Development of Convolutional Neural Networks for an Electron-Tracking Compton Camera

Abstract

Electron-tracking Compton camera, which is a complete Compton camera with tracking Compton scattering electron by a gas micro time projection chamber, is expected to open up MeV gamma-ray astronomy. The technical challenge for achieving several degrees of the point spread function is the precise determination of the electron-recoil direction and the scattering position from track images. We attempted to reconstruct these parameters using convolutional neural networks. Two network models were designed to predict the recoil direction and the scattering position. These models marked 41 degrees of the angular resolution and 2.1 mm of the position resolution for 75 keV electron simulation data in Argon-based gas at 2 atm pressure. In addition, the point spread function of ETCC was improved to 15 degrees from 22 degrees for experimental data of 662 keV gamma-ray source. These performances greatly surpassed that using the traditional analysis.

References in corpus (15)

• The BATSE Gamma-Ray Burst Spectral Catalog. I. High Time Resolution Spectroscopy of Bright Bursts using High Energy Resolution Data
• The first COMPTEL Source Catalogue
• Gamma-ray spectroscopy of Positron Annihilation in the Milky Way
• Observations of GRB 990123 by the Compton Gamma-Ray Observatory
• The BL Lac heart of Centaurus A
• Gamma-rays from Type Ia supernova SN2014J
• An Electron-Tracking Compton Telescope for a Survey of the Deep Universe by MeV gamma-rays
• Establishment of Imaging Spectroscopy of Nuclear Gamma-Rays based on Geometrical Optics
• Detection of the 511 keV Galactic positron annihilation line with COSI
• Spectral Energy Distributions of 3C 279 revisited: BeppoSAX observations and variability models
• New readout and data-acquisition system in an electron-tracking Compton camera for MeV gamma-ray astronomy (SMILE-II)
• SPI Measurements of Galactic 26Al
• Demonstration of background rejection using deep convolutional neural networks in the NEXT experiment
• A convolutional neural network approach for reconstructing polarization information of photoelectric X-ray polarimeters
• First measurement of nuclear recoil head-tail sense in a fiducialised WIMP dark matter detector

Cited by in corpus (2)

• First observation of MeV gamma-ray universe with bijective imaging spectroscopy using the Electron-Tracking Compton Telescope aboard SMILE-2+
• Low-Energy Electron-Track Imaging for a Liquid Argon Time-Projection-Chamber Telescope Concept using Probabilistic Deep Learning