Journal of clinical monitoring and computing
-
J Clin Monit Comput · Apr 2024
Artificial intelligence model predicting postoperative pain using facial expressions: a pilot study.
This study aimed to assess whether an artificial intelligence model based on facial expressions can accurately predict significant postoperative pain. ⋯ ML models using facial expressions can accurately predict the presence of significant postoperative pain and have the potential to screen patients in need of rescue analgesia.
-
J Clin Monit Comput · Apr 2024
FiO2 prediction formula during low flow oxygen therapy in an adult model: a bench study.
During low-flow oxygen therapy, the true value of inspired oxygen fraction (FiO2) is generally unknown. Knowledge of delivered FiO2 values may be useful as well as to adjust oxygen therapy, as well as to predict patient deterioration. This study proposes a New FiO2 Prediction Formula (NFiO2) for low-flow oxygenation and compares its predictive value to precedent formulas. ⋯ Bias and limits of agreement between predicted FiO2 and benchtop FiO2 highlighted consistent differences between different FiO2 prediction formulas. The NFiO2 and the Duprez Formula 2018 seemed to be the most accurate formulas, followed by the Vincent Formula, and lastly the Shapiro Formula. A New FiO2 Prediction Formula was developed using clinical readily available variables (RR and O2 Flow rate) which showed good accuracy in predicting FiO2 during oxygenation at low flow.
-
J Clin Monit Comput · Apr 2024
An in-depth analysis of parameter settings and probability distributions of specific ordinal patterns in the Shannon permutation entropy during different states of consciousness in humans.
As electrical activity in the brain has complex and dynamic properties, the complexity measure permutation entropy (PeEn) has proven itself to reliably distinguish consciousness states recorded by the EEG. However, it has been shown that the focus on specific ordinal patterns instead of all of them produced similar results. Moreover, parameter settings influence the resulting PeEn value. ⋯ With the EEG data, we demonstrated that the probability P of monotonous patterns performs like PeEn in lower embedding dimension (m = 3, AUC = 0.88, [0.7, 1] in both), whereas the probability P of non-occurring patterns outperforms both methods in higher embedding dimensions (m = 5, PeEn: AUC = 0.91, [0.77, 1]; P(non-occurring patterns): AUC = 1, [1, 1]). We showed that the accuracy of PeEn in distinguishing consciousness states changes with different parameter settings. Furthermore, we demonstrated that for the purpose of separating wake from anaesthesia EEG solely pieces of information used for PeEn calculation, i.e., the probability of monotonous patterns or the number of non-occurring patterns may be equally functional.
-
J Clin Monit Comput · Apr 2024
Early prediction of mortality at sepsis diagnosis time in critically ill patients by using interpretable machine learning.
This study applied machine learning for the early prediction of 30-day mortality at sepsis diagnosis time in critically ill patients. Retrospective study using data collected from the Medical Information Mart for Intensive Care IV database. The data of the patient cohort was divided on the basis of the year of hospitalization, into training (2008-2013), validation (2014-2016), and testing (2017-2019) datasets. 24,377 patients with the sepsis diagnosis time < 24 h after intensive care unit (ICU) admission were included. ⋯ The calibration plot for the model revealed a slope of 1.03 (95% CI 0.94-1.12) and intercept of 0.14 (95% CI 0.04-0.25). The SHAP revealed the top three most significant features, namely age, increased red blood cell distribution width, and respiratory rate. Our study demonstrated the feasibility of using the interpretable machine learning model to predict mortality at sepsis diagnosis time.
-
Dynamic preload parameters are used to guide perioperative fluid management. However, reported cut-off values vary and the presence of a gray zone complicates clinical decision making. Measurement error, intrinsic to the calculation of pulse pressure variation (PPV) has not been studied but could contribute to this level of uncertainty. ⋯ The predicted range in reference PPV-value changes for a measured absolute change of 1% was [- 1.3%, 3.3%] and [- 1.9%, 4%] for these two methods. We showed that all methods that calculate PPV come with varying degrees of uncertainty. Accounting for bias and precision may have important implications for the interpretation of measured PPV-values or PPV-changes.