DAYANE APARECIDA DOS SANTOS

Título da Tese: METODOLOGIAS ANALÍTICAS NA AVALIAÇÃO DE CORTES COMERCIAIS DE TILÁPIA DO NILO (Oreochromis niloticus)

Orientador: Prof. Dr. Ricardo Pereira Ribeiro

Data da Defesa: 27/05/2020

RESUMO GERAL:

Tilapia (Oreochromis niloticus) is the most cultivated fish species in Brazil. However, rapid analytical methodologies that aim to evaluate parameters related to tilapia are still scarce. In this sense, this work proposed to develop alternative methodologies for determining omega-3 (n-3) and omega-6 (n-6) and to determine the water retention capacity (WHC) and total lipid (TL) in fillets of Nile tilapia. In the determination of n-3 and n-6, an alternative methodology based on spectroscopy in the near-infrared region (NIR) is proposed through multivariate calibration by partial least squares (PLS). The spectra were acquired from vacuum-packed fillets, and the results achieved through external validation and parameters of merit indicate that the NIR/PLS models can be an alternative to the conventional method of n-3 and n-6 determination, highlighting the quickness of analysis, no sample preparation needs for the acquisition of NIR spectra, and (once the models have been validated) the non-generation of toxic waste. For the determination of WHC and TL parameters, the proposed alternative methodology is based on the construction of models based on regression by support vector machines (SVM) from parameters such as age at slaughter (AS), body weight at slaughter (BWS), fillet weight (FW) and fillet yield (FY). The results of the SVM model were compared with those of the reference method by the paired t test, which demonstrated that there were no significant differences between the methods at the 95% confidence level. This proposed methodology can help in the future to explore genetic and phenotypic characteristics in Nile tilapia.
KEYWORDS: Fish; NIR; PLS; SVM; Chemometrics.

Artigos Publicados Vinculados a Tese:

https://doi.org/10.21577/0103-5053.20200082