Prediction of functional capacity in chronic heart failure patients using myocardial tissue Doppler

Background. Exercise therapy improves functional capacity in CHF, but selection and individualization of training would be helped by a simple non-invasive marker of peak VO2. Peak VO2 in these pts is difficult to predict without direct measurement, and LV ejection fraction is a poor predictor. Myocardial tissue velocities are less load-dependent, and may be predictive of the exercise response in CHF pts. We sought to use tissue velocity as a predictor of peak VO2 in CHF pts. Methods. Resting 2D-echocardiography and tissue Doppler imaging were performed in 182 CHF pts (159 male, age 62±10 years) before and after metabolic exercise testing. The majority of these patients (129, 71%) had an ischemic cardiomyopathy, with resting EF of 35±13% and a peak VO2 of 13.5±4.7 ml/kg/min. Results. Neither resting EF (r=0.15) nor peak EF (r=0.18, both p=NS) were correlated with peak VO2. However, peak VO2 correlated with peak systolic velocity in septal (Vss, r=0.31) and lateral walls (Vsl, r=0.26, both p=0.01). In a general linear model (r2 = 0.25), peak VO2 was calculated from the following equation: 9.6 + 0.68*Vss - 0.09*age + 0.06*maximum HR. This model proved to be a superior predictor of peak VO2 (r=0.51, p=0.01) than the standard prediction equations of Wasserman (r= -0.12, p=0.01). Conclusions. Resting tissue Doppler, age and maximum heart rate may be used to predict functional capacity in CHF patients. This may be of use in selecting and following the response to therapy, including for exercise training.

Hemispheric asymmetry of neuroticism in the prediction of disinhibited work performance

Disinhibition is usually defined as a combination of high extraversion and high neuroticism or high extraversion and low neuroticism. The hypothesis that neuroticism interacts with aural preference (preferred-ear for listening) in the prediction of everyday types of disinhibited behaviour is tested. The importance of aural preference rests on the assumption that it is a readily available proxy measure of contra-hemispheric preference such that a left aural preference is indicative of right hemispheric preference and vice versa. Since the left hemisphere acts to initiate approach behaviour, the hypothesis investigates a model in which preference for the left hemisphere, together with high neuroticism, provides an alternative mechanism of disinhibition. This study provides evidence of the mechanism in the predicdon of disinhibited telesales performance.

Prediction of steady-state flux through variably saturated zones within a septic absorption trench

This paper describes effluent flow dynamics within a septic absorption system and the prediction of flow through the biomat and sub-biomat zone. Using soil hydraulic properties in a one dimensional model we demonstrate how soil hydraulic properties interact with biomat resistances to determine long-term acceptance rate (LTAR). The LTAR is a key parameter used in the Australian and New Zealand Standard AS1547:2000 to calculate the area of trench required to ensure trenches are not overloaded. Results show that several orders of magnitude variation in saturated hydraulic conductivity (Ks) collapse to a one order of magnitude variation in LTAR. These results are calculated from a model using basic flow theory, allowing LTAR to be estimated for any combination of biomat resistance and soil hydraulic properties. To increase the reliability of prediction of septic trench hydrology, HYDRUS 2D was used to model two dimensional flow. For more permeable soils, the exfiltration zone above sidewall biomat growth is shown to be a key pathway for excess effluent flow.

Exploiting sequence dependencies in the prediction of peroxisomal proteins

Prediction of peroxisomal matrix proteins generally depends on the presence of one of two distinct motifs at the end of the amino acid sequence. PTS1 peroxisomal proteins have a well conserved tripeptide at the C-terminal end. However, the preceding residues in the sequence arguably play a crucial role in targeting the protein to the peroxisome. Previous work in applying machine learning to the prediction of peroxisomal matrix proteins has failed W capitalize on the full extent of these dependencies. We benchmark a range of machine learning algorithms, and show that a classifier - based on the Support Vector Machine - produces more accurate results when dependencies between the conserved motif and the preceding section are exploited. We publish an updated and rigorously curated data set that results in increased prediction accuracy of most tested models.