Single-shot self-supervised object detection in microscopy
Artikel i vetenskaplig tidskrift, 2022

Object detection is a fundamental task in digital microscopy, where machine learning has made great strides in overcoming the limitations of classical approaches. The training of state-of-the-art machine-learning methods almost universally relies on vast amounts of labeled experimental data or the ability to numerically simulate realistic datasets. However, experimental data are often challenging to label and cannot be easily reproduced numerically. Here, we propose a deep-learning method, named LodeSTAR (Localization and detection from Symmetries, Translations And Rotations), that learns to detect microscopic objects with sub-pixel accuracy from a single unlabeled experimental image by exploiting the inherent roto-translational symmetries of this task. We demonstrate that LodeSTAR outperforms traditional methods in terms of accuracy, also when analyzing challenging experimental data containing densely packed cells or noisy backgrounds. Furthermore, by exploiting additional symmetries we show that LodeSTAR can measure other properties, e.g., vertical position and polarizability in holographic microscopy.

Författare

Benjamin Midtvedt

Göteborgs universitet

Jesús Pineda

Göteborgs universitet

Fredrik Skärberg

Göteborgs universitet

Erik Olsén

Chalmers, Fysik, Nano- och biofysik

Harshith Bachimanchi

Göteborgs universitet

Emelie Vilhelmsson Wesén

Chalmers, Biologi och bioteknik, Kemisk biologi

Elin Esbjörner Winters

Chalmers, Biologi och bioteknik, Kemisk biologi

Erik Selander

Göteborgs universitet

Fredrik Höök

Chalmers, Fysik, Nano- och biofysik

Daniel Midtvedt

Göteborgs universitet

Giovanni Volpe

Göteborgs universitet

Nature Communications

2041-1723 (ISSN) 20411723 (eISSN)

Vol. 13 1 7492

Ämneskategorier

Annan data- och informationsvetenskap

Robotteknik och automation

Datorseende och robotik (autonoma system)

DOI

10.1038/s41467-022-35004-y

PubMed

36470883

Mer information

Senast uppdaterat

2023-10-27