Research@Lincoln
    • Login
     
    View Item 
    •   Research@Lincoln Home
    • Theses and Dissertations
    • Doctoral (PhD) Theses
    • View Item
    •   Research@Lincoln Home
    • Theses and Dissertations
    • Doctoral (PhD) Theses
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Automatic Deep Prediction: Computational early detection of dairy cattle Mastitis using semi-supervised learning : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University 

    Khamaysa Hajaya, Moaath
    Abstract
    Dairy cattle Mastitis is considered as one of the most notable and costly diseases in dairy industry worldwide; this disease severely affects dairy cattle and results in a costly cattle treatment, and causes a huge decrease in milk produced from sick cows. Mastitis also may cause other diseases that may affect the herd due to the bacterial infection. The total Mastitis cost to dairy industry in New Zealand is up near $280 million a year; this cost includes drop in milk production, low-grade milk quality, cattle treatment cost and other costs. This research includes the examination and analysis of data collected from a commercial robotic dairy farm, in order to design and build computational models that can help early and accurate detection of Mastitis in dairy cattle herds. Unlike previous studies, this research approaches the early detection by studying pre-clinical Mastitis. Pre-clinical Mastitis includes sick instances in any form prior to clinical Mastitis, including the form of subclinical Mastitis, which is the earliest form of the disease. This approach of defining early detection complies with the understanding of cattle Mastitis as a continuum of different forms/stages, which is more realistic than thinking of subclinical Mastitis as an isolated status of the disease. Early and accurate Mastitis detection helps cut treatment costs, control the disease, retain milk production levels and maintain milk quality grade. In addition to cutting financial costs, early detection helps cows by protecting them and relieving pain caused by the disease. Computational models can help achieve these by helping farmers to adopt timely and suitable cattle treatment regime, and by preventing healthy cows from being infected. The results of this research allow viewing dairy cattle Mastitis detection from a different angle, which brings about a broader understanding of some of the early signs and symptoms, leading to better control and management of the disease.... [Show full abstract]
    Keywords
    cattle; pre-clinical; subclinical; neural networks; deep learning; Keras; Tensorflow; clustering; classification; unsupervised; manifold learning; dimensionality reduction; time series; DeLaval; robotic; dairy; mastitis; Rectified Linear Unit (ReLU); t-distributed Stochastic Neighbour Embedding (t-SNE); Density-Based Spatial Clustering of Applications with Noise (DBSCAN); self-organizing map; self-organizing map (SOM); voluntary milking system (VMS)
    Fields of Research
    080108 Neural, Evolutionary and Fuzzy Computation; 080109 Pattern Recognition and Data Mining; 080110 Simulation and Modelling; 070703 Veterinary Diagnosis and Diagnostics; 070203 Animal Management
    Date
    2018
    Type
    Thesis
    Access Rights
    Restricted item - embargoed until 01 Dec 2021
    Collections
    • Doctoral (PhD) Theses [873]
    • Department of Environmental Management [1057]
    Share this

    on Twitter on Facebook on LinkedIn on Reddit on Tumblr by Email

    View/Open
    file embargoed until 2021-12-02
    file embargoed until 2021-12-02
    Metadata
     Expand record
    This service is managed by Learning, Teaching and Library
    • Archive Policy
    • Copyright and Reuse
    • Deposit Guidelines and FAQ
    • Contact Us
     

     

    Browse

    All of Research@LincolnCommunities & CollectionsTitlesAuthorsKeywordsBy Issue DateThis CollectionTitlesAuthorsKeywordsBy Issue Date

    My Account

    LoginRegister

    Statistics

    View Usage Statistics
    This service is managed by Learning, Teaching and Library
    • Archive Policy
    • Copyright and Reuse
    • Deposit Guidelines and FAQ
    • Contact Us