Notice: Function _load_textdomain_just_in_time was called incorrectly. Translation loading for the advanced-post-block domain was triggered too early. This is usually an indicator for some code in the plugin or theme running too early. Translations should be loaded at the init action or later. Please see Debugging in WordPress for more information. (This message was added in version 6.7.0.) in /var/www/wordpress/wp-includes/functions.php on line 6114

Notice: Function _load_textdomain_just_in_time was called incorrectly. Translation loading for the lazy-blocks domain was triggered too early. This is usually an indicator for some code in the plugin or theme running too early. Translations should be loaded at the init action or later. Please see Debugging in WordPress for more information. (This message was added in version 6.7.0.) in /var/www/wordpress/wp-includes/functions.php on line 6114
Rapid detection of microbial antibiotic susceptibility via deep learning supported analysis of angle-resolved scattered-light images of picoliter droplet cultivations – Applied Systems Biology – HKI Jena

Rapid detection of microbial antibiotic susceptibility via deep learning supported analysis of angle-resolved scattered-light images of picoliter droplet cultivations

Sarkar A*, Graf M*, Svensson CM, Munser A-S, Schröder S, Hengoju S, Rosenbaum M, Figge MT

Abstract

The progressive increase in microbial resistance to antibiotics is a global health threat that requires solutions for rapid and reliable determination of antibiotic susceptibility in order to select appropriate antibiotics and dosages prior to treatment. We have established a screening platform that enables the detection of cell growth after just a few cell divisions. Our methodological approach for a robust phenotypic antibiotic susceptibility testing is based on the innovative combination of three cutting-edge technologies: (i) a high-throughput microfluidic platform where individual bacterial cells are encapsulated in picoliter-sized droplets, (ii) a 2D angle-resolved light scattering sensor to perform label-free hourly screening of the droplets, and (iii) a computational image analysis approach based on convolutional neural networks to evaluate the dynamics of microbial growth in droplets. For the gram-positive Staphylococcus aureus and gram-negative Escherichia coli, we demonstrate that microbial growth in droplets can be successfully detected within one to two hours. Furthermore, the potential of this platform for rapid phenotypic antibiotic susceptibility testing is demonstrated as a proof-of-concept with the clinically relevant bacterium S. aureus under various concentrations of the antibiotic tetracycline. Notably, we reach a robust phenotypic decision regarding the sensitivity to this antibiotic within two hours.

Related projects

12319 Next