<p class="MsoNormal">In this paper, the concentration and color of a sample of water were determined using machine learning and linear regression principles paired with an optical biosensor that has the ability to detect the amount of red, blue, and green pigment in a sample. The setup of the optical biosensor consisted of a small, 3D-printed cavity. A three-color LED, and a photoresistor sensor are integrated into the cavity. The LED and the photoresistor are situated so that the light emitted from the LED passes through the cuvette that contains the sample. The transmitted light is detected by the photoresistor. A wireless-enabled microcontroller (ESP32) is used to control the LED’s light color and intensity and read the photoresistor’s signal. The microcontroller allows blue, green, and red light to be flashed momentarily into a cuvette of a sample which would then result in the concentration of each being displayed on the interface of the circuit made to perform this function. At this stage of the presented research, this data would then be sent to a computer via a micro-USB cord; however, the work to use the wireless capabilities of the microcontroller is in progress. Using the data, the color of the sample and its concentration are successfully detected using a two-stage AI/ML architecture. First, a classification artificial neural network is trained to detect the color of the sample. Then, knowing the color of the sample, an appropriately trained ML model is selected to determine the concentration of the sample from the red-blue-green values. To perform an evaluation of color and concentration, a neural network built-in MATLAB was trained using five sets of colors, each in color concentration intervals of eighteen percent of their respective color, resulting in six different concentration samples for each of the five colors. The status of each sample was used not only to train the neural network but also to create a linear model of the concentration with respect to the amount of red, blue, and green pigment in the sample. After entering the status of each sample enough times to reduce error in predictability, a MATLAB script was used to evaluate a sample of water and estimate the color, concentration of the color, and the confidence in the evaluation. Ideas explored in this project raise the possibility of machines more efficiently predicting other traits and concentrations of a fluid using machine learning and artificial intelligence with an optical biosensor.</p>

Design of a Cost-Effective Optical Biosensor Powered by Artificial Intelligence and Machine Leaning

Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2021/volume/85543

ASME 2021 International Mechanical Engineering Congress and Exposition (IMECE2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.