A novel time series analysis approach for prediction of dialysis in critically ill patients using echo-state networks.

BACKGROUND: Echo-state networks (ESN) are part of a group of reservoir computing methods and are basically a form of recurrent artificial neural networks (ANN). These methods can perform classification tasks on time series data. The recurrent ANN of an echo-state network has an 'echo-state' characteristic. This 'echo-state' functions as a fading memory: samples that have been introduced into the network in a further past, are faded away. The echo-state approach for the training of recurrent neural networks was first described by Jaeger H. et al. In clinical medicine, until this moment, no original research articles have been published to examine the use of echo-state networks. METHODS: This study examines the possibility of using an echo-state network for prediction of dialysis in the ICU. Therefore, diuresis values and creatinine levels of the first three days after ICU admission were collected from 830 patients admitted to the intensive care unit (ICU) between May 31 th 2003 and November 17th 2007. The outcome parameter was the performance by the echo-state network in predicting the need for dialysis between day 5 and day 10 of ICU admission. Patients with an ICU length of stay <10 days or patients that received dialysis in the first five days of ICU admission were excluded. Performance by the echo-state network was then compared by means of the area under the receiver operating characteristic curve (AUC) with results obtained by two other time series analysis methods by means of a support vector machine (SVM) and a naive Bayes algorithm (NB). RESULTS: The AUC's in the three developed echo-state networks were 0.822, 0.818, and 0.817. These results were comparable to the results obtained by the SVM and the NB algorithm. CONCLUSIONS: This proof of concept study is the first to evaluate the performance of echo-state networks in an ICU environment. This echo-state network predicted the need for dialysis in ICU patients. The AUC's of the echo-state networks were good and comparable to the performance of other classification algorithms. Moreover, the echo-state network was more easily configured than other time series modeling technologies.

Support vector machine versus logistic regression modeling for prediction of hospital mortality in critically ill patients with haematological malignancies.

BACKGROUND: Several models for mortality prediction have been constructed for critically ill patients with haematological malignancies in recent years. These models have proven to be equally or more accurate in predicting hospital mortality in patients with haematological malignancies than ICU severity of illness scores such as the APACHE II or SAPS II 1. The objective of this study is to compare the accuracy of predicting hospital mortality in patients with haematological malignancies admitted to the ICU between models based on multiple logistic regression (MLR) and support vector machine (SVM) based models. METHODS: 352 patients with haematological malignancies admitted to the ICU between 1997 and 2006 for a life-threatening complication were included. 252 patient records were used for training of the models and 100 were used for validation. In a first model 12 input variables were included for comparison between MLR and SVM. In a second more complex model 17 input variables were used. MLR and SVM analysis were performed independently from each other. Discrimination was evaluated using the area under the receiver operating characteristic (ROC) curves (+/- SE). RESULTS: The area under ROC curve for the MLR and SVM in the validation data set were 0.768 (+/- 0.04) vs. 0.802 (+/- 0.04) in the first model (p = 0.19) and 0.781 (+/- 0.05) vs. 0.808 (+/- 0.04) in the second more complex model (p = 0.44). SVM needed only 4 variables to make its prediction in both models, whereas MLR needed 7 and 8 variables in the first and second model respectively. CONCLUSION: The discriminative power of both the MLR and SVM models was good. No statistically significant differences were found in discriminative power between MLR and SVM for prediction of hospital mortality in critically ill patients with haematological malignancies.

Determinants of the quality of basic life support by hospital nurses.

INTRODUCTION: Good quality basic life support (BLS) results in better survival. BLS is a core competence of nurses but despite regular refresher training, the quality of BLS is often poor and the reasons for this are not well known. We therefore investigated the relation between BLS quality and some of its potential determinants. MATERIALS AND METHODS: During a BLS refresher course, 296 nurses from non-critical care wards completed a questionnaire including demographic data and a "self-confidence" score. Subsequently, they performed a BLS test on a manikin connected to a PC using Skillreporting System software (Laerdal, Norway). The following variables were recorded: number of ventilations/min, tidal volume, number of compressions/min, compression rate, compression depth, "good ventilation" (n >or=4 min(-1) and tidal volume=700-1000 ml) and "good compression" (n >or=40 min(-1) and rate=80-120 min(-1) and compression depth=40-50mm). To detect independent determinants of BLS quality, associations between the demographic data and the objective variables of BLS quality were examined. RESULTS: Forty-three percent of the nurses rated their confidence as good or very good. Male gender was associated with good compression (P<0.001). Greater self-confidence was also associated with good ventilation (P<0.03) and with good compression (P<0.001). A short time since last BLS training was associated with a higher number of ventilations/min (P=0.01). A short time since last experience of CPR was associated with a higher number of compressions (P<0.01). CONCLUSIONS: Male gender, greater self-confidence, recent BLS training and recent CPR were associated with better quality of BLS.

Cardiovascular changes during endoscopic third ventriculostomy.

During an endoscopic third ventriculostomy (ETV) a sudden increase in intracranial pressure (ICP) may occur at any time. In the literature little attention has been paid to the early detection of such an increase. In particular the occurrence of a 'Cushing reflex' has not been discussed in this context. Therefore, we have now analysed retrospectively the anesthesia charts of 88 patients with obstructive hydrocephalus who had undergone ETV under general anesthesia. Monitoring included invasive blood pressure, electrocardiogram, end-expired carbon dioxide, pulse oximetry and heart rate. These variables were now evaluated before and after the introduction of the endoscope and during and after the occurrence of any change. In 67 patients the procedure had been uneventful. In 6 patients the occurrence of tachycardia and hypertension followed by bradycardia and hypertension was clearly the result of an increase in ICP, which we call a Cushing reflex. In his classical description of this pressure response Cushing reported the occurrence of hypertension, bradycardia and apnoea. However, many investigators have shown that beside systemic hypertension, both tachycardia and bradycardia are essential components of the Cushing reflex. Waiting for a persistent bradycardia to alert the surgeon during ETV can allow a fatal asystole.