Title:
PATIENT POSITION DETERMINATION SYSTEM, METHOD, AND PROGRAM
Document Type and Number:
WIPO Patent Application WO/2020/217569
Kind Code:
A1
Abstract:
The purpose of the present invention is to rapidly and accurately establish the amount of deviation in the position/orientation of a patient with respect to a three-dimensional image of the patient acquired prior to radiation therapy. A radiation therapy system (100) comprises X-ray imaging devices (5A, 5B) for acquiring an X-ray fluoroscopic image of a patient on a reference plane, and a position determination device (10). The position determination device (10) calculates a parameter of a region estimation model, with a reference fluoroscopic image acquired prior to radiation therapy as input data, and a reference ROI image acquired prior to radiation therapy with respect to the reference fluoroscopic image as teacher data. During radiation therapy, the position determination device (10) estimates a region of interest with respect to the X-ray fluoroscopic image and a DRR image on the basis of the parameter and the X-ray fluoroscopic image, establishes the degree of coincidence for the region of interest between the X-ray fluoroscopic image and the DRR image while virtually varying the position/orientation relationship between a CT image and the reference plane, and establishes the amount of deviation in position/orientation between the patient and the CT image on the basis of the position/orientation relationship when the degree of coincidence satisfies a predetermined condition.
More Like This:
Inventors:
MIYAZAKI KOICHI (JP)
UMEKAWA TORU (JP)
UMEKAWA TORU (JP)
Application Number:
PCT/JP2019/040787
Publication Date:
October 29, 2020
Filing Date:
October 17, 2019
Export Citation:
Assignee:
HITACHI LTD (JP)
International Classes:
A61N5/10
Domestic Patent References:
| WO2018083844A1 | 2018-05-11 |
Foreign References:
| JP2010187991A | 2010-09-02 | |||
| JP2018089065A | 2018-06-14 | |||
| JP2018153299A | 2018-10-04 | |||
| JP2010246883A | 2010-11-04 |
Other References:
LONG, JONATHANEVAN SHELHAMERTREVOR DARRELL: "Fully convolutional networks for semantic segmentation", PROC. OF IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR, 2015
See also references of EP 3960235A4
See also references of EP 3960235A4
Attorney, Agent or Firm:
YKI INTELLECTUAL PROPERTY ATTORNEYS (JP)
Download PDF:
Previous Patent: INORGANIC COMPOSITION RESISTANT TO DEGRADATION BY RADIATION, AND FIBER THEREOF
Next Patent: PRINTING PARAMETER ACQUISITION DEVICE
Next Patent: PRINTING PARAMETER ACQUISITION DEVICE