Huili Xia

Bacterial infections remain a serious global health threat, exacerbated by the widespread use of antibiotics leading to bacterial resistance. Faced with the continuous evolution and accelerated spread of drug-resistant bacteria, there is an urgent need to establish an efficient and automated system to accelerate bacterial resistance identification and combination therapy screening in clinical practice. Here, we present an Automated Real-time Antimicrobial Susceptibility Testing (AR-AST) system, which represents an innovation in the AST field by enabling continuous, high-resolution monitoring of microbial growth. This system demonstrated exceptional adaptability, stability, and robustness in monitoring the growth kinetics of six common pathogenic bacteria. Through the establishment of a combined interpretation system of “early bacterial growth kinetics + PPT threshold + slope verification”, it effectively reduces the risk of misclassification. This approach achieved a reduction in the time required for resistance identification of clinical Acinetobacter baumannii 186 and Escherichia coli 15,017 by approximately 75.54% (95% CI: 64.54% − 86.55%) in a hospital setting. Furthermore, when applied to screen pairwise combinations of polymyxin B, polymyxin E, and kanamycin against resistant E. coli1 5017, a significant synergistic effect was observed for the kanamycin and polymyxin B combination (FIC index = 0.42 < 0.5). The complete inhibition rate in the binary combination was 100% (95% CI: 47.86% − 100%), with a time saving of approximately 68.03% (95% CI: 53.20% − 82.86%). Additionally, the AR-AST System successfully facilitated the quantitative analysis of E. coli 15017, exhibiting a broad detection range of six orders of magnitude (10¹ to 10⁶ CFU/mL), high accuracy (|Δ Log₁₀| < ± 0.5), and good stability ( P > 0.05), thereby providing effective auxiliary validation for evaluating combination therapy efficacy.