Deep Learning-Enabled Detection and Classification of Bacterial Colonies Using a Thin-Film Transistor (TFT) Image Sensor

Early detection and identification of pathogenic bacteria such as Escherichia coli (E. coli) is an essential task for public health. The conventional culture-based methods bacterial colony usually take >24 hours to get the final read-out. Here, we demonstrate a colony-forming-unit (CFU) system exploiting thin-film-transistor (TFT)-based image sensor array that saves ~12 compared Environmental Protection Agency (EPA)-approved methods. To efficacy this CFU system, lensfree imaging modality was built using TFT with sample field-of-view ~10 cm^2. Time-lapse images colonies cultured on chromogenic agar plates were automatically collected at 5-minute intervals. Two deep neural networks used detect count growing identify their species. When blindly tested 265 E. other coliform (i.e., Citrobacter Klebsiella pneumoniae), our reached average rate 97.3% 9 incubation recovery 91.6% hours. This TFT-based can be applied various microbiological Due large scalability, ultra-large field-of-view, low cost sensors, platform integrated each plate disposed after automated count. cost-effectively increased >100 cm^2 provide massive throughput using, e.g., roll-to-roll manufacturing TFTs in flexible display industry.