350
M. Manley and P. J. Williams
with an SEP result of 0.35 and an RPD value of 3.82. NIR spectroscopy meat and
meat product applications have been comprehensively reviewed by Prieto et al. [5].
Discrimination studies in meat receive continuous attention with the main issues
being replacement of meat cuts of high value with cuts less costly. The potential of
NIR spectroscopy to distinguish between different meat cuts has been considered
extensively over the years. Discrimination between raw pork, chicken and turkey [4]
as well as kangaroo and beef [23] have been demonstrated. McElhinney et al. [24]
determined the lamb content mixed in raw minced beef. Classification accuracies of
more than 85% were obtained when raw beef, lamb, pork and chicken were classified
[4]. Meat properties such as intramuscular fat, fatty acids as well as muscle structure
and type of muscle fibres would have contributed to the discrimination in these
studies.
One of the earlier meat adulteration studies addressed the concern of selling
frozen-then-thawed meat as fresh meat cuts [4], which is still evaluated today [25].
A more recent study also included assessment of mince beef adulteration [26] while
the most recent study classified turkey meat products [27].
15.4 Fish and Fish Products
Quantification of moisture, fat and free fatty acids is the most common analysis
performed with NIR spectroscopy on fish [28]. Because fish is highly perishable,
microbial spoilage and freshness are important quality characteristics. Selling frozenthen-thawed fish as fresh is an important concern for the fish industry. Zhou et al. [29]
demonstrated the use of NIR spectroscopy to determine freshness in fish flesh. As
was the case for meat, heterogeneity of the fish samples resulted in NIR spectroscopy
having limited potential for prediction of sensory properties. Subjectivity of the taste
panel, which is the reference method in this case, could also have contributed to
inaccurate predictions. The first application of the use of a small (miniature) handheld
NIR device was demonstrated in fish authenticity studies [30] and is now increasingly
evaluated [31] with the added advantage of onsite analysis.
15.5 Milk and Milk Products
Milk, a turbid and opaque liquid, is a challenging commodity to analyse with NIR
spectroscopy. This is due to milk being a suspension containing fat globules and
casein micelles which cause it to be a highly scattering medium [32]. NIR spectroscopy has initially only been used on low moisture products such as milk powders.
Nowadays, it is widely used covering the entire range of dairy products. Holroyd [33]
extensively reviewed the application of NIR spectroscopy in milk and milk products
while Cattaneo and Holroyd [34] reviewed determination of adulteration and contamination in milk and milk powder. Chemical composition predictions in cheese, i.e. dry
M. Manley and P. J. Williams
with an SEP result of 0.35 and an RPD value of 3.82. NIR spectroscopy meat and
meat product applications have been comprehensively reviewed by Prieto et al. [5].
Discrimination studies in meat receive continuous attention with the main issues
being replacement of meat cuts of high value with cuts less costly. The potential of
NIR spectroscopy to distinguish between different meat cuts has been considered
extensively over the years. Discrimination between raw pork, chicken and turkey [4]
as well as kangaroo and beef [23] have been demonstrated. McElhinney et al. [24]
determined the lamb content mixed in raw minced beef. Classification accuracies of
more than 85% were obtained when raw beef, lamb, pork and chicken were classified
[4]. Meat properties such as intramuscular fat, fatty acids as well as muscle structure
and type of muscle fibres would have contributed to the discrimination in these
studies.
One of the earlier meat adulteration studies addressed the concern of selling
frozen-then-thawed meat as fresh meat cuts [4], which is still evaluated today [25].
A more recent study also included assessment of mince beef adulteration [26] while
the most recent study classified turkey meat products [27].
15.4 Fish and Fish Products
Quantification of moisture, fat and free fatty acids is the most common analysis
performed with NIR spectroscopy on fish [28]. Because fish is highly perishable,
microbial spoilage and freshness are important quality characteristics. Selling frozenthen-thawed fish as fresh is an important concern for the fish industry. Zhou et al. [29]
demonstrated the use of NIR spectroscopy to determine freshness in fish flesh. As
was the case for meat, heterogeneity of the fish samples resulted in NIR spectroscopy
having limited potential for prediction of sensory properties. Subjectivity of the taste
panel, which is the reference method in this case, could also have contributed to
inaccurate predictions. The first application of the use of a small (miniature) handheld
NIR device was demonstrated in fish authenticity studies [30] and is now increasingly
evaluated [31] with the added advantage of onsite analysis.
15.5 Milk and Milk Products
Milk, a turbid and opaque liquid, is a challenging commodity to analyse with NIR
spectroscopy. This is due to milk being a suspension containing fat globules and
casein micelles which cause it to be a highly scattering medium [32]. NIR spectroscopy has initially only been used on low moisture products such as milk powders.
Nowadays, it is widely used covering the entire range of dairy products. Holroyd [33]
extensively reviewed the application of NIR spectroscopy in milk and milk products
while Cattaneo and Holroyd [34] reviewed determination of adulteration and contamination in milk and milk powder. Chemical composition predictions in cheese, i.e. dry