Rapid Detection of the Changes in Fresh Beef Using Portable Near Infrared Spectrometer

Abstract

Proses pascapanen daging sapi segar dimulai segera setelah sapi disembelih. Penanganan pascapanen mempengaruhi kualitas dan keamanan karkas melalui rantai pasok pascapanen mulai dari rumah pemotongan hewan, distribusi, pengolahan dan konsumsi. Penanganan pascapanen daging sapi mempengaruhi sifat fisikokimia seperti pH, warna, dan pertumbuhan mikroba. Penilaian kualitas daging sapi segar perlu berpindah dari laboratorium ke tempat pengolahan atau sistem pascapanen, termasuk rumah potong hewan, pasar tradisional dan supermarket, dan konsumen. Oleh karena itu, kebutuhan akan prediksi cepat di tempat tentang kualitas dan keamanan daging sapi di tempat pemrosesan adalah yang terpenting. Tujuan utama dari penelitian ini adalah untuk mengembangkan model prediksi secara cepat untuk mendeteksi perubahan kualitas daging sapi segar menggunakan portabel NIR spektrometer Sebanyak 70 sampel daging sapi segar dibeli dari pasar tradisional di Bogor. Data spektral diukur menggunakan spektrometer Scio Vis-NIR portabel dengan rentang panjang gelombang 740 hingga 1070 nm. Dua pertiga dari spektrum digunakan untuk kalibrasi dan sepertiga untuk validasi. Partial Least Square Regression (PLSR) dikombinasikan dengan cross-validation digunakan dengan spektrum asli dan pretreated untuk mengembangkan model prediksi. Hasil penelitian menunjukkan bahwa pH daging sapi segar menurun dari 5,68 menjadi 5,31 pada lama penyimpanan meningkat pada suhu ruang rata-rata 29 oC. Kontaminasi bakteri diamati sebagai jumlah piring total (TPC) dalam daging sapi segar meningkat dari 7,0 x 102 menjadi 1,12 x 106 cfug-1. TPC melebihi batas maksimum TPC yang dapat diterima, 1.0 x 106 cfug-1 yang ditetapkan oleh Badan Standar Nasional Indonesia SNI No. 3932: 2008. Model PLS terbaik untuk memprediksi pH dikembangkan dari spektrum yang telah diolah menggunakan turunan pertama dengan R2p = 0,71, SEP = 0,11% dan RPD = 1,95. Sedangkan untuk TPC model PLS terbaik dikembangkan dari spektrum asli (absorbansi) dengan R2p = 0,48, SEP = 0,62% dan RPD = 1,64. Selanjutnya, lima kualitas dan status keamanan daging sapi segar diidentifikasi dalam penelitian ini, mulai dari segar (0-2 jam), kurang segar (3-4 jam), sedikit busuk (5- 6 jam), busuk (7-8 jam) dan kering keras gelap (9 jam). Kinerja prediktif spektrometer Vis-NIR portabel Scio gelombang pendek menunjukkan prospek untuk penilaian kualitas dan keamanan daging sapi yang lebih baik di lokasi. Penelitian lebih lanjut harus dilakukan untuk mengetahui jenis bakteri individu yang ada di TPC untuk mengembangkan model prediksi yang spesifik.

The postharvest process of fresh beef begins immediately after the beef is slaughtered. Postharvest handling affects the quality and safety of carcasses through the post-harvest supply chain starting from slaughterhouses, distribution, processing and consumption. Postharvest handling of beef affects physicochemical properties such as pH and color, and microbial growth. Assessment of the quality of fresh beef needs to move from the laboratory to the processing site or postharvest system, including abattoirs, traditional markets and supermarkets, and consumers. Therefore, the need for rapid on-site predictions of beef quality and safety at the processing site is paramount. The main objective of this study was to develop a predictive model to rapidly detect changes in the quality of fresh beef using a portable NIR spectrometer. A total of 70 samples of fresh beef were purchased from a traditional market in Bogor. Spectral data were measured using a portable Scio Vis-NIR spectrometer with a wavelength range of 740 to 1070 nm. Two thirds of the spectra were used for calibration and one third for validation. Partial Least Square Regression (PLSR) combined with cross-validation was used with the original and pretreated spectra to develop predictive models. The results showed that the pH of fresh beef decreased from 5.68 to 5.31 at increased storage time at room temperature with an average of 29 oC. Bacterial contamination was observed as the total plate count (TPC) in fresh beef which increased from 7.0 x 102 to 1.12 x 106 cfug-1. TPC exceeds the maximum acceptable TPC limit, 1.0 x 106 cfug-1 set by the Indonesian National Standards Agency SNI No. 3932:2008. The best PLS model for predicting pH was developed from the processed spectrum using the first derivative with R2p = 0.71, SEP = 0.11% and RPD = 1.95. As for the TPC, the best PLS model was developed from the original spectrum (absorbance) with R2p = 0.48, SEP = 0.62% and RPD = 1.64. Furthermore, five qualities and safety status of fresh beef were identified in this study, ranging from fresh (0-2 hours), less fresh (3-4 hours), slightly rotten (5-6 hours), rotten (7-8 hours) and dry hard dark (9 hours). The predictive performance of the short-wavelength portable Scio Vis-NIR spectrometer shows prospects for better on-site beef quality and safety assessment. Further research should be carried out to determine the individual bacterial types present in TPC so as to develop specific predictive models.