Application of NIRS adaptive technology for prediction of in vitro digestibility parameters of feed ingredients from Fermented Cocoa Pods

Abstract

Near Infrared Reflectance Spectroscopy (NIRS) is an alternative method that can be applied in the evaluation of feed ingredients. The purpose of this study was to evaluate the in vitro digestibility value of fermented cocoa pods using the NIRS method of feed analysis. This study used 18 samples of fermented cocoa shell waste, where the material used was fermented cocoa pod skin (KBK) then in vitro analysis was carried out. Samples were acquired spectrum by Thermo Integrative. Chemical analysis is intended to compare the data from chemical analysis and the NIRS method. The parameters analyzed were pH, dry matter digestibility (KCBK), and organic matter digestibility (KCBO). Spectrum data were processed using PLS with the pre-treatment methods of multiplicative scatter correction (MSC) and DeTrending (DT). Based on the analysis that has been carried out using the NIRS method with a prediction model that has been built on pH parameters, KCBK has a good predictive model with the DT pre-treatment method and KCBO has a good predictive model with the MSC method where (pH = LV: 8, r: 0 .79, R2: 0.62, RMSEC: 0.04 and RPD: 2.00; KCBK= LV: 8, r: 0.86, R2: 0.74, RMSEC: 1.30 and RPD: 2.02 ; KCBO= LV: 8, r: 0.88, R2: 0.78, RMSEC: 1.39 and RPD: 2.21). Rough predictions for pH were obtained with Non pre-treatment and MSC pre-treatment (pH and MSC = LV: 8, r: 0.74, R2: 0.55, RMSEC: 0.05, RPD 1.58) and rough predictions on KCBK were obtained with Non Preatment (KCBK=LV: 8, r: 0.83 R2:0.70, RMSEC: 1.30, and RPD 1.87).