Prediction of 3-year risk of diabetic kidney disease using machine learning based on electronic medical records

Machine learning-based prediction of in-hospital mortality for critically ill patients with sepsis-associated acute kidney injury

OBJECTIVES

This study aims to develop and validate a prediction model in-hospital mortality in critically ill patients with sepsis-associated acute kidney injury (SA-AKI) based on machine learning algorithms.

METHODS

Patients who met the criteria for inclusion were identified in the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database and divided according to the validation (n = 2440) and development (n = 9756, 80%) queues. Ensemble stepwise feature selection method was used to screen for effective features. The prediction models of short-term mortality were developed by seven machine learning algorithms. Ten-fold cross-validation was used to verify the performance of the algorithm in the development queue. The area under the receiver operating characteristic curve (ROC-AUC) was used to evaluate the differentiation accuracy and performance of the prediction model in the validation queue. The best-performing model was interpreted by Shapley additive explanations (SHAP).

RESULTS

A total of 12,196 patients were enrolled in this study. Eleven variables were finally chosen to develop the prediction model. The AUC of the random forest (RF) model was the highest value both in the Ten-fold cross-validation and evaluation (AUC: 0.798, 95% CI: 0.774-0.821). According to the SHAP plots, old age, low Glasgow Coma Scale (GCS) score, high AKI stage, reduced urine output, high Simplified Acute Physiology Score (SAPS II), high respiratory rate, low temperature, low absolute lymphocyte count, high creatinine level, dysnatremia, and low body mass index (BMI) increased the risk of poor prognosis.

CONCLUSIONS

The RF model developed in this study is a good predictor of in-hospital mortality for patients with SA-AKI in the intensive care unit (ICU), which may have potential applications in mortality prediction.

Effects of Various Heavy Metal Exposures on Insulin Resistance in Non-diabetic Populations: Interpretability Analysis from Machine Learning Modeling Perspective

Increasing and compelling evidence has been proved that heavy metal exposure is involved in the development of insulin resistance (IR). We trained an interpretable predictive machine learning (ML) model for IR in the non-diabetic populations based on levels of heavy metal exposure. A total of 4354 participants from the NHANES (2003-2020) with complete information were randomly divided into a training set and a test set. Twelve ML algorithms, including random forest (RF), XGBoost (XGB), logistic regression (LR), GaussianNB (GNB), ridge regression (RR), support vector machine (SVM), multilayer perceptron (MLP), decision tree (DT), AdaBoost (AB), Gradient Boosting Decision Tree (GBDT), Voting Classifier (VC), and K-Nearest Neighbour (KNN), were constructed for IR prediction using the training set. Among these models, the RF algorithm had the best predictive performance, showing an accuracy of 80.14%, an AUC of 0.856, and an F1 score of 0.74 in the test set. We embedded three interpretable methods, the permutation feature importance analysis, partial dependence plot (PDP), and Shapley additive explanations (SHAP) in RF model for model interpretation. Urinary Ba, urinary Mo, blood Pb, and blood Cd levels were identified as the main influencers of IR. Within a specific range, urinary Ba (0.56-3.56 µg/L) and urinary Mo (1.06-20.25 µg/L) levels exhibited the most pronounced upwards trend with the risk of IR, while blood Pb (0.05-2.81 µg/dL) and blood Cd (0.24-0.65 µg/L) levels showed a declining trend with IR. The findings on the synergistic effects demonstrated that controlling urinary Ba levels might be more crucial for the management of IR. The SHAP decision plot offered personalized care for IR based on heavy metal control. In conclusion, by utilizing interpretable ML approaches, we emphasize the predictive value of heavy metals for IR, especially Ba, Mo, Pb, and Cd.

Trends in the prevalence of osteoporosis and effects of heavy metal exposure using interpretable machine learning

There is limited evidence that heavy metals exposure contributes to osteoporosis. Multi-parameter scoring machine learning (ML) techniques were developed using National Health and Nutrition Examination Survey data to predict osteoporosis based on heavy metal exposure levels. For generating an optimal predictive model for osteoporosis, 12 ML models were used. Identification was carried out using the model that performed the best. For interpretation of models, Shapley additive explanation (SHAP) methods and partial dependence plots (PDP) were integrated into a pipeline and incorporated into the ML pipeline. By regressing osteoporosis on survey cycles, logistic regression was used to evaluate linear trends in osteoporosis over time. For the purpose of training and validating predictive models, 5745 eligible participants were randomly selected into training and testing set. It was evident from the results that the gradient boosting decision tree model performed the best among the predictive models, attributing to an accuracy rate of 89.40 % in the testing set. Based on the model results, the area under the curve and F1 score were 0.88 and 0.39, respectively. As a result of the SHAP analysis, urinary Co, urinary Tu, blood Cd, and urinary Hg levels were identified as the most influential factors influencing osteoporosis. Urinary Co (0.20-6.10 μg/mg creatinine), urinary Tu (0.06-1.93 μg/mg creatinine), blood Cd (0.07-0.50 μg/L), and urinary Hg (0.06-0.75 μg/mg creatinine) levels displayed a distinctive upward trend with risk of osteoporosis as values increased. Our analysis revealed that urinary Co, urinary Tu, blood Cd, and urinary Hg played a significant role in predictability.

Revolutionary Point-of-Care Wearable Diagnostics for Early Disease Detection and Biomarker Discovery through Intelligent Technologies

Early-stage disease detection, particularly in Point-Of-Care (POC) wearable formats, assumes pivotal role in advancing healthcare services and precision-medicine. Public benefits of early detection extend beyond cost-effectively promoting healthcare outcomes, to also include reducing the risk of comorbid diseases. Technological advancements enabling POC biomarker recognition empower discovery of new markers for various health conditions. Integration of POC wearables for biomarker detection with intelligent frameworks represents ground-breaking innovations enabling automation of operations, conducting advanced large-scale data analysis, generating predictive models, and facilitating remote and guided clinical decision-making. These advancements substantially alleviate socioeconomic burdens, creating a paradigm shift in diagnostics, and revolutionizing medical assessments and technology development. This review explores critical topics and recent progress in development of 1) POC systems and wearable solutions for early disease detection and physiological monitoring, as well as 2) discussing current trends in adoption of smart technologies within clinical settings and in developing biological assays, and ultimately 3) exploring utilities of POC systems and smart platforms for biomarker discovery. Additionally, the review explores technology translation from research labs to broader applications. It also addresses associated risks, biases, and challenges of widespread Artificial Intelligence (AI) integration in diagnostics systems, while systematically outlining potential prospects, current challenges, and opportunities.

From bytes to nephrons: AI's journey in diabetic kidney disease

Diabetic kidney disease (DKD) is a significant complication of type 2 diabetes, posing a global health risk. Detecting and predicting diabetic kidney disease at an early stage is crucial for timely interventions and improved patient outcomes. Artificial intelligence (AI) has demonstrated promise in healthcare, and several tools have recently been developed that utilize Machine Learning with clinical data to detect and predict DKD. This review aims to explore the current landscape of AI and machine learning applications in DKD, specifically examining existing literature on risk scores and machine learning approaches for predicting DKD development. A literature search was conducted using Medline (PubMed), Google Scholar, and Scopus databases until July 2023. Relevant keywords were used to extract studies that described the role of AI in DKD. The review revealed that AI and machine learning have been successfully used to predict DKD progression, outperforming traditional risk score models. Artificial intelligence-driven research for DKD extends beyond prediction models, offering opportunities for integrating genetic and epigenetic data, advancing understanding of the disease's molecular basis, personalizing treatment strategies, and fostering the development of novel drugs. However, challenges remain, including the requirement for large datasets and the lack of standardization in AI-driven tools for DKD. Artificial intelligence and machine learning have the potential to revolutionize the management and care of DKD patients, surpassing the limitations of traditional methods reliant on existing knowledge. Future research should address the challenges associated with AI and machine learning in DKD and focus on developing AI-driven tools for clinical practice.

An Ensemble Machine Learning and Data Mining Approach to Enhance Stroke Prediction

Stroke poses a significant health threat, affecting millions annually. Early and precise prediction is crucial to providing effective preventive healthcare interventions. This study applied an ensemble machine learning and data mining approach to enhance the effectiveness of stroke prediction. By employing the cross-industry standard process for data mining (CRISP-DM) methodology, various techniques, including random forest, ExtraTrees, XGBoost, artificial neural network (ANN), and genetic algorithm with ANN (GANN) were applied on two benchmark datasets to predict stroke based on several parameters, such as gender, age, various diseases, smoking status, BMI, HighCol, physical activity, hypertension, heart disease, lifestyle, and others. Due to dataset imbalance, Synthetic Minority Oversampling Technique (SMOTE) was applied to the datasets. Hyperparameter tuning optimized the models via grid search and randomized search cross-validation. The evaluation metrics included accuracy, precision, recall, F1-score, and area under the curve (AUC). The experimental results show that the ensemble ExtraTrees classifier achieved the highest accuracy (98.24%) and AUC (98.24%). Random forest also performed well, achieving 98.03% in both accuracy and AUC. Comparisons with state-of-the-art stroke prediction methods revealed that the proposed approach demonstrates superior performance, indicating its potential as a promising method for stroke prediction and offering substantial benefits to healthcare.

Machine learning prediction models for diabetic kidney disease: systematic review and meta-analysis

BACKGROUND

Machine learning is increasingly recognized as a viable approach for identifying risk factors associated with diabetic kidney disease (DKD). However, the current state of real-world research lacks a comprehensive systematic analysis of the predictive performance of machine learning (ML) models for DKD.

OBJECTIVES

The objectives of this study were to systematically summarize the predictive capabilities of various ML methods in forecasting the onset and the advancement of DKD, and to provide a basic outline for ML methods in DKD.

METHODS

We have searched mainstream databases, including PubMed, Web of Science, Embase, and MEDLINE databases to obtain the eligible studies. Subsequently, we categorized various ML techniques and analyzed the differences in their performance in predicting DKD.

RESULTS

Logistic regression (LR) was the prevailing ML method, yielding an overall pooled area under the receiver operating characteristic curve (AUROC) of 0.83. On the other hand, the non-LR models also performed well with an overall pooled AUROC of 0.80. Our t-tests showed no statistically significant difference in predicting ability between LR and non-LR models (t = 1.6767, p > 0.05).

CONCLUSION

All ML predicting models yielded relatively satisfied DKD predicting ability with their AUROCs greater than 0.7. However, we found no evidence that non-LR models outperformed the LR model. LR exhibits high performance or accuracy in practice, while it is known for algorithmic simplicity and computational efficiency compared to others. Thus, LR may be considered a cost-effective ML model in practice.

Machine learning techniques to predict the risk of developing diabetic nephropathy: a literature review

Purpose

Diabetes is a major public health challenge with widespread prevalence, often leading to complications such as Diabetic Nephropathy (DN)-a chronic condition that progressively impairs kidney function. In this context, it is important to evaluate if Machine learning models can exploit the inherent temporal factor in clinical data to predict the risk of developing DN faster and more accurately than current clinical models.

Methods

Three different databases were used for this literature review: Scopus, Web of Science, and PubMed. Only articles written in English and published between January 2015 and December 2022 were included.

Results

We included 11 studies, from which we discuss a number of algorithms capable of extracting knowledge from clinical data, incorporating dynamic aspects in patient assessment, and exploring their evolution over time. We also present a comparison of the different approaches, their performance, advantages, disadvantages, interpretation, and the value that the time factor can bring to a more successful prediction of diabetic nephropathy.

Conclusion

Our analysis showed that some studies ignored the temporal factor, while others partially exploited it. Greater use of the temporal aspect inherent in Electronic Health Records (EHR) data, together with the integration of omics data, could lead to the development of more reliable and powerful predictive models.

Prediction and analysis of risk factors for diabetic retinopathy based on machine learning and interpretable models

Objective

Diabetic retinopathy is one of the major complications of diabetes. In this study, a diabetic retinopathy risk prediction model integrating machine learning models and SHAP was established to increase the accuracy of risk prediction for diabetic retinopathy, explain the rationality of the findings from model prediction and improve the reliability of prediction results.

Methods

Data were preprocessed for missing values and outliers, features selected through information gain, a diabetic retinopathy risk prediction model established using the CatBoost and the outputs of the mode interpreted using the SHAP model.

Results

One thousand early warning data of diabetes complications derived from diabetes complication early warning dataset from the National Clinical Medical Sciences Data Center were used in this study. The CatBoost-based model for diabetic retinopathy prediction performed the best in the comparative model test. ALB_CR, HbA1c, UPR_24, NEPHROPATHY and SCR were positively correlated with diabetic retinopathy, while CP, HB, ALB, DBILI and CRP were negatively correlated with diabetic retinopathy. The relationships between HEIGHT, WEIGHT and ESR characteristics and diabetic retinopathy were not significant.

Conclusion

The risk factors for diabetic retinopathy include poor renal function, elevated blood glucose level, liver disease, hematonosis and dysarteriotony, among others. Diabetic retinopathy can be prevented by monitoring and effectively controlling relevant indices. In this study, the influence relationships between the features were also analyzed to further explore the potential factors of diabetic retinopathy, which can provide new methods and new ideas for the early prevention and clinical diagnosis of subsequent diabetic retinopathy.

Machine Learning Models for Prediction of Diabetic Microvascular Complications

IMPORTANCE AND AIMS

Diabetic microvascular complications significantly impact morbidity and mortality. This review focuses on machine learning/artificial intelligence (ML/AI) in predicting diabetic retinopathy (DR), diabetic kidney disease (DKD), and diabetic neuropathy (DN).

METHODS

A comprehensive PubMed search from 1990 to 2023 identified studies on ML/AI models for diabetic microvascular complications. The review analyzed study design, cohorts, predictors, ML techniques, prediction horizon, and performance metrics.

RESULTS

Among the 74 identified studies, 256 featured internally validated ML models and 124 had externally validated models, with about half being retrospective. Since 2010, there has been a rise in the use of ML for predicting microvascular complications, mainly driven by DKD research across 27 countries. A more modest increase in ML research on DR and DN was observed, with publications from fewer countries. For all microvascular complications, predictive models achieved a mean (standard deviation) c-statistic of 0.79 (0.09) on internal validation and 0.72 (0.12) on external validation. Diabetic kidney disease models had the highest discrimination, with c-statistics of 0.81 (0.09) on internal validation and 0.74 (0.13) on external validation, respectively. Few studies externally validated prediction of DN. The prediction horizon, outcome definitions, number and type of predictors, and ML technique significantly influenced model performance.

CONCLUSIONS AND RELEVANCE

There is growing global interest in using ML for predicting diabetic microvascular complications. Research on DKD is the most advanced in terms of publication volume and overall prediction performance. Both DR and DN require more research. External validation and adherence to recommended guidelines are crucial.

Predicting CKD progression using time-series clustering and light gradient boosting machines

Predicting the transition of kidney function in chronic kidney disease is difficult as specific symptoms are lacking and often overlooked, and progress occurs due to complicating factors. In this study, we applied time-series cluster analysis and a light gradient boosting machine to predict the trajectories of kidney function in non-dialysis dependent chronic kidney disease patients with baseline estimated glomerular filtration rate (GFR) ≥ 45 mL/min/1.73 m2. Based on 5-year changes in estimated GFR, participants were stratified into groups with similar trajectories by cluster analysis. Next, we applied the light gradient boosting machine algorithm and Shapley addictive explanation to develop a prediction model for clusters and identify important parameters for prediction. Data from 780 participants were available for analysis. Participants were classified into five classes (Class 1: n = 78, mean [± standard deviation] estimated GFR 100 ± 19.3 mL/min/1.73 m2; Class 2: n = 176, 76.0 ± 9.3 mL/min/1.73 m2; Class 3: n = 191, 59.8 ± 5.9 mL/min/1.73 m2; Class 4: n = 261, 52.7 ± 4.6 mL/min/1.73 m2; and Class 5: n = 74, 53.5 ± 12.0 mL/min/1.73 m2). Declines in estimated GFR were 8.9% in Class 1, 12.2% in Class 2, 4.9% in Class 3, 12.0% in Class 4, and 45.1% in Class 5 during the 5-year period. The accuracy of prediction was 0.675, and the top three most important Shapley addictive explanation values were 1.61 for baseline estimated GFR, 0.12 for hemoglobin, and 0.11 for body mass index. The estimated GFR transition of patients with preserved chronic kidney disease mostly depended on baseline estimated GFR, and the borderline for estimated GFR trajectory was nearly 50 mL/min/1.73 m2.

Machine learning models for early prediction of potassium lowering effectiveness and adverse events in patients with hyperkalemia

The aim of this study was to develop a model for early prediction of adverse events and treatment effectiveness in patients with hyperkalemia. We collected clinical data from patients with hyperkalemia in the First Hospital of Zhejiang University School of Medicine between 2015 and 2021. The least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression were used to analyze the predictors on the full dataset. We randomly divided the data into a training group and a validation group, and used LASSO to filter variables in the training set. Six machine learning methods were used to develop the models. The best model was selected based on the area under the curve (AUC). Shapley additive exPlanations (SHAP) values were used to explain the best model. A total of 1074 patients with hyperkalemia were finally enrolled. Diastolic blood pressure (DBP), breathing, oxygen saturation (SPO2), Glasgow coma score (GCS), liver disease, oliguria, blood sodium, international standardized ratio (ISR), and initial blood potassium were the predictors of the occurrence of adverse events; peripheral edema, estimated glomerular filtration rate (eGFR), blood sodium, actual base residual, and initial blood potassium were the predictors of therapeutic effect. Extreme gradient boosting (XGBoost) model achieved the best performance (adverse events: AUC = 0.87; therapeutic effect: AUC = 0.75). A model based on clinical characteristics was developed and validated with good performance.

The application of explainable artificial intelligence (XAI) in electronic health record research: A scoping review

Machine Learning (ML) and Deep Learning (DL) models show potential in surpassing traditional methods including generalised linear models for healthcare predictions, particularly with large, complex datasets. However, low interpretability hinders practical implementation. To address this, Explainable Artificial Intelligence (XAI) methods are proposed, but a comprehensive evaluation of their effectiveness is currently limited. The aim of this scoping review is to critically appraise the application of XAI methods in ML/DL models using Electronic Health Record (EHR) data. In accordance with PRISMA scoping review guidelines, the study searched PUBMED and OVID/MEDLINE (including EMBASE) for publications related to tabular EHR data that employed ML/DL models with XAI. Out of 3220 identified publications, 76 were included. The selected publications published between February 2017 and June 2023, demonstrated an exponential increase over time. Extreme Gradient Boosting and Random Forest models were the most frequently used ML/DL methods, with 51 and 50 publications, respectively. Among XAI methods, Shapley Additive Explanations (SHAP) was predominant in 63 out of 76 publications, followed by partial dependence plots (PDPs) in 11 publications, and Locally Interpretable Model-Agnostic Explanations (LIME) in 8 publications. Despite the growing adoption of XAI methods, their applications varied widely and lacked critical evaluation. This review identifies the increasing use of XAI in tabular EHR research and highlights a deficiency in the reporting of methods and a lack of critical appraisal of validity and robustness. The study emphasises the need for further evaluation of XAI methods and underscores the importance of cautious implementation and interpretation in healthcare settings.

Relationships of multidimensional factors to self-management in patients with diabetes: A Cross-sectional, correlational study

Poor self-management in patients with diabetes is a consistent issue, leading to diabetes complications. The purpose of this study was to examine the relationships of demographic, cognitive, psychological, social, and physiological factors to self-management in patients with diabetes. In this cross-sectional, correlational study, data were collected from 98 patients through REDCap in 2023. Multiple regression analysis was used to address the study purpose. In the sample, adherence to adequate self-management was poor. Knowledge, self-efficacy, and body mass index (BMI) were associated with overall diabetes self-management. Self-efficacy and BMI were associated with diet and exercise. Knowledge and self-efficacy were associated with blood glucose testing. No factors were associated with smoking and foot care. In conclusion, different factors were associated with different types of self-management, but BMI and self-efficacy were associated with most types of self-management. Clinicians and researchers can target BMI, self-efficacy, and knowledge to improve self-management in patients with diabetes.

An enhanced diabetes prediction amidst COVID-19 using ensemble models

In the contemporary landscape of healthcare, the early and accurate prediction of diabetes has garnered paramount importance, especially in the wake of the COVID-19 pandemic where individuals with diabetes exhibit increased vulnerability. This research embarked on a mission to enhance diabetes prediction by employing state-of-the-art machine learning techniques. Initial evaluations highlighted the Support Vector Machines (SVM) classifier as a promising candidate with an accuracy of 76.62%. To further optimize predictions, the study delved into advanced feature engineering techniques, generating interaction and polynomial features that unearthed hidden patterns in the data. Subsequent correlation analyses, visualized through heatmaps, revealed significant correlations, especially with attributes like Glucose. By integrating the strengths of Decision Trees, Gradient Boosting, and SVM in an ensemble model, we achieved an accuracy of 93.2%, showcasing the potential of harmonizing diverse algorithms. This research offers a robust blueprint for diabetes prediction, holding profound implications for early diagnosis, personalized treatments, and preventive care in the context of global health challenges and with the goal of increasing life expectancy.

The automatic detection of diabetic kidney disease from retinal vascular parameters combined with clinical variables using artificial intelligence in type-2 diabetes patients

BACKGROUND

Diabetic kidney disease (DKD) has become the largest cause of end-stage kidney disease. Early and accurate detection of DKD is beneficial for patients. The present detection depends on the measurement of albuminuria or the estimated glomerular filtration rate, which is invasive and not optimal; therefore, new detection tools are urgently needed. Meanwhile, a close relationship between diabetic retinopathy and DKD has been reported; thus, we aimed to develop a novel detection algorithm for DKD using artificial intelligence technology based on retinal vascular parameters combined with several easily available clinical parameters in patients with type-2 diabetes.

METHODS

A total of 515 consecutive patients with type-2 diabetes mellitus from Xiangyang Central Hospital were included. Patients were stratified by DKD diagnosis and split randomly into either the training set (70%, N = 360) or the testing set (30%, N = 155) (random seed = 1). Data from the training set were used to develop the machine learning algorithm (MLA), while those from the testing set were used to validate the MLA. Model performances were evaluated.

RESULTS

The MLA using the random forest classifier presented optimal performance compared with other classifiers. When validated, the accuracy, sensitivity, specificity, F1 score, and AUC for the optimal model were 84.5%(95% CI 83.3-85.7), 84.5%(82.3-86.7), 84.5%(82.7-86.3), 0.845(0.831-0.859), and 0.914(0.903-0.925), respectively.

CONCLUSIONS

A new machine learning algorithm for DKD diagnosis based on fundus images and 8 easily available clinical parameters was developed, which indicated that retinal vascular changes can assist in DKD screening and detection.

Association between co-exposure to phenols, phthalates, and polycyclic aromatic hydrocarbons with the risk of frailty

The phenomenon of population aging has brought forth the challenge of frailty. Nevertheless, the contribution of environmental exposure to frailty remains ambiguous. Our objective was to investigate the association between phenols, phthalates (PAEs), and polycyclic aromatic hydrocarbons (PAHs) with frailty. We constructed a 48-item frailty index using data from the National Health and Nutrition Examination Survey (NHANES). The exposure levels of 20 organic contaminants were obtained from the survey circle between 2005 and 2016. The association between individual organic contaminants and the frailty index was assessed using negative binomial regression models. The combined effect of organic contaminants was examined using weighted quantile sum (WQS) regression. Dose-response patterns were modeled using generalized additive models (GAMs). Additionally, an interpretable machine learning approach was employed to develop a predictive model for the frailty index. A total of 1566 participants were included in the analysis. Positive associations were observed between exposure to MIB, P02, ECP, MBP, MHH, MOH, MZP, MC1, and P01 with the frailty index. WQS regression analysis revealed a significant increase in the frailty index with higher levels of the mixture of organic contaminants (aOR, 1.12; 95% CI, 1.05-1.20; p < 0.001), with MIB, ECP, COP, MBP, P02, and P01 identified as the major contributors. Dose-response relationships were observed between MIB, ECP, MBP, P02, and P01 exposure with an increased risk of frailty (both with p < 0.05). The developed predictive model based on organic contaminants exposure demonstrated high performance, with an R2 of 0.9634 and 0.9611 in the training and testing sets, respectively. Furthermore, the predictive model suggested potential synergistic effects in the MIB-MBP and P01-P02 pairs. Taken together, these findings suggest a significant association between exposure to phthalates and PAHs with an increased susceptibility to frailty.

Implementing a Novel Machine Learning System for Nutrition Education in Diabetes Mellitus Nutritional Clinic: Predicting 1-Year Blood Glucose Control

(1) Background: Persistent hyperglycemia in diabetes mellitus (DM) increases the risk of death and causes cardiovascular disease (CVD), resulting in significant social and economic costs. This study used a machine learning (ML) technique to build prediction models with the factors of lifestyle, medication compliance, and self-control in eating habits and then implemented a predictive system based on the best model to forecast whether blood glucose can be well-controlled within 1 year in diabetic patients attending a DM nutritional clinic. (2) Methods: Data were collected from outpatients aged 20 years or older with type 2 DM who received nutrition education in Chi Mei Medical Center. Multiple ML algorithms were used to build the predictive models. (3) Results: The predictive models achieved accuracies ranging from 0.611 to 0.690. The XGBoost model with the highest area under the curve (AUC) of 0.738 was regarded as the best and used for the predictive system implementation. SHAP analysis was performed to interpret the feature importance in the best model. The predictive system, evaluated by dietitians, received positive feedback as a beneficial tool for diabetes nutrition consultations. (4) Conclusions: The ML prediction model provides a promising approach for diabetes nutrition consultations to maintain good long-term blood glucose control, reduce diabetes-related complications, and enhance the quality of medical care.

[Validation and comparison of diabetic retinopathy-based diagnostic models for diabetic nephropathy]

OBJECTIVE

To validate and compare the efficacy of two noninvasive diagnostic models for diabetic nephropathy (DN) based on diabetic retinopathy (DR).

METHODS

A total of 565 patients with type 2 diabetes undergoing kidney biopsy in the Department of Nephrology, PLA General Hospital from January, 1993 to December, 2014 were studied. The patients were divided into DN group and non-diabetic nephropathy (NDRD) group according to renal pathological diagnosis. The data from the 22-year period were divided into 3 stages based on chronological order: early stage (from 1993 to 2003), middle stage (from 2004 to April, 2012), and late stage (from May, 2012 to December, 2014). The changes in clinical features and pathological diagnosis of the patients with renal biopsy over the 22 years were analyzed. The published DNT model and JDB model, both based on DR, were validated and compared for diagnostic effectiveness of DN, and the characteristics of the misdiagnosed cases were analyzed.

RESULTS

The incidences of hypertension and DR and levels of glycosylated hemoglobin (HbA1c), creatinine and 24-h urinary protein were all significantly higher, while hemoglobin and triglyceride levels were lower in DN group than in NDRD group (P<0.05). The proportion of NDRD cases increased gradually over time, with IgA nephropathy and membranous nephropathy as the main pathological types. The AUC of JDB model was 0.946, similar to that of NDT model (0.925; P=0.198). The disease course of diabetes, hematuria and incidence of DR were important clinical features affecting the diagnostic accuracy of the models.

CONCLUSION

The clinical features and pathological diagnosis of DR change over time. The non-invasive diagnostic models based on DR have good diagnostic efficacy for DN.

Prediction of 5-year overall survival of tongue cancer based machine learning

OBJECTIVE

We aimed to develop a 5-year overall survival prediction model for patients with oral tongue squamous cell carcinoma based on machine learning methods.

SUBJECTS AND METHODS

The data were obtained from electronic medical records of 224 OTSCC patients at the PLA General Hospital. A five-year overall survival prediction model was constructed using logistic regression, Support Vector Machines, Decision Tree, Random Forest, Extreme Gradient Boosting, and Light Gradient Boosting Machine. Model performance was evaluated according to the area under the curve (AUC) of the receiver operating characteristic curve. The output of the optimal model was explained using the Python package (SHapley Additive exPlanations, SHAP).

RESULTS

After passing through the grid search and secondary modeling, the Light Gradient Boosting Machine was the best prediction model (AUC = 0.860). As explained by SHapley Additive exPlanations, N-stage, age, systemic inflammation response index, positive lymph nodes, plasma fibrinogen, lymphocyte-to-monocyte ratio, neutrophil percentage, and T-stage could perform a 5-year overall survival prediction for OTSCC. The 5-year survival rate was 42%.

CONCLUSION

The Light Gradient Boosting Machine prediction model predicted 5-year overall survival in OTSCC patients, and this predictive tool has potential prognostic implications for patients with OTSCC.

Prediction of Diabetic Kidney Disease in Newly Diagnosed Type 2 Diabetes Mellitus

Background

Diabetic kidney disease (DKD), a common microvascular complication of diabetes mellitus (DM), is always asymptomatic until it develops to the advanced stage. Thus, we aim to develop a nomogram prediction model for progression to DKD in newly diagnosed type 2 diabetes mellitus (T2DM).

Methods

This was a single-center analysis of prospective data collected from 521 newly diagnosed patients with T2DM. All related clinical records were incorporated, including the triglyceride-glucose index (TyG index). The least absolute shrinkage and selection operator (LASSO) was used to build a prediction model. In addition, discrimination, calibration, and clinical practicality of the nomogram were evaluated.

Results

In this study, 156 participants were incorporated as the validation set, while the remaining 365 were incorporated into the training set. The predictive factors included in the individualized nomogram prediction model included 5 variables. The area under the curve (AUC) for the prediction model was 0.826 (95% CI 0.775 to 0.876), indicating excellent discrimination performance. The model performed exceptionally well in terms of predictive accuracy and clinical applicability, according to calibration curves and decision curve analysis.

Conclusion

The predictive nomogram for the risk of DKD in newly diagnosed T2DM patients had outstanding discrimination and calibration, which could help in clinical practice.

Effects of heavy metal exposure on hypertension: A machine learning modeling approach

Heavy metal exposure is a common risk factor for hypertension. To develop an interpretable predictive machine learning (ML) model for hypertension based on levels of heavy metal exposure, data from the NHANES (2003-2016) were employed. Random forest (RF), support vector machine (SVM), decision tree (DT), multilayer perceptron (MLP), ridge regression (RR), AdaBoost (AB), gradient boosting decision tree (GBDT), voting classifier (VC), and K-nearest neighbour (KNN) algorithms were utilized to generate an optimal predictive model for hypertension. Three interpretable methods, the permutation feature importance analysis, partial dependence plot (PDP), and Shapley additive explanations (SHAP) methods, were integrated into a pipeline and embedded in ML for model interpretation. A total of 9005 eligible individuals were randomly allocated into two distinct sets for predictive model training and validation. The results showed that among the predictive models, the RF model demonstrated the highest performance, achieving an accuracy rate of 77.40% in the validation set. The AUC and F1 score for the model were 0.84 and 0.76, respectively. Blood Pb, urinary Cd, urinary Tl, and urinary Co levels were identified as the main influencers of hypertension, and their contribution weights were 0.0504 ± 0.0482, 0.0389 ± 0.0256, 0.0307 ± 0.0179, and 0.0296 ± 0.0162, respectively. Blood Pb (0.55-2.93 μg/dL) and urinary Cd (0.06-0.15 μg/L) levels exhibited the most pronounced upwards trend with the risk of hypertension within a specific value range, while urinary Tl (0.06-0.26 μg/L) and urinary Co (0.02-0.32 μg/L) levels demonstrated a declining trend with hypertension. The findings on the synergistic effects indicated that Pb and Cd were the primary determinants of hypertension. Our findings underscore the predictive value of heavy metals for hypertension. By utilizing interpretable methods, we discerned that Pb, Cd, Tl, and Co emerged as noteworthy contributors within the predictive model.

Prediction for chronic kidney disease by categorical and non_categorical attributes using different machine learning algorithms

Chronic kidney disease (CKD) is a common disease as it is difficult to diagnose early due to its lack of symptoms. The main goal is to first diagnose kidney failure, which is a requirement for dialysis or a kidney transplant. This model teaches patients how to live a healthy life, helps doctors identify the risk and severity of disease, and how plan future treatments. Machine learning algorithms are often used in health care to predict and manage the disease. The purpose of this study is to develop a model for the early detection of CKD, which has three parts: (a) applying baseline classifiers on categorical attributes, (b) applying baseline classifiers on non_categorical attributes, (c) applying baseline classifiers on both categorical and non_categorical attributes, and (d) improving the results of the proposed model by combing the results of above three classifiers based on a majority vote. The proposed model based on baseline classifiers and the majority voting method shows a 3% increase in accuracy over the other existing models. The results provide support for increased accuracy in the current classification of chronic kidney disease.

Machine Learning Models to Predict the Risk of Rapidly Progressive Kidney Disease and the Need for Nephrology Referral in Adult Patients with Type 2 Diabetes

Early detection of rapidly progressive kidney disease is key to improving the renal outcome and reducing complications in adult patients with type 2 diabetes mellitus (T2DM). We aimed to construct a 6-month machine learning (ML) predictive model for the risk of rapidly progressive kidney disease and the need for nephrology referral in adult patients with T2DM and an initial estimated glomerular filtration rate (eGFR) ≥ 60 mL/min/1.73 m2. We extracted patients and medical features from the electronic medical records (EMR), and the cohort was divided into a training/validation and testing data set to develop and validate the models on the basis of three algorithms: logistic regression (LR), random forest (RF), and extreme gradient boosting (XGBoost). We also applied an ensemble approach using soft voting classifier to classify the referral group. We used the area under the receiver operating characteristic curve (AUROC), precision, recall, and accuracy as the metrics to evaluate the performance. Shapley additive explanations (SHAP) values were used to evaluate the feature importance. The XGB model had higher accuracy and relatively higher precision in the referral group as compared with the LR and RF models, but LR and RF models had higher recall in the referral group. In general, the ensemble voting classifier had relatively higher accuracy, higher AUROC, and higher recall in the referral group as compared with the other three models. In addition, we found a more specific definition of the target improved the model performance in our study. In conclusion, we built a 6-month ML predictive model for the risk of rapidly progressive kidney disease. Early detection and then nephrology referral may facilitate appropriate management.

Development and validation of a machine learning-based predictive model for secondary post-tonsillectomy hemorrhage

Background

The main obstacle to a patient's recovery following a tonsillectomy is complications, and bleeding is the most frequent culprit. Predicting post-tonsillectomy hemorrhage (PTH) allows for accurate identification of high-risk populations and the implementation of protective measures. Our study aimed to investigate how well machine learning models predict the risk of PTH.

Methods

Data were obtained from 520 patients who underwent a tonsillectomy at The 940th Hospital of Joint Logistics Support Force of Chinese People's Liberation Army. The age range of the patients was 2-57 years, and 364 (70%) were male. The prediction models were developed using five machine learning models: decision tree, support vector machine (SVM), extreme gradient boosting (XGBoost), random forest, and logistic regression. The performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC). Shapley additive explanation (SHAP) was used to interpret the results of the best-performing model.

Results

The frequency of PTH was 11.54% among the 520 patients, with 10.71% in the training group and 13.46% in the validation set. Age, BMI, season, smoking, blood type, INR, combined secretory otitis media, combined adenoidectomy, surgical wound, and use of glucocorticoids were selected by mutual information (MI) method. The XGBoost model had best AUC (0.812) and Brier score (0.152). Decision curve analysis (DCA) showed that the model had a high clinical utility. The SHAP method revealed the top 10 variables of MI according to the importance ranking, and the average of the age was recognized as the most important predictor variable.

Conclusion

This study built a PTH risk prediction model using machine learning. The XGBoost model is a tool with potential to facilitate population management strategies for PTH.

Machine learning models for prediction of HF and CKD development in early-stage type 2 diabetes patients

Chronic kidney disease (CKD) and heart failure (HF) are the first and most frequent comorbidities associated with mortality risks in early-stage type 2 diabetes mellitus (T2DM). However, efficient screening and risk assessment strategies for identifying T2DM patients at high risk of developing CKD and/or HF (CKD/HF) remains to be established. This study aimed to generate a novel machine learning (ML) model to predict the risk of developing CKD/HF in early-stage T2DM patients. The models were derived from a retrospective cohort of 217,054 T2DM patients without a history of cardiovascular and renal diseases extracted from a Japanese claims database. Among algorithms used for the ML, extreme gradient boosting exhibited the best performance for CKD/HF diagnosis and hospitalization after internal validation and was further validated using another dataset including 16,822 patients. In the external validation, 5-years prediction area under the receiver operating characteristic curves for CKD/HF diagnosis and hospitalization were 0.718 and 0.837, respectively. In Kaplan-Meier curves analysis, patients predicted to be at high risk showed significant increase in CKD/HF diagnosis and hospitalization compared with those at low risk. Thus, the developed model predicted the risk of developing CKD/HF in T2DM patients with reasonable probability in the external validation cohort. Clinical approach identifying T2DM at high risk of developing CKD/HF using ML models may contribute to improved prognosis by promoting early diagnosis and intervention.

Predictive models for small-for-gestational-age births in women exposed to pesticides before pregnancy based on multiple machine learning algorithms

Background

The association between prenatal pesticide exposures and a higher incidence of small-for-gestational-age (SGA) births has been reported. No prediction model has been developed for SGA neonates in pregnant women exposed to pesticides prior to pregnancy.

Methods

A retrospective cohort study was conducted using information from the National Free Preconception Health Examination Project between 2010 and 2012. A development set (n = 606) and a validation set (n = 151) of the dataset were split at random. Traditional logistic regression (LR) method and six machine learning classifiers were used to develop prediction models for SGA neonates. The Shapley Additive Explanation (SHAP) model was applied to determine the most influential variables that contributed to the outcome of the prediction.

Results

757 neonates in total were analyzed. SGA occurred in 12.9% (n = 98) of cases overall. With an area under the receiver-operating-characteristic curve (AUC) of 0.855 [95% confidence interval (CI): 0.752–0.959], the model based on category boosting (CatBoost) algorithm obtained the best performance in the validation set. With the exception of the LR model (AUC: 0.691, 95% CI: 0.554–0.828), all models had good AUCs. Using recursive feature elimination (RFE) approach to perform the feature selection, we included 15 variables in the final model based on CatBoost classifier, achieving the AUC of 0.811 (95% CI: 0.675–0.947).

Conclusions

Machine learning algorithms can develop satisfactory tools for SGA prediction in mothers exposed to pesticides prior to pregnancy, which might become a tool to predict SGA neonates in the high-risk population.