Single-shot self-supervised object detection in microscopy

• Published in: Nature communications Vol. 13; no. 1; pp. 7492 - 13
• Main Authors: Midtvedt, Benjamin, Pineda, Jesús, Skärberg, Fredrik, Olsén, Erik, Bachimanchi, Harshith, Wesén, Emelie, Esbjörner, Elin K., Selander, Erik, Höök, Fredrik, Midtvedt, Daniel, Volpe, Giovanni
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 05.12.2022; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114/1564; 639/766/930/328; Accuracy; Algorithms; Datasets; Deep learning; Experimental data; Fysik; Holography; Humanities and Social Sciences; Learning algorithms; Localization; Machine Learning; Microscopy; multidisciplinary; Object recognition; Physical Sciences; Pixels; Polarizability; Position measurement; Science; Science (multidisciplinary); Training; Translation; Translations; Vertical orientation; Vertical polarization
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-022-35004-y

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. Object detection using machine learning universally requires vast amounts of training datasets. Midtvedt et al. proposes a deep-learning method that enables detecting microscopic objects with sub-pixel accuracy from a single unlabeled image by exploiting the roto-translational symmetries of the problem.