Artificial intelligence assisted photonic bio sensing for rapid bacterial diseases

Abstract Combining artificial intelligence (AI) and photonic biosensors is a new method of high-accuracy bacterial detection. In the present work, a decision tree classifier is used, aimed at the classification of bacterial species by taking readings from the wavelength measurements extracted from photonic sensor simulations performed using Rsoft. The data set is processed through univariate analysis, Kernel density estimation (KDE) and box plot evaluation, and optimized feature selection as well as outlier removal. The classifier is trained with a 70.27 % classification accuracy. Performance evaluation using a confusion matrix highlighted the classification efficiency. The obtained findings show the promise of AI based photonic bio sensing for the bacterial infectious diseases.