The standard GTM (generative topographic mapping) algorithm assumes that the data on which it is trained consists of independent, identically distributed (i.i.d.) vectors. For time series, however, the i.i.d. assumption is a poor apprcncimation. In this paper we show how the GTM algorithm can be extended to model time series by incorporating it as the emission density in a hidden Markov model. Since GTM has discrete hidden states we are able to find a tractable EM algorithm, based on the forward-backward algorithm, to train the model. We illustrate the performance of GTM through time using flight recorder data from a helicopter.

No abstract

No abstract

BioQT provides fully automated analysis, with confidence values for self-checking, on very large data sets such as Holter recordings. Automatic templating and expert reannotation of a small number of templates lead to a reduction in the sample size requirements for definitive QT studies.

Purpose: We sought to evaluate if incorporating an early warning system (EWS), the Visensia Safety Index (VSI) and the National Early Warning Systems 2 (NEWS2), may lead to earlier identification of rapid response team (RRT) patients. Methods: This was a retrospective study (2015-2018) of patients experiencing RRT activation within a tertiary care network. We evaluated the proportion of patients with an EWS alert prior to RRT activation and their associated outcomes (primary: hospital mortality). Results: There were 6,346 RRT activations over the study period. Of these, 2042 (50.8%) patients would have had a VSI alert prior to RRT activation, with a median advanced time of 3.6 (IQR 0.5-12.8) hours, compared to 2351 (58.4%) patients and 9.8 (IQR 2.0-18.7) hours for NEWS2. Patients with a potential alert prior to RRT activation had an increased odds of mortality for both VSI (OR 1.2, 95%CI 1.1-1.3) and NEWS2 (OR 2.7, 95% CI 2.4-3.1). Prognostic accuracy for hospital mortality was similar between groups. Conclusion: Utilization of an EWS by an RRT has potential to provide earlier recognition of deterioration and mortality risk among hospitalized inpatients.