Forest yield prediction under different climate change scenarios using data intelligent models in Pakistan / Previsão de produção florestal em diferentes cenários de mudanças climáticas usando modelos inteligentes de dados no Paquistão

Abstract This study aimed to develop and evaluate data driven models for prediction of forest yield under different climate change scenarios in the Gallies forest division of district Abbottabad, Pakistan. The Random Forest (RF) and Kernel Ridge Regression (KRR) models were developed and evaluated using yield data of two species (Blue pine and Silver fir) as an objective variable and climate data (temperature, humidity, rainfall and wind speed) as predictive variables. Prediction accuracy of both the models were assessed by means of root mean squared error (RMSE), mean absolute error (MAE), correlation coefficient (r), relative root mean squared error (RRMSE), Legates-McCabe's (LM), Willmott's index (WI) and Nash-Sutcliffe (NSE) metrics. Overall, the RF model outperformed the KRR model due to its higher accuracy in forecasting of forest yield. The study strongly recommends that RF model should be applied in other regions of the country for prediction of forest growth and yield, which may help in the management and future planning of forest productivity in Pakistan.

Resumo Este estudo teve como objetivo desenvolver e avaliar modelos baseados em dados para previsão da produção florestal em diferentes cenários de mudanças climáticas na divisão florestal Gallies do distrito de Abbottabad, Paquistão. Os modelos Random Forest (RF) e Kernel Ridge Regression (KRR) foram desenvolvidos e avaliados usando dados de produção de duas espécies (pinheiro-azul e abeto-prateado) como uma variável objetiva e dados climáticos (temperatura, umidade, precipitação e velocidade do vento) como preditivos variáveis. A precisão da previsão de ambos os modelos foi avaliada por meio de erro quadrático médio (RMSE), erro absoluto médio (MAE), coeficiente de correlação (r), erro quadrático médio relativo (RRMSE), Legates-McCabe's (LM), índice de Willmott (WI) e métricas Nash-Sutcliffe (NSE). No geral, o modelo RF superou o modelo KRR devido à sua maior precisão na previsão do rendimento florestal. O estudo recomenda fortemente que o modelo RF seja aplicado em outras regiões do país para previsão do crescimento e produtividade florestal, o que pode ajudar no manejo e planejamento futuro da produtividade florestal no Paquistão.

Prediction of visceral leishmaniasis incidence using the Seasonal Autoregressive Integrated Moving Average model (SARIMA) in the state of Maranhão, Brazil / Previsão da incidência da leishmaniose visceral usando o modelo de média móvel integrado autorregressivo sazonal (SARIMA) no Maranhão, Brasil

Abstract Visceral leishmaniasis (VL) is an infectious disease predominant in countries located in the tropics. The prediction of occurrence of infectious diseases through epidemiologic modeling has revealed to be an important tool in the understanding of its occurrence dynamic. The objective of this study was to develop a forecasting model for the incidence of VL in Maranhão using the Seasonal Autoregressive Integrated Moving Average model (SARIMA). We collected monthly data regarding VL cases from the National Disease Notification System (SINAN) corresponding to the period between 2001 and 2018. The Box-Jenkins method was applied in order to adjust a SARIMA prediction model for VL general incidence and by sex (male or female) for the period between January 2019 and December 2013. For 216 months of this time series, 10,431 cases of VL were notified in Maranhão, with an average of 579 cases per year. With regard to age range, there was a higher incidence among the pediatric public (0 to 14 years of age). There was a predominance in male cases, 6437 (61.71%). The Box-Pierce test figures for overall, male and female genders supported by the results of the Ljung-Box test suggest that the autocorrelations of residual values act as white noise. Regarding monthly occurrences in general and by gender, the SARIMA models (2,0,0) (2,0,0), (0,1,1) (0,1,1) and (0,1,1) (2, 0, 0) were the ones that mostly adjusted to the data respectively. The model SARIMA has proven to be an adequate tool for predicting and analyzing the trends in VL incidence in Maranhão. The time variation determination and its prediction are decisive in providing guidance in health measure intervention.

Resumo A leishmaniose visceral (LV) é uma doença de natureza infecciosa, predominante em países de zonas tropicais. A predição de ocorrência de doenças infecciosas através da modelagem epidemiológica tem se revelado uma importante ferramenta no entendimento de sua dinâmica de ocorrência. O objetivo deste estudo foi desenvolver um modelo de previsão da incidência da LV no Maranhão usando o modelo de Média Móvel Integrada Autocorrelacionada Sazonal (SARIMA). Foram coletados os dados mensais de casos de LV através do Sistema de Informação de Agravos de Notificação (SINAN) correspondentes ao período de 2001 a 2018. O método de Box-Jenkins foi aplicado para ajustar um modelo de predição SARIMA para incidência geral e por sexo (masculino e feminino) de LV para o período de janeiro de 2019 a dezembro de 2023. Durante o período de 216 meses dessa série temporal, foram registrados 10.431 casos de LV no Maranhão, com uma média de 579 casos por ano. Em relação à faixa etária, houve maior registro no público pediátrico (0 a 14 anos). Houve predominância do sexo masculino, com 6437 casos (61,71%). Os valores do teste de Box-Pierce para incidência geral, sexo masculino e feminino reforçados pelos resultados do teste Ljung-Box sugerem que as autocorrelações de resíduos apresentam um comportamento de ruído branco. Para incidência mensal geral e por sexo masculino e feminino, os modelos SARIMA (2,0,0) (2,0,0), (0,1,1) (0,1,1) e (0,1,1) (2, 0, 0) foram os que mais se ajustaram aos dados, respectivamente. O modelo SARIMA se mostrou uma ferramenta adequada de previsão e análise da tendência de incidência da LV no Maranhão. A determinação da variação temporal e sua predição são determinantes no norteamento de medidas de intervenção em saúde.

A disease predictive model based on epidemiological factors for the management of bacterial leaf blight of rice / Um modelo preditivo de doenças baseado em fatores epidemiológicos para o manejo da queima bacteriana das folhas do arroz

Rice is a widely consumed staple food for a large part of the world's human population. Approximately 90% of the world's rice is grown in Asian continent and constitutes a staple food for 2.7 billion people worldwide. Bacterial leaf blight (BLB) caused by Xanthomonas oryzae pv. oryzae is one of the devastating diseases of rice. A field experiment was conducted during the year 2016 and 2017 to investigate the influence of different meteorological parameters on BLB development as well as the computation of a predictive model to forecast the disease well ahead of its appearance in the field. The seasonal dataset of disease incidence and environmental factors was used to assess five rice varieties/ cultivars (Basmati-2000, KSK-434, KSK-133, Super Basmati, and IRRI-9). The accumulated effect of two year environmental data; maximum and minimum temperature, relative humidity, wind speed, and rainfall, was studied and correlated with disease incidence. Average temperature (maximum & minimum) showed a negative significant correlation with BLB disease and all other variables; relative humidity, rainfall, and wind speed had a positive correlation with BLB disease development on individual varieties. Stepwise regression analysis was performed to indicate potentially useful predictor variables and to rule out incompetent parameters. Environmental data from the growing seasons of July to October 2016 and 2017 revealed that, with the exception of the lowest temperature, all environmental factors contributed to disease development throughout the cropping season. A disease prediction multiple regression model was developed based on two-year data (Y = 214.3-3.691 Max T-0.508 Min T + 0.767 RH + 2.521 RF + 5.740 WS), which explained 95% variability. This disease prediction model will not only help farmers in early detection and timely management of bacterial leaf blight disease of rice but may also help reduce input costs and improve product quality and quantity. The model will be both farmer and environmentally friendly.

O arroz é um alimento básico amplamente consumido por grande parte da população humana mundial. Aproximadamente 90% do arroz do mundo é cultivado no continente asiático e constitui um alimento básico para 2,7 bilhões de pessoas em todo o mundo. O crestamento bacteriano das folhas (BLB) causado por Xanthomonas oryzae pv. oryzae é uma das doenças devastadoras do arroz. Um experimento de campo foi realizado durante os anos de 2016 e 2017 para investigar a influência de diferentes parâmetros meteorológicos no desenvolvimento do BLB, bem como o cálculo de um modelo preditivo para prever a doença bem antes de seu aparecimento em campo. O conjunto de dados sazonais de incidência de doenças e fatores ambientais foi usado para avaliar cinco variedades/cultivares de arroz (Basmati-2000, KSK-434, KSK-133, Super Basmati e IRRI-9). O efeito acumulado de dados ambientais de dois anos; temperatura máxima e mínima, umidade relativa do ar, velocidade do vento e precipitação pluviométrica foram estudados e correlacionados com a incidência da doença. A temperatura média (máxima e mínima) apresentou correlação significativa negativa com a doença BLB e todas as outras variáveis; umidade relativa, precipitação e velocidade do vento tiveram uma correlação positiva com o desenvolvimento da doença BLB em variedades individuais. A análise de regressão stepwise foi realizada para indicar variáveis preditoras potencialmente úteis e para descartar parâmetros incompetentes. Os dados ambientais das safras de julho a outubro de 2016 e 2017 revelaram que, com exceção da temperatura mais baixa, todos os fatores ambientais contribuíram para o desenvolvimento da doença ao longo da safra. Um modelo de regressão múltipla de previsão de doença foi desenvolvido com base em dados de dois anos (Y = 214,3-3,691 Max T-0,508 Min T + 0,767 RH + 2,521 RF + 5,740 WS), que explicou 95% de variabilidade. Este modelo de previsão de doenças não só ajudará os agricultores na detecção precoce e gestão atempada da doença bacteriana das folhas do arroz, mas também pode ajudar a reduzir os custos de insumos e melhorar a qualidade e a quantidade do produto. O modelo será agricultor e ambientalmente amigável.

Predicting the Need for Blood Transfusions in Cardiac Surgery: A Comparison between Machine Learning Algorithms and Established Risk Scores in the Brazilian Population

ABSTRACT Introduction: Blood transfusion is a common practice in cardiac surgery, despite its well-known negative effects. To mitigate blood transfusion-associated risks, identifying patients who are at higher risk of needing this procedure is crucial. Widely used risk scores to predict the need for blood transfusions have yielded unsatisfactory results when validated for the Brazilian population. Methods: In this retrospective study, machine learning (ML) algorithms were compared to predict the need for blood transfusions in a cohort of 495 cardiac surgery patients treated at a Brazilian reference service between 2019 and 2021. The performance of the models was evaluated using various metrics, including the area under the curve (AUC), and compared to the commonly used Transfusion Risk and Clinical Knowledge (TRACK) and Transfusion Risk Understanding Scoring Tool (TRUST) scoring systems. Results: The study found that the model had the highest performance, achieving an AUC of 0.7350 (confidence interval [CI]: 0.7203 to 0.7497). Importantly, all ML algorithms performed significantly better than the commonly used TRACK and TRUST scoring systems. TRACK had an AUC of 0.6757 (CI: 0.6609 to 0.6906), while TRUST had an AUC of 0.6622 (CI: 0.6473 to 0.6906). Conclusion: The findings of this study suggest that ML algorithms may offer a more accurate prediction of the need for blood transfusions than the traditional scoring systems and could enhance the accuracy of predicting blood transfusion requirements in cardiac surgery patients. Further research could focus on optimizing and refining ML algorithms to improve their accuracy and make them more suitable for clinical use.

Development and internal validation of a clinical prediction model for spontaneous abortion risk in early pregnancy

Abstract Objective: This study aimed to develop and internally validate a prediction model for estimating the risk of spontaneous abortion in early pregnancy. Methods: This prospective cohort study included 9,895 pregnant women who received prenatal care at a maternal health facility in China from January 2021 to December 2022. Data on demographics, medical history, lifestyle factors, and mental health were collected. A multivariable logistic regression analysis was performed to develop the prediction model with spontaneous abortion as the outcome. The model was internally validated using bootstrapping techniques, and its discrimination and calibration were assessed. Results: The spontaneous abortion rate was 5.95% (589/9,895) 1. The final prediction model included nine variables: maternal age, history of embryonic arrest, thyroid dysfunction, polycystic ovary syndrome, assisted reproduction, exposure to pollution, recent home renovation, depression score, and stress score 1. The model showed good discrimination with a C-statistic of 0.88 (95% CI 0.87‒0.90) 1, and its calibration was adequate based on the Hosmer-Lemeshow test (p = 0.27). Conclusions: The prediction model demonstrated good performance in estimating spontaneous abortion risk in early pregnancy based on demographic, clinical, and psychosocial factors. Further external validation is recommended before clinical application.

Fully independent validation of eleven prognostic scores predicting progression to critically ill condition in hospitalized patients with COVID-19

Abstract Introduction COVID-19 remains an important threat to global health and maintains the challenge of COVID-19 hospital care. To assist decision making regarding COVID-19 hospital care many instruments to predict COVID-19 progression to critical condition were developed and validated. Objective To validate eleven COVID-19 progression prediction scores for critically ill hospitalized patients in a Brazilian population. Methodology Observational study with retrospective follow-up, including 301 adults confirmed for COVID-19 sequentially. Participants were admitted to non-critical units for treatment of the disease, between January and April 2021 and between September 2021 and February 2022. Eleven prognostic scores were applied using demographic, clinical, laboratory and imaging data collected in the first 48 of the hospital admission. The outcomes of greatest interest were as originally defined for each score. The analysis plan was to apply the instruments, estimate the outcome probability reproducing the original development/validation of each score, then to estimate performance measures (discrimination and calibration) and decision thresholds for risk classification. Results The overall outcome prevalence was 41.8 % on 301 participants. There was a greater risk of the occurrence of the outcomes in older and male patients, and a linear trend with increasing comorbidities. Most of the patients studied were not immunized against COVID-19. Presence of concomitant bacterial infection and consolidation on imaging increased the risk of outcomes. College of London COVID-19 severity score and the 4C Mortality Score were the only with reasonable discrimination (ROC AUC 0.647 and 0.798 respectively) and calibration. The risk groups (low, intermediate and high) for 4C score were updated with the following thresholds: 0.239 and 0.318 (https://pedrobrasil.shinyapps.io/INDWELL/). Conclusion The 4C score showed the best discrimination and calibration performance among the tested instruments. We suggest different limits for risk groups. 4C score use could improve decision making and early therapeutic management at hospital care.

Forest yield prediction under different climate change scenarios using data intelligent models in Pakistan

Abstract This study aimed to develop and evaluate data driven models for prediction of forest yield under different climate change scenarios in the Gallies forest division of district Abbottabad, Pakistan. The Random Forest (RF) and Kernel Ridge Regression (KRR) models were developed and evaluated using yield data of two species (Blue pine and Silver fir) as an objective variable and climate data (temperature, humidity, rainfall and wind speed) as predictive variables. Prediction accuracy of both the models were assessed by means of root mean squared error (RMSE), mean absolute error (MAE), correlation coefficient (r), relative root mean squared error (RRMSE), Legates-McCabes (LM), Willmotts index (WI) and Nash-Sutcliffe (NSE) metrics. Overall, the RF model outperformed the KRR model due to its higher accuracy in forecasting of forest yield. The study strongly recommends that RF model should be applied in other regions of the country for prediction of forest growth and yield, which may help in the management and future planning of forest productivity in Pakistan.

Resumo Este estudo teve como objetivo desenvolver e avaliar modelos baseados em dados para previsão da produção florestal em diferentes cenários de mudanças climáticas na divisão florestal Gallies do distrito de Abbottabad, Paquistão. Os modelos Random Forest (RF) e Kernel Ridge Regression (KRR) foram desenvolvidos e avaliados usando dados de produção de duas espécies (pinheiro-azul e abeto-prateado) como uma variável objetiva e dados climáticos (temperatura, umidade, precipitação e velocidade do vento) como preditivos variáveis. A precisão da previsão de ambos os modelos foi avaliada por meio de erro quadrático médio (RMSE), erro absoluto médio (MAE), coeficiente de correlação (r), erro quadrático médio relativo (RRMSE), Legates-McCabes (LM), índice de Willmott (WI) e métricas Nash-Sutcliffe (NSE). No geral, o modelo RF superou o modelo KRR devido à sua maior precisão na previsão do rendimento florestal. O estudo recomenda fortemente que o modelo RF seja aplicado em outras regiões do país para previsão do crescimento e produtividade florestal, o que pode ajudar no manejo e planejamento futuro da produtividade florestal no Paquistão.

Prediction of visceral leishmaniasis incidence using the Seasonal Autoregressive Integrated Moving Average model (SARIMA) in the state of Maranhão, Brazil

Abstract Visceral leishmaniasis (VL) is an infectious disease predominant in countries located in the tropics. The prediction of occurrence of infectious diseases through epidemiologic modeling has revealed to be an important tool in the understanding of its occurrence dynamic. The objective of this study was to develop a forecasting model for the incidence of VL in Maranhão using the Seasonal Autoregressive Integrated Moving Average model (SARIMA). We collected monthly data regarding VL cases from the National Disease Notification System (SINAN) corresponding to the period between 2001 and 2018. The Box-Jenkins method was applied in order to adjust a SARIMA prediction model for VL general incidence and by sex (male or female) for the period between January 2019 and December 2013. For 216 months of this time series, 10,431 cases of VL were notified in Maranhão, with an average of 579 cases per year. With regard to age range, there was a higher incidence among the pediatric public (0 to 14 years of age). There was a predominance in male cases, 6437 (61.71%). The Box-Pierce test figures for overall, male and female genders supported by the results of the Ljung-Box test suggest that the autocorrelations of residual values act as white noise. Regarding monthly occurrences in general and by gender, the SARIMA models (2,0,0) (2,0,0), (0,1,1) (0,1,1) and (0,1,1) (2, 0, 0) were the ones that mostly adjusted to the data respectively. The model SARIMA has proven to be an adequate tool for predicting and analyzing the trends in VL incidence in Maranhão. The time variation determination and its prediction are decisive in providing guidance in health measure intervention.

Resumo A leishmaniose visceral (LV) é uma doença de natureza infecciosa, predominante em países de zonas tropicais. A predição de ocorrência de doenças infecciosas através da modelagem epidemiológica tem se revelado uma importante ferramenta no entendimento de sua dinâmica de ocorrência. O objetivo deste estudo foi desenvolver um modelo de previsão da incidência da LV no Maranhão usando o modelo de Média Móvel Integrada Autocorrelacionada Sazonal (SARIMA). Foram coletados os dados mensais de casos de LV através do Sistema de Informação de Agravos de Notificação (SINAN) correspondentes ao período de 2001 a 2018. O método de Box-Jenkins foi aplicado para ajustar um modelo de predição SARIMA para incidência geral e por sexo (masculino e feminino) de LV para o período de janeiro de 2019 a dezembro de 2023. Durante o período de 216 meses dessa série temporal, foram registrados 10.431 casos de LV no Maranhão, com uma média de 579 casos por ano. Em relação à faixa etária, houve maior registro no público pediátrico (0 a 14 anos). Houve predominância do sexo masculino, com 6437 casos (61,71%). Os valores do teste de Box-Pierce para incidência geral, sexo masculino e feminino reforçados pelos resultados do teste Ljung-Box sugerem que as autocorrelações de resíduos apresentam um comportamento de ruído branco. Para incidência mensal geral e por sexo masculino e feminino, os modelos SARIMA (2,0,0) (2,0,0), (0,1,1) (0,1,1) e (0,1,1) (2, 0, 0) foram os que mais se ajustaram aos dados, respectivamente. O modelo SARIMA se mostrou uma ferramenta adequada de previsão e análise da tendência de incidência da LV no Maranhão. A determinação da variação temporal e sua predição são determinantes no norteamento de medidas de intervenção em saúde.

A disease predictive model based on epidemiological factors for the management of bacterial leaf blight of rice

Abstract Rice is a widely consumed staple food for a large part of the worlds human population. Approximately 90% of the worlds rice is grown in Asian continent and constitutes a staple food for 2.7 billion people worldwide. Bacterial leaf blight (BLB) caused by Xanthomonas oryzae pv. oryzae is one of the devastating diseases of rice. A field experiment was conducted during the year 2016 and 2017 to investigate the influence of different meteorological parameters on BLB development as well as the computation of a predictive model to forecast the disease well ahead of its appearance in the field. The seasonal dataset of disease incidence and environmental factors was used to assess five rice varieties/ cultivars (Basmati-2000, KSK-434, KSK-133, Super Basmati, and IRRI-9). The accumulated effect of two year environmental data; maximum and minimum temperature, relative humidity, wind speed, and rainfall, was studied and correlated with disease incidence. Average temperature (maximum & minimum) showed a negative significant correlation with BLB disease and all other variables; relative humidity, rainfall, and wind speed had a positive correlation with BLB disease development on individual varieties. Stepwise regression analysis was performed to indicate potentially useful predictor variables and to rule out incompetent parameters. Environmental data from the growing seasons of July to October 2016 and 2017 revealed that, with the exception of the lowest temperature, all environmental factors contributed to disease development throughout the cropping season. A disease prediction multiple regression model was developed based on two-year data (Y = 214.3-3.691 Max T-0.508 Min T + 0.767 RH + 2.521 RF + 5.740 WS), which explained 95% variability. This disease prediction model will not only help farmers in early detection and timely management of bacterial leaf blight disease of rice but may also help reduce input costs and improve product quality and quantity. The model will be both farmer and environmentally friendly.

Resumo O arroz é um alimento básico amplamente consumido por grande parte da população humana mundial. Aproximadamente 90% do arroz do mundo é cultivado no continente asiático e constitui um alimento básico para 2,7 bilhões de pessoas em todo o mundo. O crestamento bacteriano das folhas (BLB) causado por Xanthomonas oryzae pv. oryzae é uma das doenças devastadoras do arroz. Um experimento de campo foi realizado durante os anos de 2016 e 2017 para investigar a influência de diferentes parâmetros meteorológicos no desenvolvimento do BLB, bem como o cálculo de um modelo preditivo para prever a doença bem antes de seu aparecimento em campo. O conjunto de dados sazonais de incidência de doenças e fatores ambientais foi usado para avaliar cinco variedades/cultivares de arroz (Basmati-2000, KSK-434, KSK-133, Super Basmati e IRRI-9). O efeito acumulado de dados ambientais de dois anos; temperatura máxima e mínima, umidade relativa do ar, velocidade do vento e precipitação pluviométrica foram estudados e correlacionados com a incidência da doença. A temperatura média (máxima e mínima) apresentou correlação significativa negativa com a doença BLB e todas as outras variáveis; umidade relativa, precipitação e velocidade do vento tiveram uma correlação positiva com o desenvolvimento da doença BLB em variedades individuais. A análise de regressão stepwise foi realizada para indicar variáveis preditoras potencialmente úteis e para descartar parâmetros incompetentes. Os dados ambientais das safras de julho a outubro de 2016 e 2017 revelaram que, com exceção da temperatura mais baixa, todos os fatores ambientais contribuíram para o desenvolvimento da doença ao longo da safra. Um modelo de regressão múltipla de previsão de doença foi desenvolvido com base em dados de dois anos (Y = 214,3-3,691 Max T-0,508 Min T + 0,767 RH + 2,521 RF + 5,740 WS), que explicou 95% de variabilidade. Este modelo de previsão de doenças não só ajudará os agricultores na detecção precoce e gestão atempada da doença bacteriana das folhas do arroz, mas também pode ajudar a reduzir os custos de insumos e melhorar a qualidade e a quantidade do produto. O modelo será agricultor e ambientalmente amigável.

Invasiveness assessment by CT quantitative and qualitative features of lung cancers manifesting ground-glass nodules in 555 patients: A retrospective cohort study / 中国胸心血管外科临床杂志

@#Objective     To explore the correlation between the quantitative and qualitative features of CT images and the invasiveness of pulmonary ground-glass nodules, providing reference value for preoperative planning of patients with ground-glass nodules. Methods    The patients with ground-glass nodules who underwent surgical treatment and were diagnosed with pulmonary adenocarcinoma from September 2020 to July 2022 at the Third Affiliated Hospital of Kunming Medical University were collected. Based on the pathological diagnosis results, they were divided into two groups: a non-invasive adenocarcinoma group with in situ and minimally invasive adenocarcinoma, and an invasive adenocarcinoma group. Imaging features were collected, and a univariate logistic regression analysis was conducted on the clinical and imaging data of the patients. Variables with statistical difference were selected for multivariate logistic regression analysis to establish a predictive model of invasive adenocarcinoma based on independent risk factors. Finally, the sensitivity and specificity were calculated based on the Youden index. Results     A total of 555 patients were collected. The were 310 patients in the non-invasive adenocarcinoma group, including 235 females and 75 males, with a meadian age of 49 (43, 58) years, and 245 patients in the invasive adenocarcinoma group, including 163 females and 82 males, with a meadian age of 53 (46, 61) years. The binary logistic regression analysis showed that the maximum diameter (OR=4.707, 95%CI 2.060 to 10.758), consolidation/tumor ratio (CTR, OR=1.027, 95%CI 1.011 to 1.043), maximum CT value (OR=1.025, 95%CI 1.004 to 1.047), mean CT value (OR=1.035, 95%CI 1.008 to 1.063), spiculation sign (OR=2.055, 95%CI 1.148 to 3.679), and vascular convergence sign (OR=2.508, 95%CI 1.345 to 4.676) were independent risk factors for the occurrence of invasive adenocarcinoma (P<0.05). Based on the independent predictive factors, a predictive model of invasive adenocarcinoma was constructed. The formula for the model prediction was: Logit(P)=–1.293+1.549×maximum diameter of lesion+0.026×CTR+0.025×maximum CT value+0.034×mean CT value+0.72×spiculation sign+0.919×vascular convergence sign. The area under the receiver operating characteristic curve of the model was 0.910 (95%CI 0.885 to 0.934), indicating that the model had good discrimination ability. The calibration curve showed that the predictive model had good calibration, and the decision analysis curve showed that the model had good clinical utility. Conclusion     The predictive model combining quantitative and qualitative features of CT has a good predictive ability for the invasiveness of ground-glass nodules. Its predictive performance is higher than any single indicator.

A postoperative in-hospital mortality risk model for elderly patients undergoing cardiac valvular surgery based on LASSO-logistic regression / 中国胸心血管外科临床杂志

@#Objective     To evaluate the risk factors for postoperative in-hospital mortality in elderly patients receiving cardiac valvular surgery, and develop a new prediction models using the least absolute shrinkage and selection operator (LASSO)-logistic regression. Methods     The patients≥65 years who underwent cardiac valvular surgery from 2016 to 2018 were collected from the Chinese Cardiac Surgery Registry (CCSR). The patients who received the surgery from January 2016 to June 2018 were allocated to a training set, and the patients who received the surgery from July to December 2018 were allocated to a testing set. The risk factors for postoperative mortality were analyzed and a LASSO-logistic regression prediction model was developed and compared with the EuroSCOREⅡ. Results     A total of 7 163 patients were collected in this study, including 3 939 males and 3 224 females, with a mean age of 69.8±4.5 years. There were 5 774 patients in the training set and 1 389 patients in the testing set. Overall, the in-hospital mortality was 4.0% (290/7 163). The final LASSO-logistic regression model included 7 risk factors: age, preoperative left ventricular ejection fraction, combined coronary artery bypass grafting, creatinine clearance rate, cardiopulmonary bypass time, New York Heart Association cardiac classification. LASSO-logistic regression had a satisfying discrimination and calibration in both training [area under the curve (AUC)=0.785, 0.627] and testing cohorts (AUC=0.739, 0.642), which was superior to EuroSCOREⅡ. Conclusion     The mortality rate for elderly patients undergoing cardiac valvular surgery is relatively high. LASSO-logistic regression model can predict the risk of in-hospital mortality in elderly patients receiving cardiac valvular surgery.

Risk assessment of nosocomial infection in patients with multiple myeloma in the first hospitalization / 公共卫生与预防医学

Objective To assess the risk of nosocomial infection in patients with multiple myeloma during their first hospitalization. Methods Totally 480 patients with multiple myeloma who were hospitalized for the first time in department of hematology of West China Hospital, Sichuan University from August 2021 to August 2022 were included, and the nosocomial infection during treatment was statistically analyzed. The patients were divided into infected group and uninfected group. The independent influencing factors of nosocomial infection were analyzed and a prediction model was established. The reliability of the prediction model was analyzed by receiver operating characteristic curve (ROC). Results The incidence rate of nosocomial infection was 31.2% among 480 patients hospitalized for the first time. There were statistically significant differences in age, ISS staging, controlling nutritional status (CONUT) score, agranulocytosis, hemoglobin, and albumin between the infected group and the uninfected group (P<0.05). Logistic multivariate regression analysis showed that age, ISS staging, CONUT score, agranulocytosis, hemoglobin level, and albumin level were all independent correlated factors of nosocomial infection in patients with multiple myeloma hospitalized for the first time (P<0.05). The area under the ROC curve (AUC), sensitivity and specificity of multivariate logistic regression prediction model were 0.88 (95%CI: 0.840-0.920), 85.00% and 76.36%, respectively. Conclusion The incidence rate of nosocomial infection is high among patients with multiple myeloma in the first hospitalization. The prediction model established according to independent correlated factors of nosocomial infection has high predictive value on the occurrence of nosocomial infection.

Development and prospects of predicting drug polymorphs technology / 药学学报

Most chemical medicines have polymorphs. The difference of medicine polymorphs in physicochemical properties directly affects the stability, efficacy, and safety of solid medicine products. Polymorphs is incomparably important to pharmaceutical chemistry, manufacturing, and control. Meantime polymorphs is a key factor for the quality of high-end drug and formulations. Polymorph prediction technology can effectively guide screening of trial experiments, and reduce the risk of missing stable crystal form in the traditional experiment. Polymorph prediction technology was firstly based on theoretical calculations such as quantum mechanics and computational chemistry, and then was developed by the key technology of machine learning using the artificial intelligence. Nowadays, the popular trend is to combine the advantages of theoretical calculation and machine learning to jointly predict crystal structure. Recently, predicting medicine polymorphs has still been a challenging problem. It is expected to learn from and integrate existing technologies to predict medicine polymorphs more accurately and efficiently.

Construction of a individualized model for predicting the risk of recurrence in patients with retinal vein occlusion induced macular edema / 国际眼科杂志(Guoji Yanke Zazhi)

AIM: To analyze the recurrence factors of patients with retinal vein occlusion(RVO)induced macular edema(ME)and construct a nomogram model.METHODS: Retrospective study. A total of 306 patients with RVO induced ME admitted to our hospital from January 2019 to June 2022 were included as study objects, and they were divided into modeling group with 214 cases(214 eyes)and 92 cases(92 eyes)in the verification group by 7:3. All patients were followed up for 1 a after receiving anti-vascular endothelial growth factor(VEGF)treatment, and patients in the modeling group were separated into a recurrence group(n=66)and a non recurrence group(n=148)based on whether they had recurrence. Clinical data were collected and multivariate Logistic regression was applied to analyze and determine the factors affecting recurrence in patients with RVO induced ME; R3.6.3 software was applied to construct a nomogram model for predicting the recurrence risk of patients with RVO induced ME; ROC curve and calibration curve were applied to evaluate the discrimination and consistency of nomogram model in predicting the recurrence risk of patients with RVO induced ME.RESULTS: There were statistically significant differences in central retinal thickness(CRT), course of disease, hyperreflective foci(HF), disorder of retinal inner layer structure, and injection frequency between the non recurrence group and the recurrence group before treatment(all P&#x0026;#x003C;0.05). The multivariate Logistic regression analysis showed that pre-treatment CRT(OR=1.011), course of disease(OR=1.104), HF(OR=5.074), retinal inner layer structural disorder(OR=4.640), and injection frequency(OR=4.036)were influencing factors for recurrence in patients with RVO induced ME(all P&#x0026;#x003C;0.01). The area under the ROC curve of the modeling group was 0.924(95%CI: 0.882-0.966), the slope of the calibration curve was close to 1, and the results of the Hosmer-Lemeshow goodness of fit test showed that χ2=11.817, P=0.160; the area under the ROC curve of the verification group was 0.939(95%CI: 0.892-0.985), the slope of the calibration curve was close to 1, and the results of the Hosmer-Lemeshow goodness of fit test showed χ2=6.082, P=0.638.CONCLUSION: Pre-treatment CRT, course of disease, HF, disorder of retinal inner layer structure, and injection frequency are independent risk factors for recurrence in patients with RVO induced ME. The nomogram model constructed based on this has a high discrimination and consistency in predicting the recurrence risk of patients with RVO induced ME.

Construction and validation of a prediction model for public acceptance of kidney xenotransplantation in China / 器官移植

Objective To explore the public attitude towards kidney xenotransplantation in China by constructing and validating the prediction model based on xenotransplantation questionnaire. Methods A convenient sampling survey was conducted among the public in China with the platform of Wenjuanxing to analyze public acceptance of kidney xenotransplantation and influencing factors. Using random distribution method, all included questionnaires (n=2 280) were divided into the training and validation sets according to a ratio of 7:3. A prediction model was constructed and validated. Results A total of 2 280 questionnaires were included. The public acceptance rate of xenotransplantation was 71.3%. Multivariate analysis showed that gender, marital status, resident area, medical insurance coverage, religious belief, vegetarianism, awareness of kidney xenotransplantation and whether on the waiting list for kidney transplantation were the independent influencing factors for public acceptance of kidney xenotransplantation (all P<0.05). The area under the curve (AUC) of receiver operating characteristic (ROC) of the prediction model in the training set was 0.773, and 0.785 in the validation set. The calibration curves in the training and validation sets indicated that the prediction models yielded good prediction value. Decision curve analysis (DCA) suggested that the prediction efficiency of the model was high. Conclusions In China, public acceptance of kidney xenotransplantation is relatively high, whereas it remains to be significantly enhanced. The prediction model based on questionnaire survey has favorable prediction efficiency, which provides reference for subsequent research.

Construction of a risk prediction model for bloodstream infection induced by carbapenem-resistant Klebsiella pneumoniae / 中国药房

OBJECTIVE To construct a risk prediction model for bloodstream infection （BSI） induced by carbapenem-resistant Klebsiella pneumoniae （CRKP）. METHODS Retrospective analysis was conducted for clinical data from 253 patients with BSI induced by K. pneumoniae in the First Hospital of Qinhuangdao from January 2019 to June 2022. Patients admitted from January 2019 to December 2021 were selected as the model group （n＝223）， and patients admitted from January 2022 to June 2022 were selected as the validation group （n＝30）. The model group was divided into the CRKP subgroup （n＝56） and the carbapenem- sensitive K. pneumoniae （CSKP） subgroup （n＝167） based on whether CRKP was detected or not. The univariate and multivariate Logistic analyses were performed on basic information such as gender， age and comorbid underlying diseases in two subgroups of patients； independent risk factors were screened for CRKP-induced BSI， and a risk prediction model was constructed. The established model was verified with patients in the validation group as the target. RESULTS Admissioning to intensive care unit （ICU）， use of immunosuppressants， empirical use of carbapenems and empirical use of antibiotics against Gram-positive coccus were independent risk factors of CRKP-induced BSI （ORs were 3.749， 3.074， 2.909， 9.419， 95%CIs were 1.639-8.572， 1.292- 7.312， 1.180-7.717， 2.877-30.840， P＜0.05）. Based on this， a risk prediction model was established with a P value of 0.365. The AUC of the receiver operating characteristic （ROC） curve of the model was 0.848 [95%CI （0.779， 0.916）， P＜0.001]， and the critical score was 6.5. In the validation group， the overall accuracy of the prediction under the model was 86.67%， and the AUC of ROC curve was 0.926 [95%CI （0.809， 1.000]， P＜0.001]. CONCLUSIONS Admission to ICU， use of immunosuppressants， empirical use of carbapenems and empirical use of antibiotics against Gram-positive coccus are independent risk factors of CRKP- induced BSI. The CRKP-induced BSI risk prediction model based on the above factors has good prediction accuracy.

PCorteSoft - Sistema para la determinación del punto de corte óptimo en escalas predictivas y la categorización de variables continuas

Al desarrollar modelos de predicción para su aplicación en la práctica clínica, los profesionales de la salud suelen categorizar las variables clínicas que son de naturaleza continua. En muchas ocasiones estos modelos constituyen la base para la confección de escalas predictivas, a partir de las cuales se estratifica a los pacientes en varias categorías atendiendo al fenómeno estudiado. En estos casos se requiere la determinación de uno o varios puntos de cortes que permitan dividir el recorrido de la variable, variables continuas o puntuaciones de una escala, en dos o más categorías. El presente trabajo tiene como objetivo la automatización de diferentes métodos para dicotomizar variables continuas en modelos de predicción clínica, donde la variable respuesta es dicotómica, y determinar el punto de corte óptimo en la estratificación de pacientes en dos categorías, a partir de escalas de predicción. Para ello se elaboró un software en el lenguaje de programación R, que implementa diferentes métodos para la determinación del punto de corte óptimo, lo cual agiliza el trabajo investigativo de los especialistas de salud en el proceso de elaboración de modelos predictivos y/o escalas de predicción.

When developing predictive models for application in clinical practice, health professionals often categorize clinical variables that are continuous in nature. In many cases, these models are the basis for the development of predictive scales from which patients are stratified into various categories according to the phenomenon under study. In both cases, it is necessary to determine one or more cut-off points that allow dividing the path of the variable, continuous variables, or scores of a scale into two or more categories. The aim of the present work is to automate different existing methods for dichotomizing continuous variables in clinical prediction models where the response variable is dichotomous, as well as to determine the optimal cut-off point for stratifying patients into two categories, based on prediction scales. For this purpose, a software was developed in the R programming language, which implements different existing methods for the determination of the optimal cut-off point, speeding up the research work of health specialists in the process of developing predictive models and/or prediction scales.

Prediction of Preterm Delivery Using Serum Ischemia Modified Albumin, Biglycan, and Decorin Levels in Women with Threatened Preterm Labor / Previsão de parto prematuro usando albumina modificada por isquemia sérica, biglicano e níveis de decorina em mulheres com ameaça de trabalho de parto prematuro

Abstract Objective The serum ischemia modified albumin (IMA), biglycan, and decorin levels of pregnant women who were hospitalized for threatened preterm labor were measured. Methods Fifty-one consecutive pregnant women with a single pregnancy between the 24th and 36th weeks with a diagnosis of threatened preterm labor were included in the present prospective cohort study. Results As a result of multivariate logistic regression analysis for predicting preterm delivery within 24 hours, 48 hours, 7 days, 14 days, ≤ 35 gestational weeks, and ≤ 37 gestational weeks after admission, area under the curve (AUC) (95% confidence interval [CI[) values were 0.95 (0.89-1.00), 0.93 (0.86-0.99), 0.91 (0.83-0.98), 0.92 (0.85-0.99), 0.82 (0.69-0.96), and 0.89 (0.80-0.98), respectively. In the present study, IMA and biglycan levels were found to be higher and decorin levels lower in women admitted to the hospital with threatened preterm labor and who gave preterm birth within 48 hours compared with those who gave birth after 48 hours. Conclusion In pregnant women admitted to the hospital with threatened preterm labor, the prediction preterm delivery of the combined model created by adding IMA, decorin, and biglycan in addition to the TVS CL measurement was higher than the TVS CL measurement alone. Clinical trial registration The present trial was registered at ClinicalTrials.gov, number NCT04451928.

Resumo Objetivo Medir os níveis séricos de albumina modificada por isquemia (IMA), biglicano e decorina de gestantes hospitalizadas por ameaça de parto prematuro. Métodos Cinquenta e uma mulheres grávidas consecutivas com uma única gravidez entre a 24ᵃ e a 36ᵃ semanas com diagnóstico de ameaça de trabalho de parto prematuro foram incluídas no presente estudo de corte prospectivo. Resultados Como resultado da análise de regressão logística multivariada para prever parto prematuro dentro de 24 horas, 48 horas, 7 dias, 14 dias, ≤ 35 semanas gestacionais e ≤ 37 semanas gestacionais após a admissão, área sob a curva (AUC) (95% de confiança os valores de intervalo [CI[) foram 0,95 (0,89-1,00), 0,93 (0,86-0,99), 0,91 (0,83-0,98), 0,92 (0,85-0,99), 0,82 (0,69-0,96) e 0,89 (0,80-0,98), respectivamente. No presente estudo, os níveis de IMA e biglican foram maiores e os níveis de decorin menores em mulheres admitidas no hospital com ameaça de trabalho de parto prematuro e que tiveram parto prematuro em 48 horas em comparação com aquelas que deram à luz após 48 horas. Conclusão Em gestantes admitidas no hospital com ameaça de trabalho de parto prematuro, a predição de parto prematuro do modelo combinado criado pela adição de IMA, decorin e biglican, além da medição do TVS CL, foi maior do que a medição do TVS CL isoladamente. Registro do ensaio clínico O presente ensaio foi registrado em ClinicalTrials.gov, número NCT04451928.

Diseño y validación de una escala de riesgo de parto pretérmino con enfoque periodontal / Design and validation of a preterm birth risk scale with a periodontal approach i: en

Introducción: La escala de riesgo diseñada para estimar la probabilidad de parto pretérmino con enfoque periodontal debe ser validada antes de su implementación en la práctica clínica. Objetivo: Diseñar y validar una escala de riesgo de parto pretérmino con enfoque periodontal. Métodos: Se realizó un estudio analítico, de casos y controles, de 1152 puérperas ingresadas en los hospitales maternos de la provincia de Santiago de Cuba en el período 2011-2022, para lo cual fueron seleccionadas 2 muestras: una de construcción del modelo (n=750) y otra de validación de la escala (n=402). Se determinaron los posibles predictores a través del análisis univariado y el cálculo del odds ratio, con un nivel de significación de p≤0,05; asimismo, se elaboró un modelo de regresión logística binaria multivariada y se obtuvo la escala de riesgo que fue validada por diferentes métodos. Resultados: La escala se obtuvo con 7 predictores y 2 estratos de riesgo. Esta alcanzó buena discriminación (80 %), así como buen nivel de ajuste y validez de constructo (p=0,72). Igualmente, aseguró una predicción correcta de más de 50 % de los partos pretérmino, valores de sensibilidad y especificidad aceptables (79,20 y 70,20 %, respectivamente), así como validez de contenido, validez interna y confiabilidad adecuadas. Conclusiones: La escala de riesgo para estratificar el riesgo de parto pretérmino incluye predictores de gravedad de la enfermedad periodontal, con buenos parámetros de validación para ser usada en la toma de decisiones para prevenir este tipo de parto.

Introduction: The risk scale designed to estimate the probability of preterm birth with periodontal approach should be validated before its implementation in the clinical practice. Objective: To design and validate a risk scale of preterm birth with periodontal approach. Methods: A cases and controls analytic study of 1152 newly-delivered women admitted to maternal hospitals in Santiago de Cuba province was carried out in the period 2011 - 2022, and 2 samples were selected: one of pattern construction (n=750) and another of scale validation(n=402). The possible predictors were determined through the single varied analysis and odds ratio calculation, with a significance level of p≤0.05; also, a multivariate binary logistical regression model was elaborated and the risk scale was obtained, which was validated by different methods. Results: The scale was obtained with 7 predictors and 2 risk stratum. It reached a good discrimination (80%), as well as a good adjustment level and construction validity (p=0.72). Likewise, it assured a correct prediction of more than 50% of preterm births, acceptable sensibility and specificity values (79.20 and 70.20%, respectively), as well as adequate content validity, internal validity and reliability. Conclusions: The risk scale to stratify the risk of preterm birth includes predictors of periodontal disease severity, with good validation parameters to be used in the decisions making to prevent this type of childbirth.

Escala predictiva de mortalidad por afección cardiovascular en adultos mayores del municipio de Santiago de Cuba / Predictive scale of mortality due to cardiovascular affection in elderly from Santiago de Cuba municipality i: en

Introducción: En el contexto del envejecimiento saludable, la prevención de las enfermedades cardiovasculares es un desafío. Objetivo: Diseñar una escala, basada en factores de riesgo identificados, para la predicción de la mortalidad por afección cardiovascular en ancianos. Métodos: Se realizó un estudio analítico, observacional y retrospectivo, de tipo caso-control, que incluyó a 536 ancianos mayores de 60 años de edad pertenecientes a 3 áreas de salud (policlínicos Ramón López Peña, Municipal y 28 de Septiembre) del municipio de Santiago de Cuba, durante el 2021. La escala se derivó del análisis de regresión logística binaria de los factores de riesgo cardiovascular identificados. Resultados: En la medida que aumentó el puntaje de la escala, se incrementó el porcentaje de pacientes fallecidos. El área bajo la curva fue de 0,836 (p= 0,000) y en la prueba de bondad de ajuste de Hosmer-Lemeshow se obtuvo p= 1,000. Conclusiones: La escala propuesta permitió predecir la mortalidad por afección cardiovascular en adultos mayores, con muy buena capacidad de discriminación y calibración.

Introduction: In the context of healthy aging, the prevention of cardiovascular diseases is a challenge. Objective: To design a scale, based on identified risk factors, for predicting mortality due to cardiovascular affection in elderly. Methods: An analytic, observational and retrospective case-control study was carried out that included 536 elderly over 60 years, belonging to 3 health areas (Ramón López Peña, Municipal and 28 de Septiembre polyclinics) from Santiago de Cuba municipality, during 2021. The scale was derived from the analysis of binary logistical regression of the identified cardiovascular risk factors. Results: As the score of the scale increased, the percentage of dead patients increased. The area under the curve was of 0.836 (p = 0.000) and in the Hosmer-Lemeshow test was obtained p = 1.000. Conclusions: The proposed scale allowed predicting mortality due to cardiovascular affection in elderly, with very good discrimination capacity and calibration.