The method is illustrated with applications to several epidemic models, including some in which it has been found that the disease may persist even if R(0)<1. It is shown that although the probability of extinction is difficult to calculate in these latter cases, it is possible to give general conditions on the parameters under which eventual extinction is certain.