Search

Cardiac safety assays incorporating label-free detection of human stem-cell derived cardiomyocyte contractility provide human relevance and medium throughput screening to assess compound-induced cardiotoxicity. In an effort to provide quantitative analysis of the large kinetic datasets resulting from these real-time studies, we applied bioinformatic approaches based on nonlinear dynamical system analysis, including limit cycle analysis and autocorrelation function, to systematically assess beat irregularity. The algorithms were integrated into a software program to seamlessly generate results for 96-well impedance-based data. Our approach was validated by analyzing dose- and time-dependent changes in beat patterns induced by known proarrhythmic compounds and screening a cardiotoxicity library to rank order compounds based on their proarrhythmic potential. We demonstrate a strong correlation for dose-dependent beat irregularity monitored by electrical impedance and quantified by autocorrelation analysis to traditional manual patch clamp potency values for hERG blockers. In addition, our platform identifies non-hERG blockers known to cause clinical arrhythmia. Our method provides a novel suite of medium-throughput quantitative tools for assessing compound effects on cardiac contractility and predicting compounds with potential proarrhythmia and may be applied to in vitro paradigms for pre-clinical cardiac safety evaluation. [ABSTRACT FROM AUTHOR]

Identification of regions on a vibrating structure which radiate energy to the far field is critical in many areas of engineering. Non-negative intensity is a means to visualize contributions of local surface regions to sound power from vibrating structures. Whilst the non-negative intensity has been used for structures under deterministic excitation due to structural forces or harmonic incident acoustic pressure excitation, it has not been considered for analyzing a structure under stochastic excitation. This work analytically formulates non-negative intensity in the wavenumber domain to investigate the surface areas on a vibrating planar structure that are contributing to the radiated sound power in the far field. The non-negative intensity is derived in terms of the cross spectrum density function of the stochastic field and the sensitivity functions of either the acoustic pressure or normal fluid particle velocity. The proposed formulation can be used for both infinite planar structure and finite plate in an infinite baffle. To demonstrate the technique, a simply supported baffled panel excited by a turbulent boundary layer as well as an acoustic diffuse field is considered and those regions contributing to the radiated sound power are identified. It is demonstrated that the non-negative intensity distribution is dependent on the stochastic excitation. It is also found that for a panel under stochastic excitation the more the non-negative intensity distribution is concentrated within the panel surface, the more efficient the panel radiates sound to the far field. [ABSTRACT FROM AUTHOR]