Reproduce “Deep convolutional framelets for dose reconstruction in BNCT with Compton camera detector” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Angelo Didonna, Dayron Ramos Lopez, Giuseppe Iaselli, Nicola Amoroso, Nicola Ferrara, Gabriella Maria Incoronata Pugliese
The authors use neural networks to refine early Compton-camera reconstructions for simulated boron neutron capture therapy (BNCT). The network takes a 10th-iteration list-mode maximum-likelihood expectation-maximization image and predicts the 60th-iteration result; a tight-frame U-Net uses Haar-wavelet pathways. On 2,520 test images, it outperformed alternatives, while total reconstruction took 4–6 minutes rather than 24–36, in simulation only.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.