Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
LEARNING DEVICE, METHOD, AND PROGRAM, MEDICAL IMAGE PROCESSING DEVICE, METHOD, AND PROGRAM, AND DISCRIMINATOR
Document Type and Number:
WIPO Patent Application WO/2020/262681
Kind Code:
A1
Abstract:
According to the present invention, an information acquisition unit acquires: an image for learning that covers an affected region; and a first teacher label that identifies the affected region included in the image for learning. A teacher label generation unit generates one or more second teacher labels that have different criteria for identifying an affected region from those of the first teacher label. On the basis of the image for learning, the first teacher label, and the one or more second teacher labels, a learning unit learns a discriminator that detects an affected region included in an object image.

Inventors:
TAKEI MIZUKI (JP)
Application Number:
PCT/JP2020/025399
Publication Date:
December 30, 2020
Filing Date:
June 26, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
FUJIFILM CORP (JP)
International Classes:
A61B6/03; A61B5/055; G01T1/161; G06T7/00; G16H30/40; G16H50/50
Foreign References:
JP2018061771A2018-04-19
US20190192096A12019-06-27
JP2019509813A2019-04-11
Other References:
SHIMABARA YUKI: "Development of cerebral aneurysm detection software utilizing deep learning", MEDICAL IMAGING AND INFORMATION SCIENCES, vol. 34, no. 2, 2017, pages 103 - 104
NATARAJAN, NAGARAJAN ET AL.: "Leaning with Noisy Labels", NIPS'13: PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING SYSTEMS, vol. 1, 2013, pages 1196 - 1204, XP055603359
HAN, JIANGFAN ET AL.: "Deep Self-Learning From Noisy Labels", DEEP SELF-LEARNING FROM NOISY LABELS, 6 August 2019 (2019-08-06), XP033723138
Attorney, Agent or Firm:
TAIYO, NAKAJIMA & KATO (JP)
Download PDF: