Machine learning and deep learning predictive models for type 2 diabetes: a systematic review

Predicting new cases of hypertension in Swedish primary care with a machine learning tool

Background

Many individuals with hypertension remain undiagnosed. We aimed to develop a predictive model for hypertension using diagnostic codes from prevailing electronic medical records in Swedish primary care.

Methods

This sex- and age-matched case-control (1:5) study included patients aged 30-65 years living in the Stockholm Region, Sweden, with a newly recorded diagnosis of hypertension during 2010-19 (cases) and individuals without a recorded hypertension diagnosis during 2010-19 (controls), in total 507,618 individuals. Patients with diagnoses of cardiovascular diseases or diabetes were excluded. A stochastic gradient boosting machine learning model was constructed using the 1,309 most registered ICD-10 codes from primary care for three years prior the hypertension diagnosis.

Results

The model showed an area under the curve (95 % confidence interval) of 0.748 (0.742-0.753) for females and 0.745 (0.740-0.751) for males for predicting diagnosis of hypertension within three years. The sensitivity was 63 % and 68 %, and the specificity 76 % and 73 %, for females and males, respectively. The 25 diagnoses that contributed the most to the model for females and males all exhibited a normalized relative influence >1 %. The codes contributing most to the model, all with an odds ratio of marginal effects >1 for both sexes, were dyslipidaemia, obesity, and encountering health services in other circumstances.

Conclusions

This machine learning model, using prevailing recorded diagnoses within primary health care, may contribute to the identification of patients at risk of unrecognized hypertension. The added value of this predictive model beyond information of blood pressure warrants further study.

A machine learning tool for identifying patients with newly diagnosed diabetes in primary care

BACKGROUND AND AIM

It is crucial to identify a diabetes diagnosis early. Create a predictive model utilizing machine learning (ML) to identify new cases of diabetes in primary health care (PHC).

METHODS

A case-control study utilizing data on PHC visits for sex-, age, and PHC-matched controls. Stochastic gradient boosting was used to construct a model for predicting cases of diabetes based on diagnostic codes from PHC consultations during the year before index (diagnosis) date and number of consultations. Variable importance was estimated using the normalized relative influence (NRI) score. Risks of having diabetes were calculated using odds ratios of marginal effects (ORME). Four groups by age and sex were studied, age-groups 35-64 years and ≥ 65 years in men and women, respectively.

RESULTS

The most important predictive factors were hypertension with NRI 21.4-29.7 %, and obesity 4.8-15.2 %. The NRI for other top ten diagnoses and administrative codes generally ranged 1.0-4.2 %.

CONCLUSIONS

Our data confirm the known risk patterns for predicting a new diagnosis of diabetes, and the need to test blood glucose frequently. To assess the full potential of ML for risk prediction purposes in clinical practice, future studies could include clinical data on life-style patterns, laboratory tests and prescribed medication.

Exploratory risk prediction of type II diabetes with isolation forests and novel biomarkers

Type II diabetes mellitus (T2DM) is a rising global health burden due to its rapidly increasing prevalence worldwide, and can result in serious complications. Therefore, it is of utmost importance to identify individuals at risk as early as possible to avoid long-term T2DM complications. In this study, we developed an interpretable machine learning model leveraging baseline levels of biomarkers of oxidative stress (OS), inflammation, and mitochondrial dysfunction (MD) for identifying individuals at risk of developing T2DM. In particular, Isolation Forest (iForest) was applied as an anomaly detection algorithm to address class imbalance. iForest was trained on the control group data to detect cases of high risk for T2DM development as outliers. Two iForest models were trained and evaluated through ten-fold cross-validation, the first on traditional biomarkers (BMI, blood glucose levels (BGL) and triglycerides) alone and the second including the additional aforementioned biomarkers. The second model outperformed the first across all evaluation metrics, particularly for F1 score and recall, which were increased from 0.61 ± 0.05 to 0.81 ± 0.05 and 0.57 ± 0.06 to 0.81 ± 0.08, respectively. The feature importance scores identified a novel combination of biomarkers, including interleukin-10 (IL-10), 8-isoprostane, humanin (HN), and oxidized glutathione (GSSG), which were revealed to be more influential than the traditional biomarkers in the outcome prediction. These results reveal a promising method for simultaneously predicting and understanding the risk of T2DM development and suggest possible pharmacological intervention to address inflammation and OS early in disease progression.

Comparing the accuracy of four machine learning models in predicting type 2 diabetes onset within the Chinese population: a retrospective study

OBJECTIVE

To evaluate the effectiveness of machine learning (ML) models in predicting 5-year type 2 diabetes mellitus (T2DM) risk within the Chinese population by retrospectively analyzing annual health checkup records.

METHODS

We included 46,247 patients (32,372 and 13,875 in training and validation sets, respectively) from a national health checkup center database. Univariate and multivariate Cox analyses were performed to identify factors influencing T2DM risk. Extreme Gradient Boosting (XGBoost), support vector machine (SVM), logistic regression (LR), and random forest (RF) models were trained to predict 5-year T2DM risk. Model performances were analyzed using receiver operating characteristic (ROC) curves for discrimination and calibration plots for prediction accuracy.

RESULTS

Key variables included fasting plasma glucose, age, and sedentary time. The LR model showed good accuracy with respective areas under the ROC (AUCs) of 0.914 and 0.913 in training and validation sets; the RF model exhibited favorable AUCs of 0.998 and 0.838. In calibration analysis, the LR model displayed good fit for low-risk patients; the RF model exhibited satisfactory fit for low- and high-risk patients.

CONCLUSIONS

LR and RF models can effectively predict T2DM risk in the Chinese population. These models may help identify high-risk patients and guide interventions to prevent complications and disabilities.

Comparative Analysis of Biomarkers in Type 2 Diabetes Patients With and Without Comorbidities: Insights Into the Role of Hypertension and Cardiovascular Disease

Background

Type 2 diabetes mellitus (T2DM) are 90% of diabetes cases, and its prevalence and incidence, including comorbidities, are rising worldwide. Clinically, diabetes and associated comorbidities are identified by biochemical and physical characteristics including glycemia, glycated hemoglobin (HbA1c), and tests for cardiovascular, eye and kidney disease.

Objectives

Diabetes may have a common etiology based on inflammation and oxidative stress that may provide additional information about disease progression and treatment options. Thus, identifying high-risk individuals can delay or prevent diabetes and its complications.

Design

In patients with or without hypertension and cardiovascular disease, as part of progression from no diabetes to T2DM, this research studied the changes in biomarkers between control and prediabetes, prediabetes to T2DM, and control to T2DM, and classified patients based on first-attendance data. Control patients and patients with hypertension, cardiovascular, and with both hypertension and cardiovascular diseases are 156, 148, 61, and 216, respectively.

Methods

Linear discriminant analysis is used for classification method and feature importance, This study examined the relationship between Humanin and mitochondrial protein (MOTSc), mitochondrial peptides associated with oxidative stress, diabetes progression, and associated complications.

Results

MOTSc, reduced glutathione and glutathione disulfide ratio (GSH/GSSG), interleukin-1β (IL-1β), and 8-isoprostane were significant (P < .05) for the transition from prediabetes to t2dm, highlighting importance of mitochondrial involvement. complement component 5a (c5a) is a biomarker associated with disease progression and comorbidities, gsh gssg, monocyte chemoattractant protein-1 (mcp-1), 8-isoprostane being most important biomarkers.

Conclusions

Comorbidities affect the hypothesized biomarkers as diabetes progresses. Mitochondrial oxidative stress indicators, coagulation, and inflammatory markers help assess diabetes disease development and provide appropriate medications. Future studies will examine longitudinal biomarker evolution.

Personalized Diabetes Management with Digital Twins: A Patient-Centric Knowledge Graph Approach

Diabetes management requires constant monitoring and individualized adjustments. This study proposes a novel approach that leverages digital twins and personal health knowledge graphs (PHKGs) to revolutionize diabetes care. Our key contribution lies in developing a real-time, patient-centric digital twin framework built on PHKGs. This framework integrates data from diverse sources, adhering to HL7 standards and enabling seamless information access and exchange while ensuring high levels of accuracy in data representation and health insights. PHKGs offer a flexible and efficient format that supports various applications. As new knowledge about the patient becomes available, the PHKG can be easily extended to incorporate it, enhancing the precision and accuracy of the care provided. This dynamic approach fosters continuous improvement and facilitates the development of new applications. As a proof of concept, we have demonstrated the versatility of our digital twins by applying it to different use cases in diabetes management. These include predicting glucose levels, optimizing insulin dosage, providing personalized lifestyle recommendations, and visualizing health data. By enabling real-time, patient-specific care, this research paves the way for more precise and personalized healthcare interventions, potentially improving long-term diabetes management outcomes.

Machine-learning algorithms in screening for type 2 diabetes mellitus: Data from Fasa Adults Cohort Study

INTRODUCTION

The application of machine learning (ML) is increasingly growing in biomedical sciences. This study aimed to evaluate factors associated with type 2 diabetes mellitus (T2DM) and compare the performance of ML methods in identifying individuals with the disease in an Iranian setting.

METHODS

Using the baseline data from Fasa Adult Cohort Study (FACS) and in a sex-stratified manner, we studied factors associated with T2DM by applying seven different ML methods including Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbours (KNN), Gradient Boosting Machine (GBM), Extreme Gradient Boosting (XGB) and Bagging classifier (BAG). We further compared the performance of these methods; for each algorithm, accuracy, precision, sensitivity, specificity, F1 score, and Area Under Curve (AUC) were calculated.

RESULTS

10,112 participants were recruited between 2014 and 2016, of whom 1246 had T2DM at baseline. 4566 (45%) participants were males, aged between 35 and 70 years. For males, age, sugar consumption, and history of hospitalization were the most weighted variables regarding their importance in screening for T2DM using the GBM model, respectively; these variables were sugar consumption, urine blood, and age for females. GBM outperformed other models for both males and females with AUC of 0.75 (0.69-0.82) and 0.76 (0.71-0.80), and F1 score of 0.33 (0.27-0.39) and 0.42 (0.38-0.46), respectively. GBM also showed a sensitivity of 0.24 (0.19-0.29) and a specificity of 0.98 (0.96-1.0) in males and a sensitivity of 0.38 (0.34-0.42) and specificity of 0.92 (0.89-0.95) in females. Notably, close performance characteristics were detected among other ML models.

CONCLUSIONS

GBM model might achieve better performance in screening for T2DM in a south Iranian population.

Validation of the Framingham Diabetes Risk Model Using Community-Based KoGES Data

BACKGROUND

An 8-year prediction of the Framingham Diabetes Risk Model (FDRM) was proposed, but the predictor has a gap with current clinical standards. Therefore, we evaluated the validity of the original FDRM in Korean population data, developed a modified FDRM by redefining the predictors based on current knowledge, and evaluated the internal and external validity.

METHODS

Using data from a community-based cohort in Korea (n = 5,409), we calculated the probability of diabetes through FDRM, and developed a modified FDRM based on modified definitions of hypertension (HTN) and diabetes. We also added clinical features related to diabetes to the predictive model. Model performance was evaluated and compared by area under the curve (AUC).

RESULTS

During the 8-year follow-up, the cumulative incidence of diabetes was 8.5%. The modified FDRM consisted of age, obesity, HTN, hypo-high-density lipoprotein cholesterol, elevated triglyceride, fasting glucose, and hemoglobin A1c. The expanded clinical model added γ-glutamyl transpeptidase to the modified FDRM. The FDRM showed an estimated AUC of 0.71, and the model's performance improved to an AUC of 0.82 after applying the redefined predictor. Adding clinical features (AUC = 0.83) to the modified FDRM further improved in discrimination, but this was not maintained in the validation data set. External validation was evaluated on population-based cohort data and both modified models performed well, with AUC above 0.82.

CONCLUSION

The performance of FDRM in the Korean population was found to be acceptable for predicting diabetes, but it was improved when corrected with redefined predictors. The validity of the modified model needs to be further evaluated.

Identifying top ten predictors of type 2 diabetes through machine learning analysis of UK Biobank data

The study aimed to identify the most predictive factors for the development of type 2 diabetes. Using an XGboost classification model, we projected type 2 diabetes incidence over a 10-year horizon. We deliberately minimized the selection of baseline factors to fully exploit the rich dataset from the UK Biobank. The predictive value of features was assessed using shap values, with model performance evaluated via Receiver Operating Characteristic Area Under the Curve, sensitivity, and specificity. Data from the UK Biobank, encompassing a vast population with comprehensive demographic and health data, was employed. The study enrolled 450,000 participants aged 40-69, excluding those with pre-existing diabetes. Among 448,277 participants, 12,148 developed type 2 diabetes within a decade. HbA1c emerged as the foremost predictor, followed by BMI, waist circumference, blood glucose, family history of diabetes, gamma-glutamyl transferase, waist-hip ratio, HDL cholesterol, age, and urate. Our XGboost model achieved a Receiver Operating Characteristic Area Under the Curve of 0.9 for 10-year type 2 diabetes prediction, with a reduced 10-feature model achieving 0.88. Easily measurable biological factors surpassed traditional risk factors like diet, physical activity, and socioeconomic status in predicting type 2 diabetes. Furthermore, high prediction accuracy could be maintained using just the top 10 biological factors, with additional ones offering marginal improvements. These findings underscore the significance of biological markers in type 2 diabetes prediction.

The Reporting Quality of Machine Learning Studies on Pediatric Diabetes Mellitus: Systematic Review

BACKGROUND

Diabetes mellitus (DM) is a major health concern among children with the widespread adoption of advanced technologies. However, concerns are growing about the transparency, replicability, biasedness, and overall validity of artificial intelligence studies in medicine.

OBJECTIVE

We aimed to systematically review the reporting quality of machine learning (ML) studies of pediatric DM using the Minimum Information About Clinical Artificial Intelligence Modelling (MI-CLAIM) checklist, a general reporting guideline for medical artificial intelligence studies.

METHODS

We searched the PubMed and Web of Science databases from 2016 to 2020. Studies were included if the use of ML was reported in children with DM aged 2 to 18 years, including studies on complications, screening studies, and in silico samples. In studies following the ML workflow of training, validation, and testing of results, reporting quality was assessed via MI-CLAIM by consensus judgments of independent reviewer pairs. Positive answers to the 17 binary items regarding sufficient reporting were qualitatively summarized and counted as a proxy measure of reporting quality. The synthesis of results included testing the association of reporting quality with publication and data type, participants (human or in silico), research goals, level of code sharing, and the scientific field of publication (medical or engineering), as well as with expert judgments of clinical impact and reproducibility.

RESULTS

After screening 1043 records, 28 studies were included. The sample size of the training cohort ranged from 5 to 561. Six studies featured only in silico patients. The reporting quality was low, with great variation among the 21 studies assessed using MI-CLAIM. The number of items with sufficient reporting ranged from 4 to 12 (mean 7.43, SD 2.62). The items on research questions and data characterization were reported adequately most often, whereas items on patient characteristics and model examination were reported adequately least often. The representativeness of the training and test cohorts to real-world settings and the adequacy of model performance evaluation were the most difficult to judge. Reporting quality improved over time (r=0.50; P=.02); it was higher than average in prognostic biomarker and risk factor studies (P=.04) and lower in noninvasive hypoglycemia detection studies (P=.006), higher in studies published in medical versus engineering journals (P=.004), and higher in studies sharing any code of the ML pipeline versus not sharing (P=.003). The association between expert judgments and MI-CLAIM ratings was not significant.

CONCLUSIONS

The reporting quality of ML studies in the pediatric population with DM was generally low. Important details for clinicians, such as patient characteristics; comparison with the state-of-the-art solution; and model examination for valid, unbiased, and robust results, were often the weak points of reporting. To assess their clinical utility, the reporting standards of ML studies must evolve, and algorithms for this challenging population must become more transparent and replicable.

Systematic reviews of machine learning in healthcare: a literature review

INTRODUCTION

The increasing availability of data and computing power has made machine learning (ML) a viable approach to faster, more efficient healthcare delivery.

METHODS

A systematic literature review (SLR) of published SLRs evaluating ML applications in healthcare settings published between1 January 2010 and 27 March 2023 was conducted.

RESULTS

In total 220 SLRs covering 10,462 ML algorithms were reviewed. The main application of AI in medicine related to the clinical prediction and disease prognosis in oncology and neurology with the use of imaging data. Accuracy, specificity, and sensitivity were provided in 56%, 28%, and 25% SLRs respectively. Internal and external validation was reported in 53% and less than 1% of the cases respectively. The most common modeling approach was neural networks (2,454 ML algorithms), followed by support vector machine and random forest/decision trees (1,578 and 1,522 ML algorithms, respectively).

EXPERT OPINION

The review indicated considerable reporting gaps in terms of the ML's performance, both internal and external validation. Greater accessibility to healthcare data for developers can ensure the faster adoption of ML algorithms into clinical practice.

Predictive model for early detection of type 2 diabetes using patients' clinical symptoms, demographic features, and knowledge of diabetes

Background and Aims

With the global rise in type 2 diabetes, predictive modeling has become crucial for early detection, particularly in populations with low routine medical checkup profiles. This study aimed to develop a predictive model for type 2 diabetes using health check-up data focusing on clinical details, demographic features, biochemical markers, and diabetes knowledge.

Methods

Data from 444 Nigerian patients were collected and analysed. We used 80% of this data set for training, and the remaining 20% for testing. Multivariable penalized logistic regression was employed to predict the disease onset, incorporating waist-hip ratio (WHR), triglycerides (TG), catalase, and atherogenic indices of plasma (AIP).

Results

The predictive model demonstrated high accuracy, with an area under the curve of 99% (95% CI = 97%-100%) for the training set and 94% (95% CI = 89%-99%) for the test set. Notably, an increase in WHR (adjusted odds ratio [AOR] = 70.35; 95% CI = 10.04-493.1, p-value < 0.001) and elevated AIP (AOR = 4.55; 95% CI = 1.48-13.95, p-value = 0.008) levels were significantly associated with a higher risk of type 2 diabetes, while higher catalase levels (AOR = 0.33; 95% CI = 0.22-0.49, p < 0.001) correlated with a decreased risk. In contrast, TG levels (AOR = 1.04; 95% CI = 0.40-2.71, p-value = 0.94) were not associated with the disease.

Conclusion

This study emphasizes the importance of using distinct clinical and biochemical markers for early type 2 diabetes detection in Nigeria, reflecting global trends in diabetes modeling, and highlighting the need for context-specific methods. The development of a web application based on these results aims to facilitate the early identification of individuals at risk, potentially reducing health complications, and improving diabetes management strategies in diverse settings.

Machine learning for predicting hepatitis B or C virus infection in diabetic patients

Highly prevalent hepatitis B and hepatitis C virus (HBV and HCV) infections have been reported among individuals with diabetes. Given the frequently asymptomatic nature of hepatitis and the challenges associated with screening in some vulnerable populations such as diabetes patients, we conducted an investigation into the performance of various machine learning models for the identification of hepatitis in diabetic patients while also evaluating the significance of features. Analyzing NHANES data from 2013 to 2018, machine learning models were evaluated; random forest (RF), support vector machine (SVM), eXtreme Gradient Boosting (XGBoost), and least absolute shrinkage and selection operator (LASSO) along with stacked ensemble model. We performed hyperparameter tuning to improve the performance of the model, and selected important predictors using the best performance model. LASSO showed the highest predictive performance (AUC-ROC = 0.810) rather than other models. Illicit drug use, poverty, and race were highly ranked as predictive factors for developing hepatitis in diabetes patients. Our study demonstrated that a machine-learning-based model performed optimally in the detection of hepatitis among diabetes patients, achieving high performance. Furthermore, models and predictors evaluated from the current study, we expect, could be supportive information for developing screening or treatment methods for hepatitis care in diabetes patients.

Establishment of a differential diagnosis method and an online prediction platform for AOSD and sepsis based on gradient boosting decision trees algorithm

OBJECTIVE

The differential diagnosis between adult-onset Still's disease (AOSD) and sepsis has always been a challenge. In this study, a machine learning model for differential diagnosis of AOSD and sepsis was developed and an online platform was developed to facilitate the clinical application of the model.

METHODS

All data were collected from 42 AOSD patients and 50 sepsis patients admitted to Affiliated Hospital of Xuzhou Medical University from December 2018 to December 2021. In addition, 5 AOSD patients and 10 sepsis patients diagnosed in our hospital after March 2022 were collected for external validation. All models were built using the scikit-learn library (version 1.0.2) in Python (version 3.9.7), and feature selection was performed using the SHAP (Shapley Additive exPlanation) package developed in Python.

RESULTS

The results showed that the gradient boosting decision tree(GBDT) optimization model based on arthralgia, ferritin × lymphocyte count, white blood cell count, ferritin × platelet count, and α1-acid glycoprotein/creatine kinase could well identify AOSD and sepsis. The training set interaction test (AUC: 0.9916, ACC: 0.9457, Sens: 0.9556, Spec: 0.9578) and the external validation also achieved satisfactory results (AUC: 0.9800, ACC: 0.9333, Sens: 0.8000, Spec: 1.000). We named this discrimination method AIADSS (AI-assisted discrimination of Still's disease and Sepsis) and created an online service platform for practical operation, the website is http://cppdd.cn/STILL1/ .

CONCLUSION

We created a method for the identification of AOSD and sepsis based on machine learning. This method can provide a reference for clinicians to formulate the next diagnosis and treatment plan.

Predicting three-month fasting blood glucose and glycated hemoglobin changes in patients with type 2 diabetes mellitus based on multiple machine learning algorithms

Fasting blood glucose (FBG) and glycosylated hemoglobin (HbA1c) are key indicators reflecting blood glucose control in type 2 diabetes mellitus (T2DM) patients. The purpose of this study is to establish a predictive model for blood glucose changes in T2DM patients after 3 months of treatment, achieving personalized treatment.A retrospective study was conducted on type 2 diabetes mellitus real-world medical data from 4 cities in Sichuan Province, China from January 2015 to December 2020. After data preprocessing, data inputting, data sampling, and feature screening, 16 kinds of machine learning methods were used to construct prediction models, and 5 prediction models with the best prediction performance were screened respectively. A total of 100,000 cases were included to establish the FBG model, and 2,169 cases were established to establish the HbA1c model. The best prediction model both of FBG and HbA1c finally obtained are realized by ensemble learning and modified random forest inputting, the AUC values are 0.819 and 0.970, respectively. The most important indicators of the FBG and HbA1c prediction model were FBG and HbA1c. Medication compliance, follow-up outcome, dietary habits, BMI, and waist circumference also had a greater impact on FBG levels. The prediction accuracy of the models of the two blood glucose control indicators is high and has certain clinical applicability.HbA1c and FBG are mutually important predictors, and there is a close relationship between them.

Machine Learning as a Support for the Diagnosis of Type 2 Diabetes

Diabetes is a chronic, metabolic disease characterized by high blood sugar levels. Among the main types of diabetes, type 2 is the most common. Early diagnosis and treatment can prevent or delay the onset of complications. Previous studies examined the application of machine learning techniques for prediction of the pathology, and here an artificial neural network shows very promising results as a possible valuable aid in the management and prevention of diabetes. Additionally, its superior ability for long-term predictions makes it an ideal choice for this field of study. We utilized machine learning methods to uncover previously undiscovered associations between an individual’s health status and the development of type 2 diabetes, with the goal of accurately predicting its onset or determining the individual’s risk level. Our study employed a binary classifier, trained on scratch, to identify potential nonlinear relationships between the onset of type 2 diabetes and a set of parameters obtained from patient measurements. Three datasets were utilized, i.e., the National Center for Health Statistics’ (NHANES) biennial survey, MIMIC-III and MIMIC-IV. These datasets were then combined to create a single dataset with the same number of individuals with and without type 2 diabetes. Since the dataset was balanced, the primary evaluation metric for the model was accuracy. The outcomes of this study were encouraging, with the model achieving accuracy levels of up to 86% and a ROC AUC value of 0.934. Further investigation is needed to improve the reliability of the model by considering multiple measurements from the same patient over time.

Predicting type 2 diabetes prevalence for people with severe mental illness in a multi-ethnic East London population

BACKGROUND AND AIMS

Prevalence of type two diabetes mellitus (T2DM) in people with severe mental illness (SMI) is 2-3 times higher than in general population. Predictive modelling has advanced greatly in the past decade, and it is important to apply cutting-edge methods to vulnerable groups. However, few T2DM prediction models account for the presence of mental illness, and none seemed to have been developed specifically for people with SMI. Therefore, we aimed to develop and internally validate a T2DM prevalence model for people with SMI.

METHODS

We utilised a large cross-sectional sample representative of a multi-ethnic population from London (674,000 adults); 10,159 people with SMI formed our analytical sample (1,513 T2DM cases). We fitted a linear logistic regression and XGBoost as stand-alone models and as a stacked ensemble. Age, sex, body mass index, ethnicity, area-based deprivation, past hypertension, cardiovascular diseases, prescribed antipsychotics, and SMI illness were the predictors.

RESULTS

Logistic regression performed well while detecting T2DM presence for people with SMI: area under the receiver operator curve (ROC-AUC) was 0.83 (95 % CI 0.79-0.87). XGBoost and LR-XGBoost ensemble performed equally well, ROC-AUC 0.83 (95 % CI 0.79-0.87), indicating a negligible contribution of non-linear terms to predictive power. Ethnicity was the most important predictor after age. We demonstrated how the derived models can be utilised and estimated a 2.14 % (95 %CI 2.03 %-2.24 %) increase in T2DM prevalence in East London SMI population in 20 years' time, driven by the projected demographic changes.

CONCLUSIONS

Primary care data, the setting where prediction models could be most fruitfully used, provide enough information for well-performing T2DM prevalence models for people with SMI. We demonstrated how thorough internal cross-validation of an ensemble of a linear and machine-learning model can quantify the predictive value of non-linearity in the data.

Development and validation of an insulin resistance model for a population without diabetes mellitus and its clinical implication: a prospective cohort study

Background

Insulin resistance (IR) is associated with diabetes mellitus, cardiovascular disease (CV), and mortality. Few studies have used machine learning to predict IR in the non-diabetic population.

Methods

In this prospective cohort study, we trained a predictive model for IR in the non-diabetic populations using the US National Health and Nutrition Examination Survey (NHANES, from JAN 01, 1999 to DEC 31, 2012) database and the Taiwan MAJOR (from JAN 01, 2008 to DEC 31, 2017) database. We analysed participants in the NHANES and MAJOR and participants were excluded if they were aged <18 years old, had incomplete laboratory data, or had DM. To investigate the clinical implications (CV and all-cause mortality) of this trained model, we tested it with the Taiwan biobank (TWB) database from DEC 10, 2008 to NOV 30, 2018. We then used SHapley Additive exPlanation (SHAP) values to explain differences across the machine learning models.

Findings

Of all participants (combined NHANES and MJ databases), we randomly selected 14,705 participants for the training group, and 4018 participants for the validation group. In the validation group, their areas under the curve (AUC) were all >0.8 (highest being XGboost, 0.87). In the test group, all AUC were also >0.80 (highest being XGboost, 0.88). Among all 9 features (age, gender, race, body mass index, fasting plasma glucose (FPG), glycohemoglobin, triglyceride, total cholesterol and high-density cholesterol), BMI had the highest value of feature importance on IR (0.43 for XGboost and 0.47 for RF algorithms). All participants from the TWB database were separated into the IR group and the non-IR group according to the XGboost algorithm. The Kaplan-Meier survival curve showed a significant difference between the IR and non-IR groups (p < 0.0001 for CV mortality, and p = 0.0006 for all-cause mortality). Therefore, the XGboost model has clear clinical implications for predicting IR, aside from CV and all-cause mortality.

Interpretation

To predict IR in non-diabetic patients with high accuracy, only 9 easily obtained features are needed for prediction accuracy using our machine learning model. Similarly, the model predicts IR patients with significantly higher CV and all-cause mortality. The model can be applied to both Asian and Caucasian populations in clinical practice.

Funding

Taichung Veterans General Hospital, Taiwan and Japan Society for the Promotion of Science KAKENHI Grant Number JP21KK0293.

Short communication: unique metabolic signature of proliferative retinopathy in the tear fluid of diabetic patients with comorbidities - preliminary data for PPPM validation

Type 2 diabetes (T2DM) defined as the adult-onset type that is primarily not insulin-dependent, comprises over 95% of all diabetes mellitus (DM) cases. According to global records, 537 million adults aged 20-79 years are affected by DM that means at least 1 out of 15 persons. This number is projected to grow by 51% by the year 2045. One of the most common complications of T2DM is diabetic retinopathy (DR) with an overall prevalence over 30%. The total number of the DR-related visual impairments is on the rise, due to the growing T2DM population. Proliferative diabetic retinopathy (PDR) is the progressing DR and leading cause of preventable blindness in working-age adults. Moreover, PDR with characteristic systemic attributes including mitochondrial impairment, increased cell death and chronic inflammation, is an independent predictor of the cascading DM-complications such as ischemic stroke. Therefore, early DR is a reliable predictor appearing upstream of this "domino effect". Global screening, leading to timely identification of DM-related complications, is insufficiently implemented by currently applied reactive medicine. A personalised predictive approach and cost-effective targeted prevention shortly - predictive, preventive and personalised medicine (PPPM / 3PM) could make a good use of the accumulated knowledge, preventing blindness and other severe DM complications. In order to reach this goal, reliable stage- and disease-specific biomarker panels are needed characterised by an easy way of the sample collection, high sensitivity and specificity of analyses. In the current study, we tested the hypothesis that non-invasively collected tear fluid is a robust source for the analysis of ocular and systemic (DM-related complications) biomarker patterns suitable for differential diagnosis of stable DR versus PDR. Here, we report the first results of the comprehensive ongoing study, in which we correlate individualised patient profiles (healthy controls versus patients with stable D as well as patients with PDR with and without co-morbidities) with their metabolic profiles in the tear fluid. Comparative mass spectrometric analysis performed has identified following metabolic clusters which are differentially expressed in the groups of comparison: acylcarnitines, amino acid & related compounds, bile acids, ceramides, lysophosphatidyl-choline, nucleobases & related compounds, phosphatidyl-cholines, triglycerides, cholesterol esters, and fatty acids. Our preliminary data strongly support potential clinical utility of metabolic patterns in the tear fluid indicating a unique metabolic signature characteristic for the DR stages and PDR progression. This pilot study creates a platform for validating the tear fluid biomarker patterns to stratify T2DM-patients predisposed to the PDR. Moreover, since PDR is an independent predictor of severe T2DM-related complications such as ischemic stroke, our international project aims to create an analytical prototype for the "diagnostic tree" (yes/no) applicable to healthrisk assessment in diabetes care.

How do nurses work in chronic management in the age of artificial intelligence? development and future prospects

AI is undeniably revolutionizing medical research and patient care across diverse fields. Chronic disease nursing care, a pivotal aspect of clinical management, has significantly reaped the benefits of AI across numerous dimensions. Understanding the operational principles of artificial intelligence before implementation is crucial, avoiding indiscriminate replacement of all tasks with AI. Nurses serve as the primary force in symptom group research, expanding beyond diabetes to encompass various chronic diseases; their primary responsibility involves recording patients' daily symptoms and vital signs. However, a substantial portion of current AI research excludes nurses from the developmental phase, encompassing them solely in user and feedback populations. The comprehensive design of the symptom analysis and long-term management approach necessitates the guidance and oversight of nurses; however, their current insufficient involvement might stem from nursing staff's comparatively limited comprehension of AI and their ambiguous perception of their role's value in AI. Therefore, an imperative exploration of nurses' roles in symptom analysis and long-term management, leveraging the latest research in these areas, is vital to pinpoint breakthroughs in nurses' AI involvement in the future.

Review: Machine learning in precision pharmacotherapy of type 2 diabetes-A promising future or a glimpse of hope?

Precision pharmacotherapy of diabetes requires judicious selection of the optimal therapeutic agent for individual patients. Artificial intelligence (AI), a swiftly expanding discipline, holds substantial potential to transform current practices in diabetes diagnosis and management. This manuscript provides a comprehensive review of contemporary research investigating drug responses in patient subgroups, stratified via either supervised or unsupervised machine learning approaches. The prevalent algorithmic workflow for investigating drug responses using machine learning involves cohort selection, data processing, predictor selection, development and validation of machine learning methods, subgroup allocation, and subsequent analysis of drug response. Despite the promising feature, current research does not yet provide sufficient evidence to implement machine learning algorithms into routine clinical practice, due to a lack of simplicity, validation, or demonstrated efficacy. Nevertheless, we anticipate that the evolving evidence base will increasingly substantiate the role of machine learning in molding precision pharmacotherapy for diabetes.

Diabetes Mellitus Disease Prediction and Type Classification Involving Predictive Modeling Using Machine Learning Techniques and Classifiers

The Diabetes-Mellitus (DM) disease is considered a persistent ailment that is triggered by excessive sugar levels in the blood of a person. It gives rise to severe health complications when left untreated and can also give rise to related diseases such as cardiac attack, nervous damage, foot problems, liver and kidney damage, and eye problems. These problems are caused by a series of factors interrelated to one another such as age, gender, family history, BMI, and Blood Glucose. Various Machine-Learning (ML) algorithms are being used in order to predict and detect the disease to avoid further complications of health. The Diabetes prediction process can be further improvised by identifying the type a person is being affected by and the probability of the occurrence of the related diseases. In order to perform the mentioned task, two types of the dataset are used in the study, namely, PIMA and a clinical survey dataset. Various ML algorithms such as Random Forest, Light Gradient Boosting Machine, Gradient Boosting Machine, Support Vector Machine, Decision Tree, and XGBoost are being used. The performance metrics used are accuracy, precision, recall, specificity, and sensitivity. Techniques such as Data Augmentation and Sampling are used. In comparison with the research conducted previously, the paper focuses on improvisation of the accuracy with a percentage of 95.20 using the LGBM Classifier, and Diabetes is also classified as Prediabetes or Diabetes using many Classification mechanisms.

Recent applications of machine learning and deep learning models in the prediction, diagnosis, and management of diabetes: a comprehensive review

Diabetes as a metabolic illness can be characterized by increased amounts of blood glucose. This abnormal increase can lead to critical detriment to the other organs such as the kidneys, eyes, heart, nerves, and blood vessels. Therefore, its prediction, prognosis, and management are essential to prevent harmful effects and also recommend more useful treatments. For these goals, machine learning algorithms have found considerable attention and have been developed successfully. This review surveys the recently proposed machine learning (ML) and deep learning (DL) models for the objectives mentioned earlier. The reported results disclose that the ML and DL algorithms are promising approaches for controlling blood glucose and diabetes. However, they should be improved and employed in large datasets to affirm their applicability.

Comparison of a chronic kidney disease predictive model for type 2 diabetes mellitus in Malaysia using Cox regression versus machine learning approach

Background

Diabetes is one of the leading causes of chronic kidney disease (CKD) and end-stage renal disease. This study aims to develop and validate different risk predictive models for incident CKD and CKD progression in people with type 2 diabetes (T2D).

Methods

We reviewed a cohort of people with T2D seeking care from two tertiary hospitals in the metropolitan cities of the state of Selangor and Negeri Sembilan from January 2012 to May 2021. To identify the 3-year predictor of developing CKD (primary outcome) and CKD progression (secondary outcome), the dataset was randomly split into a training and test set. A Cox proportional hazards (CoxPH) model was developed to identify predictors of developing CKD. The resultant CoxPH model was compared with other machine learning models on their performance using C-statistic.

Results

The cohorts included 1992 participants, of which 295 had developed CKD and 442 reported worsening of kidney function. Equation for the 3-year risk of developing CKD included gender, haemoglobin A1c, triglyceride and serum creatinine levels, estimated glomerular filtration rate, history of cardiovascular disease and diabetes duration. For risk of CKD progression, the model included systolic blood pressure, retinopathy and proteinuria. The CoxPH model was better at prediction compared with other machine learning models examined for incident CKD (C-statistic: training 0.826; test 0.874) and CKD progression (C-statistic: training 0.611; test 0.655). The risk calculator can be found at https://rs59.shinyapps.io/071221/.

Conclusions

The Cox regression model was the best performing model to predict people with T2D who will develop a 3-year risk of incident CKD and CKD progression in a Malaysian cohort.

A novel method to derive personalized minimum viable recommendations for type 2 diabetes prevention based on counterfactual explanations

Despite the growing availability of artificial intelligence models for predicting type 2 diabetes, there is still a lack of personalized approaches to quantify minimum viable changes in biomarkers that may help reduce the individual risk of developing disease. The aim of this article is to develop a new method, based on counterfactual explanations, to generate personalized recommendations to reduce the one-year risk of type 2 diabetes. Ten routinely collected biomarkers extracted from Electronic Medical Records of 2791 patients at low risk and 2791 patients at high risk of type 2 diabetes were analyzed. Two regions characterizing the two classes of patients were estimated using a Support Vector Data Description classifier. Counterfactual explanations (i.e., minimal changes in input features able to change the risk class) were generated for patients at high risk and evaluated using performance metrics (availability, validity, actionability, similarity, and discriminative power) and a qualitative survey administered to seven expert clinicians. Results showed that, on average, the requested minimum viable changes implied a significant reduction of fasting blood sugar, systolic blood pressure, and triglycerides and a significant increase of high-density lipoprotein in patients at risk of diabetes. A significant reduction in body mass index was also recommended in most of the patients at risk, except in females without hypertension. In general, greater changes were recommended in hypertensive patients compared to non-hypertensive ones. The experts were overall satisfied with the proposed approach although in some cases the proposed recommendations were deemed insufficient to reduce the risk in a clinically meaningful way. Future research will focus on a larger set of biomarkers and different comorbidities, also incorporating clinical guidelines whenever possible. Development of additional mathematical and clinical validation approaches will also be of paramount importance.

Artificial Intelligence Methodologies Applied to Technologies for Screening, Diagnosis and Care of the Diabetic Foot: A Narrative Review

Diabetic foot syndrome is a multifactorial pathology with at least three main etiological factors, i.e., peripheral neuropathy, peripheral arterial disease, and infection. In addition to complexity, another distinctive trait of diabetic foot syndrome is its insidiousness, due to a frequent lack of early symptoms. In recent years, it has become clear that the prevalence of diabetic foot syndrome is increasing, and it is among the diabetes complications with a stronger impact on patient's quality of life. Considering the complex nature of this syndrome, artificial intelligence (AI) methodologies appear adequate to address aspects such as timely screening for the identification of the risk for foot ulcers (or, even worse, for amputation), based on appropriate sensor technologies. In this review, we summarize the main findings of the pertinent studies in the field, paying attention to both the AI-based methodological aspects and the main physiological/clinical study outcomes. The analyzed studies show that AI application to data derived by different technologies provides promising results, but in our opinion future studies may benefit from inclusion of quantitative measures based on simple sensors, which are still scarcely exploited.

Diabetes Mellitus Disease Prediction Using Machine Learning Classifiers with Oversampling and Feature Augmentation

The technical improvements in healthcare sector today have given rise to many new inventions in the field of artificial intelligence. Patterns for disease identification are carried out, and the onset of prediction of many diseases is detected. Diseases include diabetes mellitus disease, fatal heart diseases, and symptomatic cancer. There are many algorithms that have played a critical role in the prediction of diseases. This paper proposes an ML based approach for diabetes mellitus disease prediction. For diabetes prediction, many ML algorithms are compared and used in the proposed work, and finally the three ML classifiers providing the highest accuracy are determined: RF, GBM, and LGBM. The accuracy of prediction is obtained using two types of datasets. They are Pima Indians dataset and a curated dataset. The ML classifiers LGBM, GB, and RF are used to build a predictive model, and the accuracy of each classifier is noted and compared. In addition to the generalized prediction mechanism, the data augmentation technique is also used, and the final accuracy of prediction is obtained for the classifiers LGBM, GB, and RF. A comparative study and demonstration between augmentation and non-augmentation are also discussed for the two datasets used in order to further improve the performance accuracy for predicting diabetes disease.

Machine learning for diabetes clinical decision support: a review

Type 2 diabetes has recently acquired the status of an epidemic silent killer, though it is non-communicable. There are two main reasons behind this perception of the disease. First, a gradual but exponential growth in the disease prevalence has been witnessed irrespective of age groups, geography or gender. Second, the disease dynamics are very complex in terms of multifactorial risks involved, initial asymptomatic period, different short-term and long-term complications posing serious health threat and related co-morbidities. Majority of its risk factors are lifestyle habits like physical inactivity, lack of exercise, high body mass index (BMI), poor diet, smoking except some inevitable ones like family history of diabetes, ethnic predisposition, ageing etc. Nowadays, machine learning (ML) is increasingly being applied for alleviation of diabetes health burden and many research works have been proposed in the literature to offer clinical decision support in different application areas as well. In this paper, we present a review of such efforts for the prevention and management of type 2 diabetes. Firstly, we present the medical gaps in diabetes knowledge base, guidelines and medical practice identified from relevant articles and highlight those that can be addressed by ML. Further, we review the ML research works in three different application areas namely—(1) risk assessment (statistical risk scores and ML-based risk models), (2) diagnosis (using non-invasive and invasive features), (3) prognosis (from normoglycemia/prior morbidity to incident diabetes and prognosis of incident diabetes to related complications). We discuss and summarize the shortcomings or gaps in the existing ML methodologies for diabetes to be addressed in future. This review provides the breadth of ML predictive modeling applications for diabetes while highlighting the medical and technological gaps as well as various aspects involved in ML-based diabetes clinical decision support.

A new predictive model for the concurrent risk of diabetic retinopathy in type 2 diabetes patients and the effect of metformin on amino acids

OBJECTIVE

This study established a model to predict the risk of diabetic retinopathy (DR) with amino acids selected by partial least squares (PLS) method, and evaluated the effect of metformin on the effect of amino acids on DR in the model.

METHODS

In Jinzhou, Liaoning Province, China, we retrieved 1031 patients with type 2 diabetes (T2D) from the First Affiliated Hospital of Liaoning Medical University. After sorting the amino acids using the PLS method, the top 10 amino acids were included in the model. Multivariate logistic regression was used to analyze the relationship between different amino acids and DR. And then the effects of metformin on amino acids were explored through interaction. Finally, Spearman's rank correlation analysis was used to analyze the correlation between different amino acids.

RESULTS

After sorting by PLS, Gly, Pro, Leu, Lyr, Glu, Phe, Tyr, His, Val and Ser were finally included in the DR risk prediction model. The predictive model after adding amino acids was statistically different from the model that only included traditional risk factors (p=0.001). Metformin had a significant effect on the relationship between DR and 7 amino acids (Gly, Glu, Phe, Tyr, His, Val, Ser, p<0.05), and the population who are not using metformin and have high levels of Glu (OR: 0.44, 95%CI: 0.27-0.71) had an additive protection effect for the occurrence of DR. And the similar results can be seen in high levels of Gly (OR: 0.46, 95%CI: 0.29-0.75), Leu (OR: 0.48, 95%CI: 0.29-0.8), His (OR: 0.46, 95%CI: 0.29-0.75), Phe (OR: 0.24, 95%CI: 0.14-0.42) and Tyr (OR: 0.41, 95%CI: 0.24 -0.68) in population who are not using metformin.

CONCLUSIONS

We established a prediction model of DR by amino acids and found that the use of metformin reduced the protective effect of amino acids on DR developing, suggesting that amino acids as biomarkers for predicting DR would be affected by metformin use.