Microfluidic Live-Cell Imaging (MLCI) enables biotechnologists to better understand the spatiotemporal growth of microbes. In the Helmholtz Imaging Platform (HIP) project SATOMI, the deep learning-based image analysis tools microbeSEG [1] and ObiWan-Microbi [2] were developed to perform offline analysis of high-quality data. These tools mainly focused on the segmentation and tracking challenges of growing colonies. One important aspect noticed afterwards was the lack of continual acquisition of high-quality data during the experiments that spanned days and weeks.

The EMSIG project addresses this problem by bringing live event detection capacities to MLCI. The goal is to develop an intelligent and adaptable event-based detection and control system that autonomously identifies predefined biological events such as cell growth, cell division or technical events such as image quality deterioration over time and triggers appropriate real-time responses. Such responses include for example refocusing or increasing spatiotemporal resolution to observe specific events.

The challenge in EMSIG (Event-driven Microscopy for Smart mIcrofluidic sinGle-cell analysis) lies in achieving robust and precise real-time classification of event triggers at the intersection of imaging and stochastic biological components. If successful, this opens the door to entirely new insights for characterizing rapid processes, advancing the level of automation of live-cell microscopy, and elevating the Technology Readiness Level (TRL) of MLCI.

[1] microbeSEG: A deep learning software tool with OMERO data management for efficient and accurate cell segmentation
[2] ObiWan-Microbi and microbeSEG: Deep Learning Tools for Microbial Image Analysis

Publications: