A new imaged-based quantitative reader for the gold immunochromatographic assay

Gold immunochromatographic assay (GICA) is widely used for quantitative measuring the concentration of the target analyte in samples. In this paper, a new low-cost quantitative reader is developed for the GICA based on the image processing technology. The image of the GICA strip is captured by the established quantitative reader, which is mainly composed by the microcontroller, image-senor and light source. Then, an automatic image processing framework, which includes image de-noising, extraction of the detection region, test and control lines segmentation and feature calculation, is carried out in the STM32F407ZGT6 microcontroller. Especially, the fuzzy cellular neural network with adaptive threshold (FCNN-AT) algorithm is presented in this paper for segmenting the test and control lines with high accuracy. Several indices are proposed to evaluate the proposed method. Results demonstrate that the presented quantitative reader based on the FCNN-AT algorithm is an effective and practical method for quantitative analysis of GICA.