A Portable UV-LED/RGB Sensor for Real-Time Bacteriological Water Quality Monitoring Using ML-Based MPN Estimation

Abstract

Bacteriological water quality monitoring is of utmost importance for safeguarding public health against waterborne diseases. Traditional methods such as membrane filtration (MF), multiple tube fermentation (MTF), and enzyme-based assays are effective in detecting fecal contamination indicators, but their time-consuming nature and reliance on specialized equipment and personnel pose significant limitations. This paper introduces a novel, portable, and cost-effective UV-LED/RGB water quality sensor that overcomes these challenges. The system is composed of a multi-well self-loading microfluidic device for sample-preparation-free analysis, RGB sensors for data acquisition, UV-LEDs for excitation, and a portable incubation system. Commercially available defined substrate technology, most probable number (MPN) analysis, and machine learning (ML) are combined for the real-time monitoring of bacteria colony-forming units (CFU) in a water sample. Fluorescence signals from individual wells are captured by the RGB sensors and analyzed using Multilayer Perceptron Neural Network (MLPNN) and Support Vector Machine (SVM) algorithms, which can quickly determine if individual wells will be positive or negative by the end of a 24 h period. The novel combination of ML and MPN analysis was shown to predict in 30 min the bacterial concentration of a water sample with a minimum prediction accuracy of 84%.

Keywords: defined substrate test (DST); machine learning (ML); most probable number (MPN); point-of-care; water quality (WQ).

Figure 1

(a) Microfluidic device configuration. (b) Dimensions and structure of acrylic sheets. (c) Fabricated microfluidic device. (d) Hydrostatic versus hydraulic resistance pressure analysis. (e) MPN table and CI of microfluidic device; red values indicate contamination beyond EPA BAV threshold.

Figure 2

(a) Case for Peltier cell and microfluidic device. (b) Temperature response with and without control system (CS). (c) RGB sensor with UV-LED to detect fluorescence in water samples. (d) Array of 8 UV-LED/RGB sensors with a black layer filter.

Figure 3

(a) The UV-LED/RGB system integrates a microfluidic device containing a water sample and transmits data monitored by the RGB sensors to a computer or smartphone, either through a wired or wireless connection. (b) Summary of results from 50 to 80 CFU pellet experimental test. (c) RGBC data from UV-LED/RGB system. (d) Microfluidic device exposed to UV light and fluorescence in wells 4, 6, and 7. The other wells (0–3 and 5) indicate negative for bacteria.

Figure 4

Machine learning flowchart: The left side depicts the MLPNN algorithm, while the right side outlines the SVM algorithm.

Figure 5

(a) UV-LED/RGB sensor number 4 “clear” color channel showing sigmoidal bacteria growth curves for different concentrations of bacteria in water samples. (b) Normalized RGB signals from the experiment measuring 37.6 CFU/100 mL. (c) First derivative curves from 37.6 CFU/100 mL experiment.

Figure 6

(a,b) Confusion matrix for each well of the UV-LED/RGB system after applying the SVM and MLPNN algorithm, respectively. (c,d) ROC curves with the AUC value for each SVM and MLPNN model used on the data acquired from each well. (e,f) Metrics of accuracy, precision, recall, and F1-score for the SVM and MLPNN models. (g,h) The metrics table for each well, along with the execution time for each ML algorithm.