Optical Sensor for Water Bacteria Detection using Machine Learning

The proposed method focuses on the detection of Escherichia coli bacterial contamination in water supplies. The proposed method uses a photonic-based sensor to examine photonic crystals in order to identify bacterial contamination in water. The system accurately predicts and diagnoses water contamination caused by bacteria by integrating an optical sensor with a graphical user interface (GUI). By attentively scrutinizing the early sensor-generated output graphs, we are able to discover variations in drainage patterns that are signaling the presence of bacteria in water samples. These graphs offer crucial insights into the unique characteristics of many samples, such as their refractive index. The sensor's sensitivity, specificity, and accuracy have all been thoroughly evaluated, and its results have been compared to those of established reference methods. A 95% accuracy rate is attained in the effective detection of the bacterium. These comparisons demonstrate the effectiveness of our AI-driven model and its intuitive graphical user interface (GUI), while also highlighting the potential of our machine learning techniques, such as Decision tree and Naive Bayes algorithm, for effective water bacterial detection.