Identification of Diffusive States in Tracking Applications Using Unsupervised Deep Learning Methods

Research output: Chapter/Conference proceedingConference proceedingAcademicpeer-review

2 Citations (Scopus)
43 Downloads (Pure)

Abstract

The most widely used method for analysis of diffusive motion in particle tracking is based on estimation of the mean squared displacement (MSD) and subsequently relevant motion parameters. This approach is only valid for a population of particles exhibiting a single type of motion (e.g., super or sub-diffusive). Thus, to deal with trajectories that describe dynamics with switching motion patterns, trajectory segmentation techniques are of major importance. Here, we propose an unsupervised trajectory segmentation technique, which employs the ideas of the state-of-the-art image denoising "noise2noise" approach. Using typical single-particle tracking data, our method is capable of unsupervised trajectory segmentation in the most difficult situations (e.g. unknown number of purely diffusive states), and computation of the relevant parameters. The applicability of the method is demonstrated using simulated and real experimental data, showing that its performance is comparable to similar top performing supervised methods.
Original languageEnglish
Title of host publication2022 19th International Symposium on Biomedical Imaging (ISBI)
Number of pages4
ISBN (Electronic)9781665429238
DOIs
Publication statusPublished - 26 Apr 2022
Event2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI) - Kolkata, India
Duration: 28 Mar 202231 Mar 2022
Conference number: 19th

Conference

Conference2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI)
Abbreviated titleIEEE ISBI
Country/TerritoryIndia
CityKolkata
Period28/03/2231/03/22

Bibliographical note

Funding Information:
Acknowledgments. Special thanks to H. Zijlstra, M. W. Paul and C. Wyman for sharing the BRCA2 SPT image data. This work was supported by the Dutch Research Council through the NWO-BBOL research program (Grant No. 737.016.014). Compliance with Ethical Standards. This is a numerical simulation study for which no ethical approval was required.

Publisher Copyright:
© 2022 IEEE.

Fingerprint

Dive into the research topics of 'Identification of Diffusive States in Tracking Applications Using Unsupervised Deep Learning Methods'. Together they form a unique fingerprint.

Cite this