Search

Automated acoustic monitoring methods are frequently used to survey bat activity around wind turbines. The algorithms are often based on spectral features or threshold values of the recordings. Due to the generality of these features, a lot of recordings are noise, making manual analysis and labelling of the recordings time consuming. In this paper, we present an approach based on convolutional neural networks to detect and classify bat calls respectively. Recordings are converted to Mel-frequency cepstral coefficients (MFCCs), which are then fed as an image into the convolutional neural networks (CNNs) for classification. A dataset consisting of 43585 recordings gathered at 5 m height was used to train and test this method. An accuracy of 99.7% was achieved on a test set for the binary classification of noise and bat calls. For the species classification, this approach achieved an accuracy of 96%. Both networks, trained on data gathered at 5 m, were also tested on recordings gathered at heights of 33 m, 65 m and 95 m. In case of the binary classification task, the results showed an increased rate of misclassifications among noise recordings. For species classification, there was a higher amount of misclassifcations among all species. [ABSTRACT FROM AUTHOR]

In order to provide an alternative treatment option to lung transplantation, we aim for the development of an implantable biohybrid lung (BHL), based on hollow fibre membrane (HFM) technology used in extracorporeal membrane oxygenators. Crucial for long-lasting durability of the BHL is the complete haemocompatibility of all blood contacting surfaces, which can be achieved by their endothelialisation. Indeed, HFM coating using fibronectin (FN) or albumin/heparin (AH) is essential to initially enable endothelial cell (EC) adhesion and monolayer confluence, but clinically relevant flow exposure results in EC detachment. We hypothesize, that optimized physiologic extracellular matrix (EM) synthesis increases flow-resilience for clinical BHL application. FN and AH coated HFM were endothelialized, cultivated under static conditions for 2 or 6 days and exposed to flow conditions for 24h. Quantitative cell counting after flow exposure was analyzed, EC confluence and EM-synthesis under static and dynamic conditions were assessed by immunofluorescence (IF) for VE-Cadherin (CD144) and Collagen-IV (ColIV). Expression levels of EM associated genes, like collagen-alpha-1(IV)chain (COL4A1), syndecan-2 (SDC2) Laminin-alpha4 (LAMA4), hyaluronan-synthase-2 (HAS2), exostosin-1 (EXT1) were analyzed by qRT-PCR. Regardless of HFM coating, IF indicated confluent monolayers and time-dependent augmentation of ColIV. Furthermore, higher cell numbers after flow exposure were detected after prolonged static cultivation. While COL4A1 expression levels remained constant under static and dynamic conditions, EM-associated genes SDC2, LAMA4, HAS2 and EXT1 were up-regulated under flow exposure. Prolonged static cultivation facilitates CollV augmentation improving monolayer resilience to flow exposure on both tested HFM. Moreover, further protein synthesis generating physiological EM requires flow stimulus. These results are valuable to further improve the HFM endothelialisation protocol for optimal EM-formation to resist clinically relevant flow conditions for long-term BHL application. [ABSTRACT FROM AUTHOR]