Tail-biting is globally recognized as a welfare concern for commercial swine production. Substantial research has been undertaken to identify risk factors and intervention methods to decrease and understand this vice. Tail-biting appears to be multifactorial and has proven difficult to predict and control. The primary objective of the scoping review was to identify and chart all available literature on the risk factors and interventions associated with tail-biting in pigs. A secondary objective was to identify gaps in the literature and identify the relevance for a systematic review. An online literature search of four databases, encompassing English, peer-reviewed and grey literature published from 1 January 1970 to 31 May 2019, was conducted. Relevance screening and charting of included articles were performed by two independent reviewers. A total of 465 citations were returned from the search strategy. Full-text screening was conducted on 118 articles, with 18 being excluded in the final stage. Interventions, possible risk factors, as well as successful and unsuccessful outcomes were important components of the scoping review. The risk factors and interventions pertaining to tail-biting were inconsistent, demonstrating the difficulty of inducing tail-biting in an experimental environment and the need for standardizing terms related to the behavior.

The time- or temperature-resolved detector signal from a thermoluminescence dosimeter can reveal additional information about circumstances of an exposure to ionising irradiation. We present studies using deep neural networks to estimate the date of a single irradiation with 12 mSv within a monitoring interval of 42 days from glow curves of novel TL-DOS personal dosimeters developed by the Materialprüfungsamt NRW in cooperation with TU Dortmund University. Using a deep convolutional network, the irradiation date can be predicted from raw time-resolved glow curve data with an uncertainty of roughly 1–2 days on a 68% confidence level without the need for a prior transformation into temperature space and a subsequent glow curve deconvolution (GCD). This corresponds to a significant improvement in prediction accuracy compared to a prior publication, which yielded a prediction uncertainty of 2–4 days using features obtained from a GCD as input to a neural network.

No abstract