Classification-based deep neural network vs mixture density network models for insulin sensitivity prediction problem

Model-based glycemic control (GC) protocols are used to treat stress-induced hyperglycaemia in intensive care units (ICUs). The STAR (Stochastic-TARgeted) glycemic control protocol - used in clinical practice in several ICUs in New Zealand, Hungary, Belgium, and Malaysia - is a model-based GC protocol using a patient-specific, model-based insulin sensitivity to describe the patient's actual state. Two neural network based methods are defined in this study to predict the patient's insulin sensitivity parameter: a classification deep neural network and a Mixture Density Network based method. Treatment data from three different patient cohorts are used to train the network models. Accuracy of neural network predictions are compared with the current model- based predictions used to guide care. The prediction accuracy was found to be the same or better than the reference. The authors suggest that these methods may be a promising alternative in model-based clinical treatment for patient state prediction. Still, more research is needed to validate these findings, including in-silico simulations and clinical validation trials.

Estimating Enhanced Endogenous Glucose Production in Intensive Care Unit Patients with Severe Insulin Resistance

BACKGROUND

Critically ill ICU patients frequently experience acute insulin resistance and increased endogenous glucose production, manifesting as stress-induced hyperglycemia and hyperinsulinemia. STAR (Stochastic TARgeted) is a glycemic control protocol, which directly manages inter- and intra- patient variability using model-based insulin sensitivity (SI). The model behind STAR assumes a population constant for endogenous glucose production (EGP), which is not otherwise identifiable.

OBJECTIVE

This study analyses the effect of estimating EGP for ICU patients with very low SI (severe insulin resistance) and its impact on identified, model-based insulin sensitivity identification, modeling accuracy, and model-based glycemic clinical control.

METHODS

Using clinical data from 717 STAR patients in 3 independent cohorts (Hungary, New Zealand, and Malaysia), insulin sensitivity, time of insulin resistance, and EGP values are analyzed. A method is presented to estimate EGP in the presence of non-physiologically low SI. Performance is assessed via model accuracy.

RESULTS

Results show 22%-62% of patients experience 1+ episodes of severe insulin resistance, representing 0.87%-9.00% of hours. Episodes primarily occur in the first 24 h, matching clinical expectations. The Malaysian cohort is most affected. In this subset of hours, constant model-based EGP values can bias identified SI and increase blood glucose (BG) fitting error. Using the EGP estimation method presented in these constrained hours significantly reduced BG fitting errors.

CONCLUSIONS

Patients early in ICU stay may have significantly increased EGP. Increasing modeled EGP in model-based glycemic control can improve control accuracy in these hours. The results provide new insight into the frequency and level of significantly increased EGP in critical illness.

Differentiating Phenotypes of Coronavirus Disease-2019 Pneumonia by Electric Impedance Tomography

Introduction

Coronavirus disease-2019 (COVID-19) pneumonia has different phenotypes. Selecting the patient individualized and optimal respirator settings for the ventilated patient is a challenging process. Electric impedance tomography (EIT) is a real-time, radiation-free functional imaging technique that can aid clinicians in differentiating the “low” (L-) and “high” (H-) phenotypes of COVID-19 pneumonia described previously.

Methods

Two patients (“A” and “B”) underwent a stepwise positive end-expiratory pressure (PEEP) recruitment by 3 cmH₂O of steps from PEEP 10 to 25 and back to 10 cmH₂O during a pressure control ventilation of 15 cmH₂O. Recruitment maneuvers were performed under continuous EIT recording on a daily basis until patients required controlled ventilation mode.

Results

Patients “A” and “B” had a 7- and 12-day long trial, respectively. At the daily baseline, patient “A” had significantly higher compliance: mean ± SD = 53 ± 7 vs. 38 ± 5 ml/cmH₂O (p < 0.001) and a significantly higher physiological dead space according to the Bohr–Enghoff equation than patient “B”: mean ± SD = 52 ± 4 vs. 45 ± 6% (p = 0.018). Following recruitment maneuvers, patient “A” had a significantly higher cumulative collapse ratio detected by EIT than patient “B”: mean ± SD = 0.40 ± 0.08 vs. 0.29 ± 0.08 (p = 0.007). In patient “A,” there was a significant linear regression between the cumulative collapse ratios at the end of the recruitment maneuvers (R² = 0.824, p = 0.005) by moving forward in days, while not for patient “B” (R² = 0.329, p = 0.5).

Conclusion

Patient “B” was recognized as H-phenotype with high elastance, low compliance, higher recruitability, and low ventilation-to-perfusion ratio; meanwhile patient “A” was identified as the L-phenotype with low elastance, high compliance, and lower recruitability. Observation by EIT was not just able to differentiate the two phenotypes, but it also could follow the transition from L- to H-type within patient “A.”

Clinical Trial Registration

www.ClinicalTrials.gov, identifier: NCT04360837.