AI-based automated active learning for discovery of hidden dynamic processes: A use case in light microscopy
This addresses the challenge of planning and executing high-throughput experiments in biomedical research, though it appears incremental as it builds on existing methods like CPN.
The paper tackles the problem of efficiently acquiring dynamic processes in biomedical light microscopy experiments by introducing two AI-based methods: EDP for predicting pseudo-time from single images and EAPDP for automating acquisition scheduling, with results showing reliable integration of a SOTA segmentation method.
In the biomedical environment, experiments assessing dynamic processes are primarily performed by a human acquisition supervisor. Contemporary implementations of such experiments frequently aim to acquire a maximum number of relevant events from sometimes several hundred parallel, non-synchronous processes. Since in some high-throughput experiments, only one or a few instances of a given process can be observed simultaneously, a strategy for planning and executing an efficient acquisition paradigm is essential. To address this problem, we present two new methods in this paper. The first method, Encoded Dynamic Process (EDP), is Artificial Intelligence (AI)-based and represents dynamic processes so as to allow prediction of pseudo-time values from single still images. Second, with Experiment Automation Pipeline for Dynamic Processes (EAPDP), we present a Machine Learning Operations (MLOps)-based pipeline that uses the extracted knowledge from EDP to efficiently schedule acquisition in biomedical experiments for dynamic processes in practice. In a first experiment, we show that the pre-trained State-Of-The- Art (SOTA) object segmentation method Contour Proposal Networks (CPN) works reliably as a module of EAPDP to extract the relevant object for EDP from the acquired three-dimensional image stack.