A harmonic linear dynamical system for prominent ECG feature extraction.

Unsupervised mining of electrocardiography (ECG) time series is a crucial task in biomedical applications. To have efficiency of the clustering results, the prominent features extracted from preprocessing analysis on multiple ECG time series need to be investigated. In this paper, a Harmonic Linear Dynamical System is applied to discover vital prominent features via mining the evolving hidden dynamics and correlations in ECG time series. The discovery of the comprehensible and interpretable features of the proposed feature extraction methodology effectively represents the accuracy and the reliability of clustering results. Particularly, the empirical evaluation results of the proposed method demonstrate the improved performance of clustering compared to the previous main stream feature extraction approaches for ECG time series clustering tasks. Furthermore, the experimental results on real-world datasets show scalability with linear computation time to the duration of the time series.

Microscopic observation drug susceptibility assay (MODS) for early diagnosis of tuberculosis in children.

MODS is a novel liquid culture based technique that has been shown to be effective and rapid for early diagnosis of tuberculosis (TB). We evaluated the MODS assay for diagnosis of TB in children in Viet Nam. 217 consecutive samples including sputum (n = 132), gastric fluid (n = 50), CSF (n = 32) and pleural fluid (n = 3) collected from 96 children with suspected TB, were tested by smear, MODS and MGIT. When test results were aggregated by patient, the sensitivity and specificity of smear, MGIT and MODS against "clinical diagnosis" (confirmed and probable groups) as the gold standard were 28.2% and 100%, 42.3% and 100%, 39.7% and 94.4%, respectively. The sensitivity of MGIT and MODS was not significantly different in this analysis (P = 0.5), but MGIT was more sensitive than MODS when analysed on the sample level using a marginal model (P = 0.03). The median time to detection of MODS and MGIT were 8 days and 13 days, respectively, and the time to detection was significantly shorter for MODS in samples where both tests were positive (P<0.001). An analysis of time-dependent sensitivity showed that the detection rates were significantly higher for MODS than for MGIT by day 7 or day 14 (P<0.001 and P = 0.04), respectively. MODS is a rapid and sensitive alternative method for the isolation of M.tuberculosis from children.