Time series analysis of hemorrhagic fever with renal syndrome in mainland China by using an XGBoost forecasting model

AI-Enhanced Prediction of Aortic Stenosis Progression: Insights From the PROGRESSA Study

Background

Aortic valve stenosis (AS) is a progressive chronic disease with progression rates that vary in patients and therefore difficult to predict.

Objectives

The aim of this study was to predict the progression of AS using comprehensive and longitudinal patient data.

Methods

Machine and deep learning algorithms were trained on a data set of 303 patients enrolled in the PROGRESSA (Metabolic Determinants of the Progression of Aortic Stenosis) study who underwent clinical and echocardiographic follow-up on an annual basis. Performance of the models was measured to predict disease progression over long (next 5 years) and short (next 2 years) terms and was compared to a standard clinical model with usually used features in clinical settings based on logistic regression.

Results

For each annual follow-up visit including baseline, we trained various supervised learning algorithms in predicting disease progression at 2- and 5-year terms. At both terms, LightGBM consistently outperformed other models with the highest average area under curves across patient visits (0.85 at 2 years, 0.83 at 5 years). Recurrent neural network-based models (Gated Recurrent Unit and Long Short-Term Memory) and XGBoost also demonstrated strong predictive capabilities, while the clinical model showed the lowest performance.

Conclusions

This study demonstrates how an artificial intelligence-guided approach in clinical routine could help enhance risk stratification of AS. It presents models based on multisource comprehensive data to predict disease progression and clinical outcomes in patients with mild-to-moderate AS at baseline.

A machine learning model for early candidemia prediction in the intensive care unit: Clinical application

Candidemia often poses a diagnostic challenge due to the lack of specific clinical features, and delayed antifungal therapy can significantly increase mortality rates, particularly in the intensive care unit (ICU). This study aims to develop a machine learning predictive model for early candidemia diagnosis in ICU patients, leveraging their clinical information and findings. We conducted this study with a cohort of 334 patients admitted to the ICU unit at Ji Ning NO.1 people's hospital in China from Jan. 2015 to Dec. 2022. To ensure the model's reliability, we validated this model with an external group consisting of 77 patients from other sources. The candidemia to bacteremia ratio is 1:1. We collected relevant clinical procedures and eighteen key examinations or tests features to support the recursive feature elimination (RFE) algorithm. These features included total bilirubin, age, platelet count, hemoglobin, CVC, lymphocyte, Duration of stay in ICU and so on. To construct the candidemia diagnosis model, we employed random forest (RF) algorithm alongside other machine learning methods and conducted internal and external validation with training and testing sets allocated in a 7:3 ratio. The RF model demonstrated the highest area under the receiver operating characteristic (AUC) with values of 0.87 and 0.83 for internal and external validation, respectively. To evaluate the importance of features in predicting candidemia, Shapley additive explanation (SHAP) values were calculated and results revealed that total bilirubin and age were the most important factors in the prediction model. This advancement in candidemia prediction holds significant promise for early intervention and improved patient outcomes in the ICU setting, where timely diagnosis is of paramount crucial.

A dynamic ensemble model for short-term forecasting in pandemic situations

During the COVID-19 pandemic, many hospitals reached their capacity limits and could no longer guarantee treatment of all patients. At the same time, governments endeavored to take sensible measures to stop the spread of the virus while at the same time trying to keep the economy afloat. Many models extrapolating confirmed cases and hospitalization rate over short periods of time have been proposed, including several ones coming from the field of machine learning. However, the highly dynamic nature of the pandemic with rapidly introduced interventions and new circulating variants imposed non-trivial challenges for the generalizability of such models. In the context of this paper, we propose the use of ensemble models, which are allowed to change in their composition or weighting of base models over time and could thus better adapt to highly dynamic pandemic or epidemic situations. In that regard, we also explored the use of secondary metadata-Google searches-to inform the ensemble model. We tested our approach using surveillance data from COVID-19, Influenza, and hospital syndromic surveillance of severe acute respiratory infections (SARI). In general, we found ensembles to be more robust than the individual models. Altogether we see our work as a contribution to enhance the preparedness for future pandemic situations.

Enhancing infectious disease prediction model selection with multi-objective optimization: an empirical study

As the pandemic continues to pose challenges to global public health, developing effective predictive models has become an urgent research topic. This study aims to explore the application of multi-objective optimization methods in selecting infectious disease prediction models and evaluate their impact on improving prediction accuracy, generalizability, and computational efficiency. In this study, the NSGA-II algorithm was used to compare models selected by multi-objective optimization with those selected by traditional single-objective optimization. The results indicate that decision tree (DT) and extreme gradient boosting regressor (XGBoost) models selected through multi-objective optimization methods outperform those selected by other methods in terms of accuracy, generalizability, and computational efficiency. Compared to the ridge regression model selected through single-objective optimization methods, the decision tree (DT) and XGBoost models demonstrate significantly lower root mean square error (RMSE) on real datasets. This finding highlights the potential advantages of multi-objective optimization in balancing multiple evaluation metrics. However, this study's limitations suggest future research directions, including algorithm improvements, expanded evaluation metrics, and the use of more diverse datasets. The conclusions of this study emphasize the theoretical and practical significance of multi-objective optimization methods in public health decision support systems, indicating their wide-ranging potential applications in selecting predictive models.

The Spectrum of Biopsy-Proven Kidney Diseases, Causes, and Renal Outcomes in Acute Kidney Injury Patients

INTRODUCTION

Acute kidney injury (AKI) is a group of highly heterogeneous, complicated clinical syndromes. Although kidney biopsy plays an irreplaceable role in evaluating complex AKI, a few studies have focused on the clinicopathology of AKI biopsies. This study analyzed the pathological disease spectrum, causes, and renal outcomes of biopsied AKI patients.

METHODS

We retrospectively included 2,027 AKI patients who underwent kidney biopsies at a national clinical research center of kidney diseases from 2013 through 2018. To compare the biopsied AKI cases without and with coexisting glomerulopathy, patients were classified into acute tubular/tubulointerstitial nephropathy-associated AKI (ATIN-AKI) and glomerular disease-associated AKI (GD-AKI) groups.

RESULTS

Of 2,027 biopsied AKI patients, 65.1% were male, with a median age of 43 years. A total of 1,590 (78.4%) patients had coexisting GD, while only 437 (21.6%) patients had ATIN alone. The AKI patients with GD mainly (53.5%) manifested as stage 1 AKI, while most ATIN-AKI patients (74.8%) had stage 3 AKI. In the ATIN-AKI group, 256 (58.6%) patients had acute interstitial nephritis (AIN), and 77 (17.6%) had acute tubular injury (ATI). ATIN-AKI was mainly caused by drugs in 85.5% of AIN and 63.6% of ATI cases, respectively. In AKI patients with coexisting GD, the leading pathological diagnoses in over 80% of patients were IgA nephropathy (IgAN, 22.5%), minimal change disease (MCD, 17.5%), focal segmental glomerulosclerosis (FSGS, 15.3%), lupus nephritis (LN, 11.9%), membranous nephropathy (MN, 10.2%), and ANCA-associated vasculitis (AAV, 4.7%). A total of 775 patients were followed up within 3 months after renal biopsy; ATIN-AKI patients achieved statistically higher complete renal recovery than the GD-AKI patients (83.5% vs. 70.5%, p < 0.001).

CONCLUSIONS

Most biopsied AKI patients have coexisting GD, while ATIN alone is seen less frequently. ATIN-AKI is mainly caused by drugs. In GD-AKI patients, IgAN, MCD, FSGS, LN, MN, and AAV are the leading diagnoses. Compared to AKI patients without GD, patients with GD suffer from worse renal function recovery.

Exploring the influence of COVID-19 on the spread of hand, foot, and mouth disease with an automatic machine learning prediction model

Hand, foot, and mouth disease (HFMD) is an important public health problem and has received concern worldwide. Moreover, the coronavirus disease 2019 (COVID-19) epidemic also increases the difficulty of understanding and predicting the prevalence of HFMD. The purpose is to prove the usability and applicability of the automatic machine learning (Auto-ML) algorithm in predicting the epidemic trend of HFMD and to explore the influence of COVID-19 on the spread of HFMD. The AutoML algorithm and the autoregressive integrated moving average (ARIMA) model were applied to construct and validate models, based on the monthly incidence numbers of HFMD and meteorological factors from May 2008 to December 2019 in Henan province, China. A total of four models were established, among which the Auto-ML model with meteorological factors had minimum RMSE and MAE in both the model constructing phase and forecasting phase (training set: RMSE = 1424.40 and MAE = 812.55; test set: RMSE = 2107.83, MAE = 1494.41), so this model has the best performance. The optimal model was used to further predict the incidence numbers of HFMD in 2020 and then compared with the reported cases. And, for analysis, 2020 was divided into two periods. The predicted incidence numbers followed the same trend as the reported cases of HFMD before the COVID-19 outbreak; while after the COVID-19 outbreak, the reported cases have been greatly reduced than expected, with an average of only about 103 cases per month, and the incidence peak has also been delayed, which has led to significant changes in the seasonality of HFMD. Overall, the AutoML algorithm is an applicable and ideal method to predict the epidemic trend of the HFMD. Furthermore, it was found that the countermeasures of COVID-19 have a certain influence on suppressing the spread of HFMD during the period of COVID-19. The findings are helpful to health administrative departments.

The research on TBATS and ELM models for prediction of human brucellosis cases in mainland China: a time series study

BACKGROUND

Human brucellosis is a serious public health concern in China. The objective of this study is to develop a suitable model for forecasting human brucellosis cases in mainland China.

METHODS

Data on monthly human brucellosis cases from January 2012 to December 2021 in 31 provinces and municipalities in mainland China were obtained from the National Health Commission of the People's Republic of China website. The TBATS and ELM models were constructed. The MAE, MSE, MAPE, and RMSE were calculated to evaluate the prediction performance of the two models.

RESULTS

The optimal TBATS model was TBATS (1, {0,0}, -, {< 12,4 >}) and the lowest AIC value was 1854.703. In the optimal TBATS model, {0,0} represents the ARIMA (0,0) model, {< 12,4 >} are the parameters of the seasonal periods and the corresponding number of Fourier terms, respectively, and the parameters of the Box-Cox transformation ω are 1. The optimal ELM model hidden layer number was 33 and the R-squared value was 0.89. The ELM model provided lower values of MAE, MSE, MAPE, and RMSE for both the fitting and forecasting performance.

CONCLUSIONS

The results suggest that the forecasting performance of ELM model outperforms the TBATS model in predicting human brucellosis between January 2012 and December 2021 in mainland China. Forecasts of the ELM model can help provide early warnings and more effective prevention and control measures for human brucellosis in mainland China.

Blood transcriptome analysis: Ferroptosis and potential inflammatory pathways in post-traumatic stress disorder

BACKGROUND

Transcriptome-wide analysis of peripheral blood in post-traumatic stress disorder (PTSD) indicates widespread changes in immune-related pathways and function. Ferroptosis, an iron-dependent regulated cell death, is closely related to oxidative stress. However, little is known as to whether ferroptosis plays a role in PTSD.

METHODS

We conducted a comprehensive analysis of combined data from six independent peripheral blood transcriptional studies in the Gene Expression Omnibus (GEO) database, covering PTSD and control individuals. Differentially expressed genes (DEGs) were extracted by comparing PTSD patients with control individuals, from which 29 ferroptosis-related genes (FRGs) were cross-matched and obtained. The weighted gene co-expression network analysis (WGCNA), the Extreme Gradient Boosting (XGBoost) model with Bayesian Optimization, and the least absolute shrinkage and selection operator (LASSO) Cox regression were utilized to construct a PTSD prediction model. Single-sample Gene Set Enrichment Analysis (ssGSEA) and CIBERSORT revealed the disturbed immunologic state in PTSD high-risk patients.

RESULTS

Three crucial FRGs (ACSL4, ACO1, and GSS) were identified and used to establish a predictive model of PTSD. The receiver operating characteristic (ROC) curve verifies its risk prediction ability. Remarkably, ssGSEA and CIBERSORT demonstrated changes in cellular immunity and antigen presentation depending on the FRGs model.

CONCLUSION

These findings collectively provide evidence that ferroptosis may change immune status in PTSD and be related to the occurrence of PTSD, which may help delineate mechanisms and discover treatment biomarkers for PTSD.

Estimating the Effects of the COVID-19 Outbreak on the Reductions in Tuberculosis Cases and the Epidemiological Trends in China: A Causal Impact Analysis

Objective

COVID-19 may have a demonstrable influence on disease patterns. However, it remained unknown how tuberculosis (TB) epidemics are impacted by the COVID-19 outbreak. The purposes of this study are to evaluate the impacts of the COVID-19 outbreak on the decreases in the TB case notifications and to forecast the epidemiological trends in China.

Methods

The monthly TB incidents from January 2005 to December 2020 were taken. Then, we investigated the causal impacts of the COVID-19 pandemic on the TB case reductions using intervention analysis under the Bayesian structural time series (BSTS) method. Next, we split the observed values into different training and testing horizons to validate the forecasting performance of the BSTS method.

Results

The TB incidence was falling during 2005–2020, with an average annual percentage change of −3.186 (95% confidence interval [CI] −4.083 to −2.281), and showed a peak in March–April and a trough in January–February per year. The BSTS method assessed a monthly average reduction of 14% (95% CI 3.8% to 24%) in the TB case notifications from January–December 2020 owing to COVID-19 (probability of causal effect=99.684%, P=0.003), and this method generated a highly accurate forecast for all the testing horizons considering the small forecasting error rates and estimated a continued downward trend from 2021 to 2035 (annual percentage change =−2.869, 95% CI −3.056 to −2.681).

Conclusion

COVID-19 can cause medium- and longer-term consequences for the TB epidemics and the BSTS model has the potential to forecast the epidemiological trends of the TB incidence, which can be recommended as an automated application for public health policymaking in China. Considering the slow downward trend in the TB incidence, additional measures are required to accelerate the progress of the End TB Strategy.