Projects per year
Abstract
Wave scattering is a challenging numerical problem, yet it is central to fields as diverse as seismology, fluid dynamics, acoustics, and photonics. Complex structures scatter waves in random yet deterministic ways. Advances in our understanding and control of scattering are key to applications such as deeptissue microscopy. However, computing the internal fields on a scale relevant to microscopy remains excessively demanding for both conventional methods and physicsbased neural networks. Here, we show how coherent scattering calculations can be scaled up to 21 × 106 cubic wavelengths by mapping the physics of multiple scattering onto a deterministic neural network that efficiently harnesses publicly available machine learning infrastructure. We refer to this as a scattering network. Memory usage, an important bottleneck to scaling beyond (10 μm)³, is kept to a minimum by the recurrent network topology and the convolutional derivatives it embodies. Tight integration with an opensource electromagnetic solver enables any researcher with an internet connection to compute complex lightwave scattering throughout volumes as large as (130 μm)³ or 25 mm2.
Original language  English 

Article number  0098 
Number of pages  10 
Journal  Intelligent Computing 
Volume  3 
DOIs  
Publication status  Published  5 Aug 2024 
Fingerprint
Dive into the research topics of 'Scaling Up Wave Calculations with a Scattering Network'. Together they form a unique fingerprint.
Correlative Refractive Index LightSheet Microscopy
Vettenburg, T. (Investigator)
1/01/20 → 1/07/27
Project: Research

DTP 20182019 Training Grant
Rowan, J. (Investigator)
Engineering and Physical Sciences Research Council
1/10/18 → 30/09/23
Project: Research
Datasets

Datasets used in Scaling up wave calculations with a Scattering Network
Valantinas, L. (Creator) & Vettenburg, T. (Creator), University of Dundee, Apr 2024
DOI: 10.15132/10000251, https://dmail.sharepoint.com/:f:/s/ResearchServicesPublicDocuments/EijcgAlOXtZOnynxiz7TMWUBx8F8yLhPwd8vOEUNKpgCfg?e=7HrIg3
Dataset