PETNet– Coincident Particle Event Detection using Spiking Neural Networks
Jan Debus 1 , Charlotte Debus 2 , Günther Dissertori 1 , and Markus Götz 2 3 · 2025
Read PETNet– Coincident Particle Event Detection using Spiking Neural Networks by Jan Debus 1 , Charlotte Debus 2 , Günther Dissertori 1 , and Markus Götz 2 3 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-06730.txt.
Description
Jan Debus and co-authors present PETNet, a 2025 method that uses spiking neural networks to find photon pairs in medical scan data. Positron emission tomography traces a radioactive marker inside the body. Pairing up photons from the same decay, while dropping stray hits, is a core cleanup step before an image forms.
Spiking networks process sparse, timed events with little energy, a natural fit for detector hits read as spike trains. PETNet learns the pairing task under supervision, with a loss shaped for the job and detector geometry built in. On simulated data it beat the best classic algorithm, reaching an F1 score of 95.2 percent. It also called pairs up to 36 times faster. Event-based computing may suit particle physics well beyond this scan.
Catalog record
| CREATOR | Jan Debus 1 , Charlotte Debus 2 , Günther Dissertori 1 , and Markus Götz 2 3 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-06730