KIT | KIT-Bibliothek | Impressum | Datenschutz

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

Friederich, Nils ORCID iD icon 1,2; Yamachui Sitcheu, Angelo Jovin ORCID iD icon 1; Neumann, Oliver 1; Eroglu-Kayıkçı, Süheyla 2; Prizak, Roshan 2; Hilbert, Lennart ORCID iD icon 3; Mikut, Ralf ORCID iD icon 1
1 Institut für Automation und angewandte Informatik (IAI), Karlsruher Institut für Technologie (KIT)
2 Institut für Biologische und Chemische Systeme (IBCS), Karlsruher Institut für Technologie (KIT)
3 Zoologisches Institut (ZOO), Karlsruher Institut für Technologie (KIT)

Abstract (englisch):

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.


Postprint §
DOI: 10.5445/IR/1000165020
Veröffentlicht am 29.11.2023
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Automation und angewandte Informatik (IAI)
Institut für Biologische und Chemische Systeme (IBCS)
Publikationstyp Proceedingsbeitrag
Publikationsmonat/-jahr 11.2023
Sprache Deutsch
Identifikator ISBN: 978-3-7315-1324-7
KITopen-ID: 1000165020
HGF-Programm 47.14.02 (POF IV, LK 01) Information Storage and Processing in the Cell Nucleus
Erschienen in Proceedings-33. Workshop Computational Intelligence: Berlin, 23.-24. November 2023
Veranstaltung 33th Workshop Computational Intelligence (2023), Berlin, 22.11.2023 – 24.11.2023
Auflage 1
Verlag KIT Scientific Publishing
Seiten 31-51
Vorab online veröffentlicht am 05.10.2023
Schlagwörter MLOps, Active Learning, Dynamic Processes, Microscope control
Nachgewiesen in Dimensions
Relationen in KITopen
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page