Exciting times ahead! This year (2022), the University is investing in a new repository which will bring with it various benefits and improvements. Whilst the system will change, the UDORA name will remain the same. For further information on what this means, please see the Repository Project pages on the Library SharePoint site. 

Thank you! 


Access to Taylor and Francis microsite for free Covid-19 literature is available here. 


Welcome to UDORA, the University of Derby Online Research Archive.

UDORA is the institutional repository of research produced by staff at the University of Derby, and an archive of our completed doctoral theses.

If you are a member of staff ready to submit your research, please see our Quick Guide to Getting Started.

We welcome any feedback. Please contact UDORA@derby.ac.uk


Select a community to browse its collections.

  • Mobility Analysis during the 2020 Pandemic in a Touristic city: the Case of Cagliari

    Ferrara, Enrico; Uras, Marco; Atzori, Luigi; Bagdasar, Ovidiu; Liotta, Antonio; University of Derby; University of Cagliari; Free University of Bozen-Bolzano, Italy (IEEE, 2021-09-20)
    The impact of the 2020 COVID-19 pandemic has been significant on every aspect of life and has drastically changed our habits. Here we analyze an extensive set of traffic traces in Cagliari, one of the most touristic cities in the Mediterranean, to quantify how the different phases of the pandemic have affected not only traffic volumes but also their patterns. We put traffic in relation to different restriction levels, finding a non-linear relation. Following a 76% traffic reduction on the first lockdown, subsequent restrictions have lead to less sudden changes. We then use the official tourist-presence figures to pinpoint the traffic stations that are influenced by tourists’ mobility the most. All in all, our analysis shows that although the absolute traffic volumes roughly followed the pandemic evolution, the weekly traffic patterns changed drastically over the time, whereas the daily ones maintained more consistency.
  • The 24-h Movement Compositions in Weekday, Weekend Day or Four-Day Periods Differentially Associate with Fundamental Movement Skills

    Roscoe, Clare M. P.; Duncan, Michael, J; Clark, Cain, C. T; University of Derby; Coventry University (MDPI AG, 2021-09-22)
    The aim of this study was to investigate the relationship between weekday, weekend day and four-day physical activity (PA) behaviours and fundamental movement skills (FMS) in British preschool children from a low socio-economic status background using compositional data analysis (CoDA). One hundred and eighty-five preschool children aged 3–4 years provided objectively assessed PA and sedentary behaviour (SB) data (GENEActiv accelerometer) and FMS (TGMD-2). The association of 24-h movement behaviours with FMS was explored using CoDA and isotemporal substitution (R Core Team, 3.6.1). When data were considered compositionally (SB, light PA (LPA), moderate and vigorous PA (MVPA)) and adjusted for age, BMI and sex, the weekday-derived composition predicted total motor competence (r2 = 0.07), locomotor (r2 = 0.08) and object control skills (r2 = 0.09); the weekend day-derived composition predicted total motor competence (r2 = 0.03) and object control skills (r2 = 0.03), the 4-day-derived composition predicted total motor competence (r2 = 0.07), locomotor (r2 = 0.07) and object control skills (r2 = 0.06) (all p < 0.05). Reallocation of 5 min of LPA at the expense of any behaviour was associated with significant improvements in total motor competence, locomotor and object control skills; for weekend-derived behaviours, MVPA was preferential. Considering movement behaviours over different time periods is required to better understand the effect of the 24-h movement composition on FMS in preschool children.
  • Particulate and drug-induced toxicity assessed in novel quadruple cell human primary hepatic disease models of steatosis and pre-fibrotic NASH.

    Kermanizadeh, Ali; Valli, Jessica; Sanchez, Katarzyna; Hutter, Simon; Pawlowska, Agnieszka; Whyte, Graeme; Moritz, Wolfgang; Stone, Vicki; University of Derby; Heriot Watt University, Edinburgh; et al. (Springer, 2021-10-20)
    In an effort to replace, reduce and refine animal experimentation, there is an unmet need to advance current in vitro models that offer features with physiological relevance and enhanced predictivity of in vivo toxicological output. Hepatic toxicology is key following chemical, drug and nanomaterials (NMs) exposure, as the liver is vital in metabolic detoxification of chemicals as well as being a major site of xenobiotic accumulation (i.e., low solubility particulates). With the ever-increasing production of NMs, there is a necessity to evaluate the probability of consequential adverse effects, not only in health but also in clinically asymptomatic liver, as part of risk stratification strategies. In this study, two unique disease initiation and maintenance protocols were developed and utilised to mimic steatosis and pre-fibrotic NASH in scaffold-free 3D liver microtissues (MT) composed of primary human hepatocytes, hepatic stellate cells, Kupffer cells and sinusoidal endothelial cells. The characterized diseased MT were utilized for the toxicological assessment of a panel of xenobiotics. Highlights from the study included: 1. Clear experimental evidence for the pre-existing liver disease is important in the augmentation of xenobiotic-induced hepatotoxicity and 2. NMs are able to activate stellate cells. The data demonstrated that pre-existing disease is vital in the intensification of xenobiotic-induced liver damage. Therefore, it is imperative that all stages of the wide spectrum of liver disease are incorporated in risk assessment strategies. This is of significant consequence, as a substantial number of the general population suffer from sub-clinical liver injury without any apparent or diagnosed manifestations.
  • Relations Between Entropy and Accuracy Trends in Complex Artificial Neural Networks

    Cavallaro, Lucia; Grassia, Marco; Fiumara, Giacomo; Mangioni, Giuseppe; De Meo, Pasquale; Carchiolo, Vincenza; Bagdasar, Ovidiu; Liotta, Antonio; University of Derby; Università degli Studi di Catania, Italy; et al. (Springer, 2022-01-01)
    Training Artificial Neural Networks (ANNs) is a non-trivial task. In the last years, there has been a growing interest in the academic community in understanding how those structures work and what strategies can be adopted to improve the efficiency of the trained models. Thus, the novel approach proposed in this paper is the inclusion of the entropy metric to analyse the training process. Herein, indeed, an investigation on the accuracy computation process in relation to the entropy of the intra-layers’ weights of multilayer perceptron (MLP) networks is proposed. From the analysis conducted on two well-known datasets with several configurations of the ANNs, we discovered that there is a connection between those two metrics (i.e., accuracy and entropy). These promising results can be helpful in defining, in the future, new criteria to evaluate the training process goodness in real-time by optimising it and allow faster detection of its trend.
  • Embedded Data Imputation for Environmental Intelligent Sensing: A Case Study

    Erhan, Laura; Di Mauro, Mario; Anjum, Ashiq; Bagdasar, Ovidiu; Song, Wei; Liotta, Antonio; University of Derby; University of Salerno, 84084 Fisciano, Italy; University of Leicester; University of Alba Iulia, 510009 Alba Iulia, Romania; et al. (MDPI AG, 2021-11-23)
    Recent developments in cloud computing and the Internet of Things have enabled smart environments, in terms of both monitoring and actuation. Unfortunately, this often results in unsustainable cloud-based solutions, whereby, in the interest of simplicity, a wealth of raw (unprocessed) data are pushed from sensor nodes to the cloud. Herein, we advocate the use of machine learning at sensor nodes to perform essential data-cleaning operations, to avoid the transmission of corrupted (often unusable) data to the cloud. Starting from a public pollution dataset, we investigate how two machine learning techniques (kNN and missForest) may be embedded on Raspberry Pi to perform data imputation, without impacting the data collection process. Our experimental results demonstrate the accuracy and computational efficiency of edge-learning methods for filling in missing data values in corrupted data series. We find that kNN and missForest correctly impute up to 40% of randomly distributed missing values, with a density distribution of values that is indistinguishable from the benchmark. We also show a trade-off analysis for the case of bursty missing values, with recoverable blocks of up to 100 samples. Computation times are shorter than sampling periods, allowing for data imputation at the edge in a timely manner.

View more