Fig. 1: Multi-bacteria classification results. | npj Biosensing

Fig. 1: Multi-bacteria classification results.

From: Realtime bacteria detection and analysis in sterile liquid products using deep learning holographic imaging

Fig. 1

a A sample of experimentally derived synthetic hologram showing our deep learning model detects and classifies six types of particles each marked with colored bounding boxes including E. coli (EC), P. aeruginosa (PA), B. subtilis (BS), E. faecalis (EF), C. jejuni (CJ), and generic particulate matters (PMs). The figure also includes an in-focused closeup hologram of each type of particle showing their distinct morphologies and diffraction patterns with a scale bar of 5 µm. b Receiver operator characteristic (ROC) curves and corresponding area under curve (AUC) values of our deep learning model for different types of particles with vertical dashed line marking 0.1% false positive rate (FPR). c The confusion matrix shows the accuracy and prediction errors of our classification for each type of particle evaluated at 0.1% FPR.

Back to article page