This study aimed to investigate the epidemiological characteristics and trends over time of carbapenemase-producing (e.g., KPC, NDM, VIM, IMP, and OXA-48) Gram-negative bacteria (CPGNB). Non-duplicated multi-drug resistant Gram-negative bacteria (MDRGNB) were collected from the First Affiliated Hospital of Zhengzhou University from April 2019 to February 2023. Species identification of each isolate was performed using the Vitek2 system and confirmed by matrix-assisted laser desorption ionization-time of flight mass spectrometry according to the manufacturer\'s instructions. PCR detected carbapenem resistance genes in the strains, strains carrying carbapenem resistance genes were categorized as CPGNB strains after validation by carbapenem inactivation assay. A total of 5705 non-repetitive MDRGNB isolates belonging to 78 different species were collected during the study period, of which 1918 CPGNB were validated, with the respiratory tract being the primary source of specimens. Epidemiologic statistics showed a significant predominance of ICU-sourced strains compared to other departments. Klebsiella pneumoniae, Escherichia coli, Acinetobacter baumannii, and Pseudomonas aeruginosa were the significant CPGNB in Henan, and KPC and NDM were the predominant carbapenemases. Carbapenem-resistant infections in Henan Province showed an overall increasing trend, and the carriage of carbapenemase genes by CPGNB has become increasingly prevalent and complicated. The growing prevalence of CPGNB in the post-pandemic era poses a significant challenge to public safety.

A rapid and accurate detection of carbapenemase-producing Gram-negative bacteria (CPGNB) has an immediate demand in the clinic. Here, we developed and validated a method for rapid detection of CPGNB using Blue-Carba combined with deep learning (designated as AI-Blue-Carba). The optimum bacterial suspension concentration and detection wavelength were determined using a Multimode Plate Reader and integrated with deep learning modeling. We examined 160 carbapenemase-producing and non-carbapenemase-producing bacteria using the Blue-Carba test and a series of time and optical density values were obtained to build and validate the machine models. Subsequently, a simplified model was re-evaluated by descending the dataset from 13 time points to 2 time points. The best suitable bacterial concentration was determined to be 1.5 optical density (OD) and the optimum detection wavelength for AI-Blue-Carba was set as 615 nm. Among the 2 models (LRM and LSTM), the LSTM model generated the higher ROC-AUC value. Moreover, the simplified LSTM model trained by short time points (0-15 min) did not impair the accuracy of LSTM model. Compared with the traditional Blue-Carba, the AI-Blue-Carba method has a sensitivity of 95.3% and a specificity of 95.7% at 15 min, which is a rapid and accurate method to detect CPGNB.