The development of a glucose prediction model in critically ill patients

Boorn, M. van den; Lagerburg, Vera; Steen, S. C. J. van; Wedzinga, R.; Bosman, R. J.; Voort, Peter H. J. van der

doi:10.1016/j.cmpb.2021.106105

Cited by 3 publications

(2 citation statements)

References 55 publications

(66 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Van den Boorn et al utilized a CGM dataset with a larger sample of ICU patients ( N = 94) that totaled 134,673 glucose measurements. Their model attempted to predict the glucose reading with a prediction horizon of 30 min based on the first and second derivatives of the CGM data, and achieved a mean squared difference of 7.39 mg/dL [ 46 ]. The time series approach to predicting glucose in the short-term horizon has not been limited to CGM data, but has also been applied to the MIMIC-III dataset from critical care unit patients.…”

Section: Machine Learning Models For Inpatient Glucose Predictionmentioning

confidence: 99%

Machine Learning Models for Inpatient Glucose Prediction

Zale

Mathioudakis

2022

Curr Diab Rep

View full text Add to dashboard Cite

Purpose of Review Glucose management in the hospital is difficult due to non-static factors such as antihyperglycemic and steroid doses, renal function, infection, surgical status, and diet. Given these complex and dynamic factors, machine learning approaches can be leveraged for prediction of glucose trends in the hospital to mitigate and prevent suboptimal hypoglycemic and hyperglycemic outcomes. Our aim was to review the clinical evidence for the role of machine learning–based models in predicting hospitalized patients’ glucose trajectory. Recent Findings The published literature on machine learning algorithms has varied in terms of population studied, outcomes of interest, and validation methods. There have been tools developed that utilize data from both continuous glucose monitors and large electronic health records (EHRs). With increasing sample sizes, inclusion of a greater number of predictor variables, and use of more advanced machine learning algorithms, there has been a trend in recent years towards increasing predictive accuracy for glycemic outcomes in the hospital setting. While current models predicting glucose trajectory offer promising results, they have not been tested prospectively in the clinical setting. Summary Accurate machine learning algorithms have been developed and validated for prediction of hypoglycemia and hyperglycemia in the hospital. Further work is needed in implementation/integration of machine learning models into EHR systems, with prospective studies to evaluate effectiveness and safety of such clinical decision support on glycemic and other clinical outcomes.

show abstract

Section: Machine Learning Models For Inpatient Glucose Predictionmentioning

confidence: 99%

Machine Learning Models for Inpatient Glucose Prediction

Zale

Mathioudakis

2022

Curr Diab Rep

View full text Add to dashboard Cite

show abstract

“…Such decision support tools range from paper-based to computer-based algorithms and include variables such as current/previous blood glucose, insulin administration, and calorie delivery [4,16]. It is intuitively appealing to include a closed-loop decision support tool that utilizes recent near-CGM results to titrate insulin administration, such that input from healthcare workers is no longer required [17]. However, recent studies have been single-center and have generally favored a computer-based decision support tool rather than a complete closed-loop system [18].…”

Section: Updates In Glucose Monitoringmentioning

confidence: 99%

Update on glucose control during and after critical illness

Deane

Plummer

Abdelhamid

2022

Current Opinion in Critical Care

View full text Add to dashboard Cite

Purpose of reviewThere is a complex bidirectional relationship between critical illness and disordered glucose metabolism. This review aims to provide a comprehensive summary of the recent evidence focused on the relationship between critical illness and disordered glucose metabolism through the distinct phases of prior to, during, and after an acute illness that requires admission to the intensive care unit (ICU).Recent findingsRecent data suggest that preexisting glucose metabolism affects the optimal blood glucose target during critical illness, with preliminary data suggesting that glucose targets should be ‘personalized’ based on preexisting glycemia. Because of the close association between critical illness and disordered glucose metabolism, there is a need to optimize glucose monitoring in the ICU with rapid, precise, and cost-efficient measurements at the bedside. Recent studies have evaluated the use of various methodologies, with a focus on the use of near-continuous glucose monitoring. For those patients with preexisting diabetes who survive ICU, nocturnal hypoglycemia may be an unrecognized and important issue when discharged to the ward. There is increasing evidence that patients with high blood glucose during their acute illness, so called ‘stress hyperglycemia’, are at increased risk of developing diabetes in the years following recovery from the inciting event. Critically ill patients with COVID-19 appear at greater risk.SummaryThere have been important recent insights in the approach to glucose monitoring and glucose targets during critical illness, monitoring and administration of glucose-lowering drugs on discharge from the ICU, and longitudinal follow-up of patients with stress hyperglycemia.

show abstract