[en] Boron neutron caputure therapy (BNCT) is a promising treatment for invasive tumors. The BNCT utilizes the energy deposition by alpha particle and ⁷Li created by the neutron capture reaction of ¹⁰B. One of the problem of BNCT is the imaging of ¹⁰B distribution during the treatment. The gamma rays with the energy of 0.478 MeV emitted by ⁷Li can be exploited for this purpose. The photon counting of the gamma rays, however, is very difficult due to the high gamma ray background caused by 2.22 MeV gamma rays originated by the neutron reactions with ¹H. To overcome this difficulty, the authors propose a method to measure gamma rays as electric current by a "transXend" detector, which had been invented by them for energy-resolved X-ray computed tomography. In this paper, the feasibility study on ¹⁰B distribution imaging in a water phantom with a tumor region is shown by simulation studies. For establishing the relationship between electric currents induced in the transXend detector by gamma rays and ¹⁰B concentration, the attenuation of gamma rays by the phantom should be taken into account. For this purpose, a neural network was employed. The results showed good agreement with the training data used for the neural network.