Reliable and Robust Model for Fast Identification and Detection of Counterfeited EBN Products by Using FTIR Microspectroscopy (talk)

Abstract

Edible birds’ nest (EBN) is a traditional Chinese cuisine and medicine made by cavedwelling birds called swiftlets; Aerodramus, Hydrochous, Schoutedenapus and Collocalia. It is a culinary delicacy, which can retail at over 4000 USD per kilogram. EBN has been scientifically proven to have nutritional properties; improvement in bone strength cell division, anti-aging and antiviral, to name a few. While EBN has been demonstrated to have many health benefits, recent studies have highlighted its health risks. High nitrite content was found in EBN collected inside caves. This increases the risk of cancer due to the increased production of carcinogenic nitrosamines. In 2013, to combat this, China imposed a trade ban on EBN with some South East Asian countries like Malaysia until healthier harvesting practices were adopted. Only a few Malaysian producers were able to meet requirement. The low supply drove up the prices of EBN. Some producers, in order to meet demand, allegedly used adulteration techniques to add impurities to increase the weight. Adulterated or fake EBN may be hazardous to the consumers. A total of 8 samples pieces of original EBN sourced from 5 different provinces in Vietnam, together with various materials such as tremella fungus, pork skin, karaya gum, fish swimming bladder, jelly, agar, monosodium glutamate and egg white used to adulterate EBN have been analyzed. The main goal of our work was to establish a reliable and robust model for fast identification and detection of counterfeited EBN products by using FTIR microspectroscopy in conjunction with R Platform for statistical computing. This model can be distributed later among Health Regulatory Authorities (HSA) and food industry laboratories.

Krzysztof Banas

Principal Research Fellow

I work as beam-line scientist at Singapore Synchrotron Light Source. My research interests include application of advanced statistical methods for hyperspectral data processing (dimension reduction, clustering and identification).