International Journal on Advanced Science, Engineering and Information Technology, Vol. 12 (2022) No. 4, pages: 1565-1573, DOI:10.18517/ijaseit.12.4.17167

Characterization of Lard Profile from Different Geographical Regions of Husbandries and Body Parts of Pig using FTIR Technique Combined with Chemometrics

N.A.M Salleh, N. Yahaya, D. Rosli, M.F. Azizan, Bohari M Yamin, M.S. Hassan

Abstract

In halal authentication, there is a necessity for vigorous, rapid, faultless, easy, and cost-worthwhile approaches to characterize adulteration materials in food or highly processed products. Lard adulteration in food and other consumer products has been a primary concern, especially among Muslims. This research has been conducted to characterize lard profile collected from different geographical regions, which is southern, northern, and central of Malaysia peninsular and different pig body parts such as belly fats (BL), back fats (BK), and shoulder fat (SF) using Fourier transform infrared spectroscopy (FTIR) technique combined with chemometrics. The chemometrics evaluation on lard spectral was conducted by dividing two data sets into training for calibration model set and test set for validation model. Comparison between the Multiplicative Scattering Correction (MSC) and Second Derivative Savitzky-Golay (2nd DSG) data transformation was conducted to resolve the baseline issues. The Hotelling T Squared (T2) observed outliers detection and determined acceptance threshold based on the first three Principals Components (PCs). Spectral from FTIR spectroscopy showed a similar pattern of lard samples from different geographical regions and pig body parts. In this research, the PCA model of collected lards was established using Hotelling T2, and the quality of the model was determined using the projection method. These outcomes of the PCA- data-driven approach in this research has resolved the issue of obtaining a valid source of raw material for the preparation of certified reference material (CRM) of lard.

Keywords:

FTIR; Hotelling T2; lard; Multivariate; PCA

Viewed: 1141 times (since abstract online)

cite this paper     download