Performance evaluation of systems for automated monitoring of udder health: would the real gold standard please stand up?

G.A. Mein; M.D. Rasmussen

Performance evaluation of systems for automated monitoring of udder health: would the real gold standard please stand up?

In: Mastitis control

Authors:

G.A. Mein

G.A. Mein ¹Sensortec Ltd, Hamilton, New Zealand

Search for other papers by G.A. Mein in
Current site
Google Scholar
PubMed

and

M.D. Rasmussen

M.D. Rasmussen ²Dept.of Agricultural Engineering, University of Aarhus, Horsens, Denmark Corresponding author: mein@netspace.net.au

Search for other papers by M.D. Rasmussen in
Current site
Google Scholar
PubMed

Type:: Chapter

Pages:: 259–266

DOI:: https://doi.org/10.3920/9789086866496_048

Purchase instant access (PDF download and unlimited online access):

€42.00€35.00 excl. VAT

Add to Cart

Redeem Access Token

Data from a recent field study in New Zealand and another in Denmark were analysed to illustrate some real or perceived problems with finding the ‘right’ reference standard to use when evaluating the performance of automated monitoring and detection systems for udder health. Our analyses imply that a robust ‘gold standard’ cannot be derived from observations of clots in foremilk at a single milking per month. Inclusion of cows with a few flecks in their foremilk on isolated occasions is likely to result in over-diagnosis of clinical mastitis. Addition of a criterion of raised cell count or CMT >3 in foremilk from one or more quarters of a cow reduces the likelihood of over-diagnosis of clinical mastitis or abnormal milk. However, inclusion of cell count or CMT information is not necessary if the definition of a true clinical is based on observation of clots at 2 of 3 consecutive milkings. Annex C of ISO 20966 implies that diagnosis can be based on the first squirts of foremilk. A more objective test would be to discard the first three squirts of foremilk before sampling to determine the presence of clots. Our recommended definition of a ‘clinical episode’ (or ‘clinical case’) is: ‘Clots (> 2mm in average diameter) persisting beyond the first three squirts of foremilk for at least 2 of 3 consecutive milkings’. This definition would provide an effective compromise between robustness of the gold standard and the practical issues of conducting an evaluation of new sensing systems. Because the incidence of clinical cases is low and variable in most herds, a 36 h evaluation period may be too short to produce reliable data for robust statistical analyses.

Citation Info

Save
Cite
Email this content

Share link with colleague or librarian

You can email a link to this page to a colleague or librarian:
Email this content
or copy the link directly:

https://brill.edhh.ma/abstract/book/9789086866496/BP000048.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully

Collapse
Expand

Mastitis control

From science to practice

E-Book ISBN:: 9789086866496

Publisher:: Wageningen Academic

Print Publication Date:: 30 Oct 2008

Subjects
- Life Sciences

Metrics

	All Time	Past 365 days	Past 30 Days
Abstract Views	105	48	4
Full Text Views	6	3	0
PDF Views & Downloads	5	5	0

African Studies	Education	Media Studies
American Studies	History	Middle East and Islamic Studies
Ancient Near East and Egypt	Human Rights and Humanitarian Law	Musicology
Art History	International Law	Philosophy
Asian Studies	International Relations	Religious Studies
Biblical Studies	Jewish Studies	Slavic and Eurasian Studies
Biology	Languages and Linguistics	Social Sciences
Book History and Cartography	Life Sciences	Theology and World Christianity
Classical Studies	Literature and Cultural Studies

Subjects

Authors

Open Access

Product Information

Company

Contact

Performance evaluation of systems for automated monitoring of udder health: would the real gold standard please stand up?

Citation Info

Share link with colleague or librarian

Mastitis control

From science to practice

Table of Contents

Metrics

Metrics