AI-based automated active learning for discovery of hidden dynamic processes: A use case in light microscopy

TL;DR

This paper introduces AI-driven methods for planning and executing efficient data acquisition in biomedical experiments involving dynamic processes, utilizing image-based predictions and an MLOps pipeline to optimize experimental scheduling.

Contribution

It presents two novel methods, EDP and EAPDP, that leverage AI and MLOps to improve the discovery and analysis of dynamic processes in biomedical light microscopy.

Findings

Pre-trained CPN reliably extracts objects for EDP.

EAPDP effectively schedules data acquisition in experiments.

Methods enhance efficiency of dynamic process discovery.

Abstract

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.