Currently undertaking a PhD focusing deep learning applications in animal health.
We designed and evaluated 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 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 patterns it is fed as input. By training the GRU-AE using environmental data that did not lead to an occurrence of respiratory disease, data that did not fit the pattern of “healthy environmental data” had a greater reconstruction error. All reconstruction errors were labelled as either normal or anomalous using threshold-based anomaly detection optimised with 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 occurrences of pigs showing symptoms of respiratory disease within 1–7 days, meaning that there is a period of time during which their keepers can act to mitigate the negative effect of respiratory diseases, such as porcine reproductive and respiratory syndrome (PRRS), a common and destructive disease endemic in pigs.
Jake Cowton, Ilias Kyriazakis, Thomas Plötz, Jaume Bacardit. A Combined Deep Learning GRU-Autoencoder for the Early Detection of Respiratory Disease in Pigs Using Multiple Environmental Sensors. Sensors 2018, 18, 2521.
Flood risk and associated impacts are major societal and policy concerns following widespread flooding in December 2015, which cost the UK economy an estimated £5 billion. Increasing advocacy for alternatives to conventional hard engineering solutions is accompanied by demands for evidence. This study provides a systematic review and meta‐analysis of direct evidence for the effect of tree cover on channel discharge. The results highlighted a deficiency in direct evidence. From 7 eligible studies of 156 papers reviewed, the results show that increasing tree cover has a small statistically significant effect on reducing channel discharge. Meta‐analysis reveals that tree cover reduces channel discharge (standardised mean difference −0.35, 95%CI, −0.71 to 0.00), but the effect was variable (I2 = 81.91%), the potential for confounding was high, and publication bias is strongly suspected (Egger Test z = 3.0568, p = .002). Due to the lack of direct evidence the overall strength of evidence is low, indicating high uncertainty. Further primary research is required to understand reasons for heterogeneity and reduce uncertainty. A Bayesian network parameterised with data from the meta‐analysis supports investment in integrated catchment management, particularly on infrastructure density and water storage (reservoirs), for effective responses to flood risk.
Jayne Carrick, Mohd Shaiful Azman Bin Abdul Rahim, Cosmos Adjei, Hassan Habib Hassan Ashraa Kalee, Steven James Banks, Friederike Charlotte Bolam, Ivone Maritza Campos Luna, Beth Clark, Jake Cowton, Israel Freitas Nongando Domingos, David Duba Golicha, Garima Gupta, Matthew Grainger, Gultakin Hasanaliyeva, David John Hodgson, Elisa Lopez‐Capel, Amelia Jo Magistrali, Ian George Merrell, Idiegberanoise Oikeh, Mwanajuma Salim Othman, Thilanka Kumari Ranathunga Ranathunga Mudiyanselage, Carl Warren Charles Samuel, Enas KH Sufar, Philip Alexander Watson, Nik Nur Azwanida Binti Zakaria, Gavin Stewart. Is planting trees the solution to reducing flood risks? Journal of Flood Risk Management 2018
Jake Cowton, Longzhi Yang. The 15th UK Workshop on Computational Intelligence. 2015. 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.