The development of image base, portable microfluidic paper-based analytical device

Microfluidic paper-based analytical devices can be used in preliminary screening and diagnosis of specific diseases. It has the advantages of low costs and quick diagnosis. After chemical reaction, the paper chip will display the disease´s relative status through color identification or electrochemistry testing. When compared to electrochemistry testing, the non-contact color identification method has the advantages of low costs, reusability and not requiring cleaning. However since research in the past have encountered many problems with the color identification method, this has led to the negation of the advantages of microfluidic paper-based analytical devices in quick testing. For this reason, this study developed a portable testing apparatus with image analysis as a foundation and uses a self-designed imaging processing calculation method to implement color identification. The hardware equipment designed for image processing in this study uses a micro embedded style system and imaging equipment as the hardware framework, along with a Web application to implement a multi-platform operating interface. Color quantization was used as the foundation for the image processing method, to calculate the paper chip´s color. Furthermore, this study also carried out a variety of color quantization methods to compare the color quantization results of the microfluidic paper-based analytical devices. Methods include uniform color quantization, median-cut algorithm, k-means clustering algorithm and self-organizing maps neural networks. During the experiment phase of this study, we found that median-cut was superior to other methods in application. This result is different from many other research works conducted on color quantization. Lastly, the experiment results showed the usability of this testing setup.