New Developments in Open-Source Software for Inverse Problems
|
Organizer(s): |
Name:
|
Affiliation:
|
Country:
|
Evangelos Papoutsellis
|
Finden Ltd, University of Manchester
|
England
|
Ander Biguri
|
University of Cambridge
|
England
|
|
|
|
|
|
|
|
|
Introduction:
| | Open-source software has had a profound impact on both industry and academia over the past 50 years, revolutionising how we develop, share, and collaborate on software solutions. Within the sphere of inverse problems, it has enabled access to many modalities of practical importance— including medical, chemical, and seismic imaging; materials science; and more. This special session aims to showcase new developments in open-source software for inverse problems and to foster collaboration across disciplines. We will highlight theory, algorithms, and applications that advance reproducible imaging pipelines—from regularisation techniques and variational models to large-scale optimisation and hybrid model-driven/data-driven approaches. Talks will demonstrate end-to-end, FAIR, and well-documented toolchains that enable trustworthy reconstruction and analysis across modalities such as CT, PET, and MRI; chemical and spectral imaging; seismic exploration; and materials characterisation. By bringing together software developers, applied mathematicians, and domain scientists, the session aims to lower barriers to entry, disseminate practical tools, address challenges of scalability and generalisation, and accelerate the translation of methods into robust, real-world imaging systems.
|
|
|
|
|
|
