Estimation of somatic cell count levels of hard cheeses using physicochemical composition and artificial neural networks

Abstract

This study addresses the prediction of the somatic cell counts of the milk used in the production of sheep cheese using artificial neural networks. To achieve this objective, the neural network was designed using 33 parameters of the physicochemical composition of the cheeses obtained after they have been matured for 12 mo as input data. The physicochemical analysis of the cheeses revealed that the somatic cell count level of the cheese has a significant influence on the amount of protein, fat, dry extract, and fatty acids. When properly set up, the neural network allows the correct classification of the cheeses (100% of correct results in both training and test phases) and therefore their samples in each of the 3 nominal output variables (low, average, and high somatic cell counts). The fatty composition of the cheeses, individual fatty acids, and fat acidity are the variables that most affect the correct operation of the neural network.

Fatty acids profile of Serra da Estrela PDO cheeses and respective atherogenic and thrombogenic indices

“…The quantified levels of FFA are in accordance with those determined by Carocho et al (2015, 2016a, 2016b) for this type of PDO cheese. Also, SEC fatty acids profiles were within the same order of magnitude of those found in semi-hard uncooked Italian cheese (Cabiddu et al , 2006), Manchego Spanish cheese (Gómez-Cortés et al , 2009), Chilean, French and Spanish commercial cheeses (Aguilar et al , 2014), Serbian white brined cheeses (Barać et al , 2018), Portuguese Terrincho ewe cheese (Pinho et al , 2003) or of hard ewe milk cheeses (Hernández-Ramos et al , 2018).…”

“…The quantified levels of FFA are in accordance with those determined by Carocho et al (2015Carocho et al ( , 2016aCarocho et al ( , 2016b for this type of PDO cheese. Also, SEC fatty acids profiles were within the same order of magnitude of those found in semi-hard uncooked Italian cheese (Cabiddu et al, 2006), Manchego Spanish cheese (G omez-Cortés et al, 2009), Chilean, French and Spanish commercial cheeses (Aguilar et al, 2014), Serbian white brined cheeses (Bara c et al, 2018), Portuguese Terrincho ewe cheese (Pinho et al, 2003) or of hard ewe milk cheeses (Hernández-Ramos et al, 2018). Fatty acids profiles of SECs (Tables I-III) were significantly influenced by cheese producer, geographical origin or moment of production, probably due to the difference in the milk used, since the FFA composition of the milk fat depends on the animals' feed, the season and the stage of lactation (Cabiddu et al, 2006;Jaramillo et al, 2008;Vera et al, 2009;G omez-Cortés et al, 2015;Giorgio et al, 2019;Nájera et al, 2017).…”

Assessing Serra da Estrela PDO cheeses’ origin-production date using fatty acids profiles

“…Finally, it should be referred that the overall fatty acids profile established in the present study was in line with other previous reports for Serra da Estrela cheeses [6,7,12,13,16]. Moreover, compared to other cheeses produced from ewe milk, similar fatty acids profiles were obtained although the individual relative abundances could be different but within the same order of magnitude, like semi-hard uncooked Italian cheese [35], Manchego Spanish cheese [36], Kashkaval Greek cheese [37], Tulum Turkish cheese [38], Chilean, French and Spanish commercial cheeses [39], Serbian white brined cheeses [1] or of hard ewe milk cheeses [40].…”

Application of artificial neural networks to prediction of new substances with antimicrobial activity againstEscherichia coli

“…The neural network tool also provides promising results from medical and pharmaceutical research, such as predicting the properties of potential drugs or predicting the onset of diseases, complications or other health problems based on patient data (Saritas et al 2010;Wnuk et al 2013;Badura et al 2019). Recently, ANNs have also become a useful predictive tool in food technology and production that allow to solve as specific problems as prediction of Buci nski et al 2004b) for example they are employed in cheese production, to predict the number of somatic cells of milk used for sheep cheese production (Poirazi et al 2007;Hernandez-Ramos et al 2019). An interesting application of this method is the use of ANN classification models to optimize the wine production process.…”

“…2007; Hernandez‐Ramos et al . 2019). An interesting application of this method is the use of ANN classification models to optimize the wine production process.…”