Materiały źródłowe

• #1 Predicting major outcomes in type 1 diabetes: a model development and validation study

  https://pmc.ncbi.nlm.nih.gov/articles/PMC4399797/

  Type 1 diabetes is associated with a higher risk of major vascular complications and death. A reliable method that predicts these outcomes early in the disease process would be helpful in risk classification. We therefore developed such a prognostic model and quantified its performance in independent cohorts. […] Major outcomes included major coronary heart disease, stroke, end-stage renal failure, amputations, blindness and all-cause death. […] Our prognostic model that uses easily accessible clinical features can discriminate between type 1 diabetes patients with good and poor prognosis. Such a prognostic model may be helpful in clinical practice and for risk stratification in clinical trials. […] Absolute risk predictions in individual patients with type 1 diabetes are important to timely identify the patients at high risk of major outcomes in order to enable strategies to prevent the development of such complications and to reduce health care costs.

• #1 Prediction models for risk of developing type 2 diabetes: systematic literature search and independent external validation study | The BMJ

  https://www.bmj.com/content/345/bmj.e5900

  Objective To identify existing prediction models for the risk of development of type 2 diabetes and to externally validate them in a large independent cohort. […] Most basic prediction models can identify people at high risk of developing diabetes in a time frame of five to 10 years. Models including biomarkers classified cases slightly better than basic ones. Most models overestimated the actual risk of diabetes. Existing prediction models therefore perform well to identify those at high risk, but cannot sufficiently quantify actual risk of future diabetes. […] An evaluation of the performance of 25 prediction models for type 2 diabetes in an independent Dutch cohort with over 10 years of follow-up showed that basic models perform similarly well in identifying individuals at high and low risk of developing diabetes. The performance was slightly better for extended models that included conventional biomarkers.

• #1 Development of a predictive risk model for all-cause mortality in patients with diabetes in Hong Kong | BMJ Open Diabetes Research & Care

  https://drc.bmj.com/content/9/1/e001950

  Patients with diabetes mellitus are at risk of premature death. […] A multiparametric model incorporating variables from different domains predicted all-cause mortality accurately in type 2 diabetes mellitus. […] The predictive and modeling capabilities of machine/deep learning survival analysis achieved more accurate predictions. […] We developed a machine learning-driven predictive risk model for type 2 diabetes mellitus using a multiparametric approach with data from different domains. […] Measures of variability of fasting glucose and HbA1c show similar predictive power for all-cause mortality, regardless of whether adjustments were made for initial values or mean values across follow-up. […] A multiparametric predictive risk model incorporating variables from different domains, including baseline demographics, comorbidities and laboratory tests, measures of variability of HbA1c and fasting blood glucose predicted all-cause mortality accurately.

• #1 Predictive Value of IL-6 and PDGF-AA for Short-term Mortality Risk in | IJGM

  https://www.dovepress.com/predictive-value-of-il-6-and-pdgf-aa-for-28-day-mortality-risk-in-crit-peer-reviewed-fulltext-article-IJGM

  In conclusion, IL-6 and PDGF-AA levels are independent risk factors associated with the 28-day mortality in critically ill patients. This study highlights the importance of monitoring serum levels of IL-6 and PDGF-AA in critically ill patients. Compared with the marker alone, combinations with other conventional risk factors have better predictive values. […] The logistic regression analysis revealed that IL-6 and PDGF-AA were both independent risk factors associated with 28-day mortality (OR=1.003, 95% CI (1.0011.005), OR=1.002, 95% CI (1.0011.003)). […] Our study showed that the HDL levels were lower in the non-survival group than in the survival group of critically ill patients. This funding is consistent with that of a previous report of critically ill patients with sepsis.2931 HDL appears to be an inflammatory marker because reduced levels reflect the intensity of the underlying inflammatory process.32 Low HDL-C levels are independently associated with higher all-cause mortality in critically ill patients.33 […] Overall, our data showed that serum IL-6 and PDGF-AA levels are significant for the diagnosis and prognosis assessment of critically ill patients. After admission, the IL-6, PDGF-AA and SOFA scores can be quickly assessed and clinicians can perform early diagnosis and treatment of patients.

• #1 Prediction model for cardiovascular disease in patients with diabetes using machine learning derived and validated in two independent Korean cohorts | Scientific Reports

  https://www.nature.com/articles/s41598-024-63798-y

  This study aimed to develop and validate a machine learning (ML) model tailored to the Korean population with type 2 diabetes mellitus (T2DM) to provide a superior method for predicting the development of cardiovascular disease (CVD), a major chronic complication in these patients. […] The random forest (RF) model exhibited the best overall performance among the models, with an AUROC of 0.830 (95% confidence interval [CI] 0.8180.842) in the discovery dataset and 0.722 (95% CI 0.6600.783) in the validation dataset. […] This study introduces a pioneering ML-based model for predicting CVD in Korean patients with T2DM, outperforming existing prediction tools and providing a groundbreaking approach for early personalized preventive medicine. […] Various predictive models using ML have recently been developed for CVD prediction in T2DM; however, their predictive power is limited because multiple risk factors have not been included in these models.

• #1 Predicting Mortality in Patients with Diabetes Starting Dialysis | PLOS One

  https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0089744

  While some prediction models have been developed for diabetic populations, prediction rules for mortality in diabetic dialysis patients are still lacking. […] A prediction model for mortality in diabetic dialysis patients could be a helpful tool in clinical decision-making. […] The primary aim of this study was to construct a prediction model to predict 1-year mortality in diabetic dialysis patients. […] A total of 394 patients were available for statistical analysis; 82 (21%) patients died within one year after baseline (3 months after starting dialysis therapy). […] The final prediction model contained seven predictors; age, smoking, history of macrovascular complications, duration of diabetes mellitus, Karnofsky scale, serum albumin and hemoglobin level. […] A prediction model containing seven predictors has been identified in order to predict 1-year mortality for diabetic incident dialysis patients.

• #1 Predicting Time to Diabetes Diagnosis Using Random Survival Forests | medRxiv

  https://www.medrxiv.org/content/10.1101/2024.02.03.24302304v1.full-text

  Type 2 Diabetes Mellitus (T2DM) is a chronic metabolic disorder with increasing population incidence. […] In this study we propose and assess a novel approach to diabetes prediction which integrates a specialized extension of the random forest algorithm known as random survival forest (RSF). […] Thus, RSF models have been shown to produce accurate timelines of diabetes onset trajectory, providing patients with quantifiable and relatable risks that are easy to understand. […] The results of our study have substantial implications for advancing machine learning in clinical decision support and patient outcome predictions, emphasizing the role of innovative models in improving predictive accuracy. […] By providing explicit time to diabetes diagnosis estimates, our model allows clinicians to better advise patients on their diabetes risk and collaborate towards personalized prevention plans for patients, thereby enhancing the efficacy of patient care strategies.

• #1 Predicting adverse outcomes due to diabetes complications with machine learning using administrative health data | npj Digital Medicine

  https://www.nature.com/articles/s41746-021-00394-8

  Across jurisdictions, government and health insurance providers hold a large amount of data from patient interactions with the healthcare system. We aimed to develop a machine learning-based model for predicting adverse outcomes due to diabetes complications using administrative health data from the single-payer health system in Ontario, Canada. Our model predicting three-year risk of adverse outcomes due to diabetes complications (hyper/hypoglycemia, tissue infection, retinopathy, cardiovascular events, amputation) included 700 features from multiple diverse data sources and had strong discrimination (average test AUC=77.7, range 77.7-77.9). Thus, predicting adverse outcomes due to diabetes complications is important for health system planning. Many prognostic models have been developed for diabetes complications in the clinical setting, including more recent applications of machine learning approaches. However, while these models are important for clinical level risk prediction, they are not easily deployed by governments or private health insurance providers at the population level which is precisely what is needed for addressing the aforementioned systemic barriers to diabetes complications care. The purpose of this study is to develop a single, large-scale machine learning model for common adverse outcome prediction due to diabetes complications that can be applied on routinely collected AHD for the purposes of public health planning and healthcare resource allocation. Our model can predict the 3-year risk of adverse outcomes due to diabetes complications (hyper/hypoglycemia, tissue infection, retinopathy, cardiovascular events, and amputation) with a test AUC of 77.7 (range 77.7-77.9). Adverse outcomes generally indicate significant negative events during a patient’s diabetes progression, and can be both detrimental to quality of life and place considerable cost burden on the healthcare system. Accurate advanced prediction can support preventative measures, and aid with how resources are deployed and managed within a health system.

• #1 Evaluation of Outcome Prediction for a Clinical Diabetes Database | SpringerLink

  https://link.springer.com/chapter/10.1007/978-3-540-30478-4_16

  Diabetes is a metabolic disorder which can be greatly affected by lifestyle. […] We investigate the potential for data mining in order to spot trends in the data and attempt to predict outcome. […] Feature selection has been used to improve the efficiency of the data mining algorithms and identify the contribution of different features to diabetes control status prediction. […] Decision trees can provide classification accuracy over 78%. […] However, while most bad control cases (90%) can be correctly classified, at least 50% of good control cases will be misclassified, which means that current feature selection and prediction models illustrate some potential but need additional refinement.

• #1 Advanced Ensemble Framework for Diabetes Outcome Forecasting

  https://www.ijraset.com/research-paper/advanced-ensemble-framework-for-diabetes-outcome-forecasting

  Diabetes is a chronic health condition characterized by high blood sugar levels. It occurs when the body either doesnt produce enough insulin or cant use it effectively. If left unmanaged, diabetes can lead to serious complications such as heart disease, kidney failure, and blindness. Developing a reliable diabetes prediction model is crucial for effective healthcare management. Machine learning techniques, such as RandomForest, KNeighborsClassifier, and LGBMClassifier, play a significant role in predicting diabetes risk. These algorithms analyze factors like age, gender, and various biochemical markers to identify patterns associated with diabetes onset. […] Moreover, understanding the significance of early diabetes detection cannot be overstated. Early diagnosis allows for timely interventions and personalized treatment plans, which can significantly improve patient outcomes and reduce the risk of complications associated with diabetes. By leveraging machine learning algorithms to accurately predict diabetes risk, healthcare professionals can provide proactive care and support to individuals at risk, ultimately leading to better healthcare management and improved quality of life.

• #1 Pima Indian Diabetes Prediction – Comet

  https://www.comet.com/site/blog/pima-indian-diabetes-prediction/

  This project aims to analyze the medical factors of a patient, such as Glucose Level, Blood Pressure, Skin Thickness, Insulin Level, and many others, to predict whether the patient has diabetes or not. […] The objective of the dataset is to diagnostically predict whether or not a patient has diabetes based on specific diagnostic measurements included in the dataset. […] The dataset consists of several medical predictor variables and one target variable, Outcome. […] The boxplot and violin plot show a strange relationship between the number of pregnancies and diabetes. According to the graphs, the increased number of pregnancies highlights an increased risk of diabetes. […] Glucose level plays a significant role in determining whether the patient has diabetes. Patients with a median glucose level of less than 120 are more likely to be nondiabetic. Patients with a median glucose level greater than 140 are more likely to be diabetic. Therefore, high glucose levels are a good indicator of diabetes.

• #2 Prediction of unfavorable outcome of acute decompensation of diabetes mellitus | Diabetology & Metabolic Syndrome | Full Text

  https://dmsjournal.biomedcentral.com/articles/10.1186/s13098-025-01605-y

  Using the method of spectral-probability analysis, to evaluate the possibility of predicting an unfavorable outcome of acute decompensation of diabetes mellitus in patients hospitalized in the intensive care unit using a mathematical model. […] Evaluation of the prognosis of patients with this pathology is important for choosing the tactics of management and observation. […] A group of mathematical criteria has been defined that makes it possible to identify patients with a high risk of an unfavorable course of the disease. […] Predicting the development of T1DM at an early preclinical stage and identifying children at high risk is important not only from the point of view of timely administration of insulin therapy to prevent acute life-threatening metabolic disorders during disease manifestation, but also the possibility of early therapeutic immunomodulatory intervention to prevent or delay the development of T1DM.

• #2 Predicting major outcomes in type 1 diabetes: a model development and validation study

  https://pmc.ncbi.nlm.nih.gov/articles/PMC4399797/

  We here describe a prognostic model based on easily accessible patient and clinical characteristics for the composite of the major outcomes and death, which would allow straightforward application of the model in practice. This model may help to identify patients with type 1 diabetes at high risk of major outcome events. […] The discriminative ability of the prognostic model was adequate in all cohorts with C-statistic of 0.79 in EDC, 0.74 in EDC recent and 0.73 in CACTI compared to a C-statistic of 0.74 in the development set. […] Predicting major outcomes permits establishment of a risk profile for individual patients with type 1 diabetes. Physicians may consider active intervention in the identified high risk patients. Such interventions can include intensifying insulin regimen and cardiovascular risk management.

• #2 Prediction models for risk of developing type 2 diabetes: systematic literature search and independent external validation study | The BMJ

  https://www.bmj.com/content/345/bmj.e5900

  All except two prediction models overestimated the absolute risk of diabetes in our validation dataset, which can partly be explained by the difference in incidence of diabetes between development and validation populations. To account for this, we adjusted the models for difference in incidence, resulting in much better calibration. Significant deviations between the predicted and observed risks, however, remained for most models. […] Results from our study show that prediction models perform well to identify those at high risk of future diabetes, being a first prerequisite for use of such models in practice as currently recommended.

• #2 Development of a predictive risk model for all-cause mortality in patients with diabetes in Hong Kong | BMJ Open Diabetes Research & Care

  https://drc.bmj.com/content/9/1/e001950

  Machine learning-driven algorithms further improved the accuracy of the predictive models. […] The findings of this study illustrate that machine/deep survival learning models can better capture the highly complex and nonlinear relationships between prognostic variables and an individual patient’s risk of mortality without prior variable selection or domain knowledge, compared with the traditional Cox analysis model. […] Application of machine/deep learning to survival analysis performs much better than the standard Cox model in predicting mortality risk of patients with diabetes mellitus.

• #2 Comprehensive Factors for Predicting the Complications of Diabetes Mellitus: A Systematic Review

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11327746/

  According to the results, fifty-nine features have been chosen as frequent features for predicting the selected CoDM. Among them, age, gender, ethnicity, weight, height, BMI, smoking history, HbA1c, SBP, eGFR, DBP, HDL, LDL, total cholesterol, triglyceride, use of insulin, duration of diabetes, and family history of diabetes CVD are determined as the persistent features in the prediction of CoDM.

• #2 Prediction model for cardiovascular disease in patients with diabetes using machine learning derived and validated in two independent Korean cohorts | Scientific Reports

  https://www.nature.com/articles/s41598-024-63798-y

  By identifying significant predictors within the model, we sought to understand the nuanced relationship between these factors and cardiovascular risk in patients with T2DM. […] Consequently, considering its superior and consistent results, the RF model emerged as the most effective predictor of CVD onset within a 3-year timeframe in patients with diabetes. […] The findings of this study can potentially improve patient outcomes by facilitating timely interventions, enhancing the understanding of contributing variables, and reducing the burden of cardiovascular complications in patients with diabetes. […] Our CVD prediction model demonstrated sufficiently good performance with a mean AUROC of 0.83, using only questionnaires, body measurements, and blood tests commonly conducted in clinical practice for patients with diabetes. […] This study used an ML-based CVD prediction model for patients with T2DM using comprehensive covariates and two independent cohorts from a multicenter registry in South Korea.

• #2 Predicting Mortality in Patients with Diabetes Starting Dialysis | PLOS One

  https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0089744

  Predictive performance of the model was good. […] This model will provide the opportunity to individualize treatment options. […] Despite these limitations this prediction model is the first model that predicts mortality in diabetic incident dialysis patients with good discriminative ability, indicated by the c-statistic of 0.810. […] In conclusion; a prediction algorithm for 1-year all-cause mortality has been developed for incident diabetic dialysis patients.

• #2 Predicting Time to Diabetes Diagnosis Using Random Survival Forests | medRxiv

  https://www.medrxiv.org/content/10.1101/2024.02.03.24302304v1.full-text

  Our model had a concordance index (C-Index) of 0.84, which is the evaluation measure for random survival forest models. […] A C-Index of 0.84 indicates high predictive performance and shows great potential in the baseline capabilities of RSF. […] This study advocates the use of random survival forest to predict the time to diabetes diagnosis, offering a quantifiable and relatable risk that is easy to comprehend. […] With a concordance index of 0.84 at baseline, random survival forest models have proven to be effective and accessible as a clinical tool with significant potential for further improvement.

• #2 Advanced Ensemble Framework for Diabetes Outcome Forecasting

  https://www.ijraset.com/research-paper/advanced-ensemble-framework-for-diabetes-outcome-forecasting

  The increasing prevalence of diabetes globally has underscored the importance of accurate prediction methods. However, existing methods often lack the precision and efficiency required for early detection. As a result, there is a growing need for a reliable system that can predict diabetes early on, aiding in better healthcare management. […] This automated system would enable healthcare professionals to identify individuals at risk of diabetes more effectively. Additionally, it would empower individuals to take proactive steps towards healthier lifestyles based on personalized risk assessments. […] The project adopts a systematic approach known as the Stacking Classifier ensemble model, which integrates the predictive capabilities of multiple base models specifically, the Random Forest Classifier, LightGBM Classifier, and KNearest Neighbors Classifier with a Logistic Regression meta-learner. This method involves a structured process of training and prediction.

• #2 Pima Indian Diabetes Prediction – Comet

  https://www.comet.com/site/blog/pima-indian-diabetes-prediction/

  Both the boxplot and violin plot provide a clear understanding of the relationship between blood pressure and diabetes. […] As observed in the boxplot, the median skin thickness is higher for diabetic patients than nondiabetic patients. […] Insulin is a major body hormone that regulates glucose metabolism. […] Both graphs highlight the role of BMI in diabetes prediction. […] The Diabetes Pedigree Function (DPF) calculates diabetes likelihood depending on the subjects age and diabetic family history. […] From the exploratory data analysis, I have concluded that the risk of diabetes depends upon the following factors: Glucose level, Number of pregnancies, Skin Thickness, Insulin level, BMI. […] With an increase in Glucose level, insulin level, BMI, and number of pregnancies, the risk of diabetes increases. […] Coming to the classification models, Logistic Regression outperformed Random Forest and SVM with 78% accuracy.

• #3 Predicting major outcomes in type 1 diabetes: a model development and validation study

  https://pmc.ncbi.nlm.nih.gov/articles/PMC4399797/

  In conclusion, a prognostic model is now available to assess the absolute risk of major outcomes in patients with type 1 diabetes. The prognostic model may be useful for providing individual risk estimates of major outcomes. The risk estimates can guide surveillance recommendations, inform patients and allow efficient design and analysis of clinical trials.

• #3 Prediction models for risk of developing type 2 diabetes: systematic literature search and independent external validation study | The BMJ

  https://www.bmj.com/content/345/bmj.e5900

  Objective To identify existing prediction models for the risk of development of type 2 diabetes and to externally validate them in a large independent cohort. […] Most basic prediction models can identify people at high risk of developing diabetes in a time frame of five to 10 years. Models including biomarkers classified cases slightly better than basic ones. Most models overestimated the actual risk of diabetes. Existing prediction models therefore perform well to identify those at high risk, but cannot sufficiently quantify actual risk of future diabetes. […] An evaluation of the performance of 25 prediction models for type 2 diabetes in an independent Dutch cohort with over 10 years of follow-up showed that basic models perform similarly well in identifying individuals at high and low risk of developing diabetes. The performance was slightly better for extended models that included conventional biomarkers.

• #3 Prognosis and Treatment Prediction of Type-2 Diabetes Using Deep Neural Network and Machine Learning Classifiers – ADS

  https://ui.adsabs.harvard.edu/abs/2023arXiv230103093K/abstract

  Type 2 Diabetes is a fast-growing, chronic metabolic disorder due to imbalanced insulin activity. The motion of this research is a comparative study of seven machine learning classifiers and an artificial neural network method to prognosticate the detection and treatment of diabetes with high accuracy, in order to identify and treat diabetes patients at an early age. […] We use performance measures such as accuracy and precision to find out the best algorithm deep ANN which outperforms with 95.14% accuracy among all other tested machine learning classifiers. […] We hope our high-performing model can be used by hospitals to predict diabetes and drive research into more accurate prediction models.

• #3 Establishing a Prediction Model for Weight Loss Outcomes After LSG in | DMSO

  https://www.dovepress.com/establishing-a-prediction-model-for-weight-loss-outcomes-after-lsg-in–peer-reviewed-fulltext-article-DMSO

  In obese Chinese patients with a BMI 32.5 kg/m2, the Inbody-based nomogram integrating REE/BW, FFMI, and WC offers an effective preoperative tool for predicting weight loss outcomes one year after LSG, facilitating surgical planning and postoperative management. […] The five most influential variables were subsequently included in a multivariate logistic regression model, which indicated that REE/BW, FFMI, and WC were independent predictors of weight loss outcomes. […] This study developed a preoperative predictive model for weight loss outcomes following LSG by integrating three key indicators: REE/BW, FFMI, and WC. […] The findings of this study are consistent with previous research. […] In conclusion, the combination of REE/BW, FFMI, and WC provides a relatively accurate preoperative prediction of weight loss outcomes one year post-LSG.

• #3 Predicting Mortality in Patients with Diabetes Starting Dialysis | PLOS One

  https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0089744

  While some prediction models have been developed for diabetic populations, prediction rules for mortality in diabetic dialysis patients are still lacking. […] A prediction model for mortality in diabetic dialysis patients could be a helpful tool in clinical decision-making. […] The primary aim of this study was to construct a prediction model to predict 1-year mortality in diabetic dialysis patients. […] A total of 394 patients were available for statistical analysis; 82 (21%) patients died within one year after baseline (3 months after starting dialysis therapy). […] The final prediction model contained seven predictors; age, smoking, history of macrovascular complications, duration of diabetes mellitus, Karnofsky scale, serum albumin and hemoglobin level. […] A prediction model containing seven predictors has been identified in order to predict 1-year mortality for diabetic incident dialysis patients.

• #3 Advanced Ensemble Framework for Diabetes Outcome Forecasting

  https://www.ijraset.com/research-paper/advanced-ensemble-framework-for-diabetes-outcome-forecasting

  The meta-learner adjusts the final prediction by weighing the predictions of the base estimators based on their individual performance, thereby leveraging the collective insights of the ensemble framework. This systematic approach ensures superior accuracy in diabetes outcome forecasting, providing valuable support for early detection and intervention strategies in diabetic patient care. […] By following this systematic approach for evaluation and analysis, the model provides actionable insights into diabetes prognosis. This helps healthcare practitioners implement personalized care strategies, ultimately improving patient outcomes. […] The future scope of diabetes prediction systems is vast. Advancements in analysis, contextual understanding, transfer learning, customization, privacy and ethics, multilingual support, and collaborative feedback will contribute to more effective and patient-centered systems, ultimately improving diabetes management and healthcare outcomes.

• #4 [2301.03093] Prognosis and Treatment Prediction of Type-2 Diabetes Using Deep Neural Network and Machine Learning Classifiers

  https://arxiv.org/abs/2301.03093

  Prognosis and Treatment Prediction of Type-2 Diabetes Using Deep Neural Network and Machine Learning Classifiers […] Type 2 Diabetes is a fast-growing, chronic metabolic disorder due to imbalanced insulin […] this research is a comparative study of seven machine learning classifiers and an artificial neural network method to prognosticate the detection and treatment of diabetes with high accuracy […] deep ANN which outperforms with 95.14% accuracy among all other tested machine learning […] hope our high-performing model can be used by hospitals to predict diabetes and drive research into more accurate prediction models.

Zobacz więcej