公开(公告)号：WO2020118429A1

公开(公告)日：2020-06-18

申请号：PCT/CA2019/051780

申请日：2019-12-10

Applicant: UNIVERSITY HEALTH NETWORK

Inventor： CHO, Young-bin ,  ISLAM, Mohammad Khairul ,  JAFFRAY, David Anthony ,  NORRLINGER, Bernhard Dieter ,  FARROKHKISH, Makan ,  HEATON, Robert K.

IPC: G01T1/17 ,  A61N5/10 ,  G01T1/02 ,  G01T1/185 ,  G06N3/02 ,  G06N3/08

Abstract: Various embodiments are described herein of radiation systems and methods for monitoring radiation dose are provided for monitoring an amount of radiation in a radiation beam generated by a radiation source for a radiation treatment session, where a radiation sensor is used to provide an actual radiation measurement and an Artificial Neural Network (ANN) engine is used to generate a predicted radiation measurement based on a plurality of feature values for features including radiation field segments from the radiation treatment plan data for the radiation treatment session. The difference between the actual radiation measurement and the predicted radiation measurement is used to determine whether the radiation system is operating in a predetermined safe operation range.