Deep Learning Approaches for Robust Time of Arrival Estimation in Acoustic Emission Monitoring.

In this work, different types of artificial neural networks are investigated for the estimation of the time of arrival (ToA) in acoustic emission (AE) signals. In particular, convolutional neural network (CNN) models and a novel capsule neural network are proposed in place of standard statistical strategies which cannot handle, with enough robustness, very noisy scenarios and, thus, cannot be sufficiently reliable when the signal statistics are perturbed by local drifts or outliers. This concept was validated with two experiments: the pure ToA identification capability was firstly assessed on synthetic signals for which a ground truth is available, showing a 10× gain in accuracy when compared to the classical Akaike information criterion (AIC). Then, the same models were tested via experimental data acquired in the framework of a localization problem to identify targets with known coordinates on a square aluminum plate, demonstrating an overreaching precision under significant noise levels.

Prediction of gene expression regulation by human microRNAs in Plasmodium falciparum.

BACKGROUND: Malaria is a disease annually causing over 400,000 deaths. Deep understanding of molecular and genetic processes underlying its life cycle and pathogenicity is required to efficiently resist it. RNA interference is a mechanism of the gene expression regulation typical for a wide variety of species. Even though the existence of this phenomenon in Plasmodium falciparum has long been rejected, several recent works pose hypotheses and provide direct and indirect evidence of the existence of mechanisms similar to RNA interference in this organism. In particular, the possibility of regulation of P. falciparum gene expression through human microRNAs is of great importance both for fundamental biology and for medicine. In the present work we address the problem of possibility of the existence in the P. falciparum genome of the nucleotide sequences such that mRNAs transcribed from genes containing these sequences could form duplexes with human microRNAs. Using bioinformatics methods we have analysed genomes of 15 P. falciparum isolates for sequences homological to these microRNAs. RESULTS: The analysis has demonstrated the existence of a vast number of genes that could potentially be regulated by the human microRNAs in the plasmodial genome. CONCLUSIONS: Despite the fact that the numbers of homological intervals vary significantly between isolates, the hsa-miR-451a and hsa-miR-223-3p microRNAs are expected to make the most notable contribution to the pathogenesis of P. falciparum malaria. The majority of homological intervals occur in genes encoding cell adhesion proteins.