Description
Laser-driven ion beams offer intrinsically ultrashort bunches and extreme instantaneous dose rates that are attractive for multidisciplinary applications spanning radiobiology, dosimetry, radiation resilience and imaging. However, translating these sources into reliable ’user beams‘ remains challenging because key beam properties (e.g. spectral shape, cut-off energy, divergence and dose) are commonly inferred from diagnostics that are invasive, rate-limiting, or otherwise incompatible with application delivery, while the source itself is highly sensitive to shot-to-shot fluctuations in laser and target conditions. Advances in machine learning (ML) techniques can now directly address some of these challenges.
In this talk, we will present a coordinated programme that combines (i) real-time source optimisation at SCAPA (the Scottish Centre for the Applications of Plasma-based Accelerators) with (ii) ML-based synthetic diagnostics, providing a practical route towards stabilised, application-ready operation. At SCAPA we have implemented automated parameter scans and multi-dimensional, ML-driven optimisation on a 350 TW, Hz-rate Ti:Sapphire proton platform [1]. A key enabler is ARISE [2], a rapid Thomson-parabola spectrum-extraction pipeline designed for high-throughput analysis that is integrated into Bayesian optimisation loops, enabling autonomous optimisation of proton-beam performance during operation.
Building on our recent demonstration of a deep neural-network ‘synthetic diagnostic’ that predicts the full proton energy spectrum (with uncertainty) from non-disruptive secondary measurements and shot metadata—removing the need for direct spectral measurement on every shot [3] —we will present progress towards a more complete surrogate diagnostic suite. Using data from a dedicated three-week campaign on the ELIMAIA–ELIMED user beamline at ELI-Beamlines, we are training improved predictors and extending the approach to additional beam metrics such as divergence. The resulting framework aims to provide accurate, real-time characterisation without disrupting beam delivery, while identifying the laser/target parameters that must be stabilised to achieve reproducible source performance for end users.
[1] R. Wilson et al., In preparation (2026).
[2] B. C. Torrance et al., High Power Laser Science and Engineering, 13, e105 (2025).
[3] C. J. McQueen et al., Communications Physics 8, 66 (2025).
