Designing a pre-analytical portable reader for on-site aflatoxin detection

Aflatoxins are a global public health problem as they are responsible for significant adverse health and economic issues affecting consumers and farmers worldwide. Aflatoxins, produced by fungi of the genus Aspergillus, are a group of toxic, mutagenic and carcinogenic fungal metabolites that routinely contaminate food and agricultural products. Instruments have been designed to detect and measure the presence of aflatoxins in foods by utilizing their chemical and structural properties. Aflatoxins are divided into 2 categories according to their response to UV-A light at a wavelength of 365nm. These are classified as blue fluorescence B1 and B2, and green fluorescence G1 and G2 [2]. It is used as a rapid method for contamination detection. Hyperspectral, QCM-based systems, machine learning, and similar methods are widely investigated in the literature. Fluorescence and FTIR spectroscopy are the methods used for aflatoxin detection. With the advancement of technological developments, different pre-analytical methods that are easy to use and provide rapid detection are being developed. In this study, spectral-based sensor systems for aflatoxin detection were investigated and their use as a pre-analytical on-site detection method was discussed. The mechanical parts of the system were designed in Autodesk FUSION 360 program and manufactured with a three-dimensional printer. Arduino UNO was used as the electronic controller of the reader system and AS7262 and ISL29125 sensors were used for fluorescence luminescence properties. The embedded software studies of the reader system were performed using the Arduino IDE platform C programming languages. The user interface and the control system on the computer were provided using the Python programming language. Different weights of the AFTOK008 contaminated pistachio sample obtained from the National Reference Food Laboratory were exposed to 365nm UV-A light and the fluorescence luminescence properties were observed on the UV-VIS Spectrometer (Ocean Insight, Flame), AS7262 (Sparkfun) and ISL29125 (Sparkfun) components. Spectrometer readings were used for validation of AS7262 (Sparkfun) and ISL29125 (Sparkfun) components. BGYF and blue fluorescence readings were provided in AS7262 sensor-based studies. A significant increase was observed between the green zone level control, 0.2, 0.3, 0.4 and 0.5 gr groups and fluorescence intensity. A significant increase was observed between the blue zone measurement level control, 0.1, 0.2, 0.3, 0.4 and 0.5 gr groups and fluorescence intensity. In ISL29125 sensor readings, the absence of green zone readings is not a positive result for the thesis. A significant increase was observed between the blue zone measurement level control, 0.1, 0.2, 0.3, 0.4 and 0.5 gr groups and fluorescence intensity. In spectrometer studies, it gave a maximum peak between 440-560 nm and 480-510 nm. As a result of this study, AS7262 sensor-based system gives more accurate results than ISL29125 based system. It was evaluated that pre-analytical portable reader design can be used as an on-site detection method for on-site aflatoxin detection.