Currently undertaking a PhD focusing deep learning applications in animal health.
We design and evaluate an assumption-free, deep learning-based methodology for animal health monitoring, specifically for the early detection of respiratory disease in growing pigs based on environmental sensor data. Two recurrent neural networks (RNNs), each comprising of gated recurrent units (GRUs), were used to create an autoencoder (GRU-AE) into which environmental data, collected from a variety of sensors, was processed to detect anomalies.An autoencoder is a type of network trained to reconstruct the patters it is fed as input. By training the GRU-AE using environmental data that did not lead to an increase in respiratory disease prevalence, data that did not fit the pattern of ``healthy environmental data'' would have a greater reconstruction error. All reconstruction errors were labelled as either normal or anomalous using a classifier optimised using particle swarm optimisation (PSO) from which alerts are raised. The results from the GRU-AE method outperformed state of the art techniques raising alerts when such predictions deviated from the actual observations. The results show that a change in the environment can result in an increase in pigs showing symptoms within 1 -- 7 days meaning that there is a period of time in which farmers can act to mitigate the negative effect of respiratory diseases such as porcine reproductive and respiratory syndrome (PRRS), a common and destructive virus found in pigs.
Jake Cowton, Ilias Kyriazakis, Thomas Plötz, Jaume Bacardit: Pre-print
Jake Cowton, Longzhi Yang (2015). The 15th UK Workshop on Computational Intelligence. IEEE
In this paper we present newly launched services for open data and for long-term preservation and reuse of high-energy-physics data analyses based on the digital library software Invenio. We track the ”data continuum” practices through several progressive data analysis phases up to the final publication. The aim is to capture for subsequent generations all digital assets and associated knowledge inherent in the data analysis process, and to make a subset available rapidly to the public. The ultimate goal of the analysis preservation platform is to capture enough information about the processing steps in order to facilitate reproduction of an analysis even many years after its initial publication, permitting to extend the impact of preserved analyses through future revalidation and recasting services. A related ”open data” service was launched for the benefit of the general public.
Jake Cowton, Sunje Dallmeier-Tiessen, Pamfilos Fokianos, L Rueda, P Herterich, J Kunčar, T Šimko, Tim Smith (2015). Open data and data analysis preservation services for LHC experiments. In Journal of Physics: Conference Series (Vol. 664, No. 3, p. 032030). IOP Publishing.