Journal of clinical monitoring and computing
-
J Clin Monit Comput · Apr 2020
Machine learning applied to multi-sensor information to reduce false alarm rate in the ICU.
Studies reveal that the false alarm rate (FAR) demonstrated by intensive care unit (ICU) vital signs monitors ranges from 0.72 to 0.99. We applied machine learning (ML) to ICU multi-sensor information to imitate a medical specialist in diagnosing patient condition. We hypothesized that applying this data-driven approach to medical monitors will help reduce the FAR even when data from sensors are missing. ⋯ While the FAR for PER with missing parameters was 0.17-0.39, it was only 0.01-0.02 for RF. When scenarios were examined separately, RF showed clear superiority in almost all combinations of scenarios and numbers of missing parameters. When sensor data are missing, specialist performance worsens with the number of missing parameters, whereas the RF model attains high accuracy and low FAR due to its ability to fuse information from available sensors, compensating for missing parameters.
-
J Clin Monit Comput · Apr 2020
Observational StudyPharmacodynamic modelling of the effect of remifentanil using the Pupillary Pain Index.
Using a targeted controlled infusion of remifentanil during total intravenous anesthesia, we investigated the effect-site concentrations of remifentanil that correlate with different values of the Pupillary Pain Index and which concentrations were necessary for achieving a Pupillary Pain Index ≤ 4 and its usefulness in titrating opioids. The Pupillary Pain Index was measured in 54 patients prior to surgery under different remifentanil effect-site concentrations and subsequently modeled. ⋯ For the probability of 80% of patients achieving a PPI score ≤ 4 the Ce of remifentanil was 4.39 ng/mL. We conclude that concentrations of remifentanil that have been shown to suppress movement in response to noxious stimulation correspond to a Pupillary Pain Index ≤ 4.
-
J Clin Monit Comput · Apr 2020
Simple calculation of the optimal insertion depth of esophageal temperature probes in children.
Placing an esophageal temperature probe (ETP) in the optimal esophageal site is important in various anesthetic and critical care settings to accurately monitor the core temperature of a pediatric patient. However, no reported study has provided a formula to calculate the optimal insertion depth of ETP placement in children based on direct measurement of the optimal depth. The aim of this study was to develop a simple and reliable method to determine the optimal depth of ETP placement in children via their mouth. ⋯ A linear regression analysis was performed to assess the correlation of the optimal depth of ETP placement with the children's age, weight, and height. The optimal depth of ETP placement had a greater correlation with height than with age or weight, and the best-fit equation was '0.180 × height + 6.749 (cm) (R2 = 0.920).' We obtained three simplified formulae, which showed no statistically significant difference in predicting the optimal depth of ETP placement: height/6 + 8 (cm), height/5 + 4 (cm), and height/5 + 5 (cm). The optimal depth of ETP via children's mouths has a close correlation with height and can be calculated with a simple formula 'height/5 + 5 (cm)'.
-
J Clin Monit Comput · Apr 2020
Determination of cardiac output from pulse pressure contour during intra-aortic balloon pumping in patients with low ejection fraction.
Evaluation of a new Windkessel model based pulse contour method (WKflow) to calculate stroke volume in patients undergoing intra-aortic balloon pumping (IABP). Preload changes were induced by vena cava occlusions (VCO) in twelve patients undergoing cardiac surgery to vary stroke volume (SV), which was measured by left ventricular conductance volume method (SVlv) and WKflow (SVwf). Twelve VCO series were carried out during IABP assist at a 1:2 ratio and seven VCO series were performed with IABP switched off. ⋯ Changes in SVlv and SVwf were directionally concordant in response to VCO's and during severe arrhythmia. (R2 = 0.868). The SVwf and SVlv methods are interchangeable with respect to measuring absolute stroke volume as well as tracking changes in stroke volume. The precision of the non-calibrated WKflow method is about 10% which improved to 7.5% after one calibration per patient.
-
J Clin Monit Comput · Apr 2020
Neuroanesthesiologists as interoperative neurophysiologists: a collaborative cognitive apprenticeship model of training in a community of clinical practice.
Directing intraoperative neurophysiologic monitoring (IONM) is a patient care activity for which no formal training programs exist, even though the need for well-trained practitioners is readily evident while caring for patients with diseases of the brain, spinal cord, spinal column, or nervous system. Here, we present the theoretical basis and institutional experience for a successful model of learning a new and complex set of skills: the medical direction of IONM. In a major academic institution, a clinical community of practice absorbed new members with professional backgrounds ranging from a recent neuroanesthesia fellowship to several decades of neuroanesthesia practice and trained them in a collaborative cognitive apprenticeship model to medically direct IONM. ⋯ The group has also trained four outside anesthesiologists-one of whom went on to become certified as a DABNM-who went on to develop the IONM program at a major children's hospital. This collaborative cognitive apprenticeship in anesthesiology to learn the medical direction of IONM is quite innovative as it integrates new members and expands the range of existing ones. In our model, the entire community is elevated by the reciprocal interactions of master clinicians, novice apprentices, and the community of practice.