Missing Wedge Completion via Unsupervised Learning with Coordinate Networks
International Journal of Molecular Sciences · 2024
DOI: 10.3390/ijms25105473 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Molecular Sciences, 2024. Verfasst von Dave Van Veen und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA. Hauptthemen laut MEDLINE: Unsupervised Machine Learning, Neural Networks, Computer, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 105 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIAID NIH HHS, Chan Zuckerberg Initiative (United States), Silicon Valley Community Foundation Chan-Zuckerberg Imaging Initiative.
Abstract (Originalsprache)
Cryogenic electron tomography (cryoET) is a powerful tool in structural biology, enabling detailed 3D imaging of biological specimens at a resolution of nanometers. Despite its potential, cryoET faces challenges such as the missing wedge problem, which limits reconstruction quality due to incomplete data collection angles. Recently, supervised deep learning methods leveraging convolutional neural networks (CNNs) have considerably addressed this issue; however, their pretraining requirements render them susceptible to inaccuracies and artifacts, particularly when representative training data is scarce. To overcome these limitations, we introduce a proof-of-concept unsupervised learning approach using coordinate networks (CNs) that optimizes network weights directly against input projections. This eliminates the need for pretraining, reducing reconstruction runtime by 3–20× compared to supervised methods. Our in silico results show improved shape completion and reduction of missing wedge artifacts, assessed through several voxel-based image quality metrics in real space and a novel directional Fourier Shell Correlation (FSC) metric. Our study illuminates benefits and considerations of both supervised and unsupervised approaches, guiding the development of improved reconstruction strategies.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
- Department of Bioengineering, Stanford University, Stanford, CA 94305, USA.
- Department of Electrical and Computer Engineering, University of Michigan, Ann Arbor, MI 48109, USA.
- Department of Biochemistry and Molecular Pharmacology, Baylor College of Medicine, Houston, TX 77030, USA.
- Department of Genetics, The Salk Institute of Biological Sciences, La Jolla, CA 92037, USA.
- Department of Radiology, Stanford University, Stanford, CA 94305, USA.
ORCID-Profile: Dave Van Veen, Jesús G Galaz-Montoya, Akshay S Chaudhari, Michael F Schmid, Wah Chiu
Förderung
- NIAID NIH HHS
- Chan Zuckerberg Initiative (United States)
- Silicon Valley Community Foundation Chan-Zuckerberg Imaging Initiative
Zitieren
APA
Van Veen, D., Galaz-Montoya, J. G., Shen, L., Baldwin, P., Chaudhari, A. S., Lyumkis, D., Schmid, M. F., Chiu, W., & Pauly, J. (2024). Missing Wedge Completion via Unsupervised Learning with Coordinate Networks. International Journal of Molecular Sciences. https://doi.org/10.3390/ijms25105473
Vancouver
Van Veen D, Galaz-Montoya JG, Shen L, Baldwin P, Chaudhari AS, Lyumkis D, et al. Missing Wedge Completion via Unsupervised Learning with Coordinate Networks. International Journal of Molecular Sciences. 2024. doi:10.3390/ijms25105473
BibTeX
@article{VanVeen2024,
author = {Van Veen, Dave and Galaz-Montoya, Jesús G and Shen, Liyue and Baldwin, Philip and Chaudhari, Akshay S and Lyumkis, Dmitry and Schmid, Michael F and Chiu, Wah and Pauly, John},
title = {Missing Wedge Completion via Unsupervised Learning with Coordinate Networks},
journal = {International Journal of Molecular Sciences},
year = {2024},
doi = {10.3390/ijms25105473}
}
EUVIMED