Cited 0 time in webofscience Cited 0 time in scopus

Recurrent Attention Models for Tissue Histopathology Image Classification

Title
Recurrent Attention Models for Tissue Histopathology Image Classification
Authors
Chikontwe, PhilipKim, JunghwanWon, Dong KyuUllah, IhsanPark, Sang Hyun
DGIST Authors
Park, Sang Hyun
Issue Date
2019-02-15
Citation
IPIU 2019: 제 31회 영상처리 및 이해에 관한 워크샵
Type
Conference
Abstract
Histopathology continues to be the de-facto gold standard for cancer recognition and diagnosis. Pathologists greatly benefit from the automation of pathology image analysis for fast diagnosis of tumors and their subtypes. To automatically analyze the pathology image, in this paper, we propose to leverage the recurrent attention model. The recurrent attention model selectively focuses on small number of locations in a large patch called a glimpse for feature extraction at each time step, updates its internal state, and chooses the next location to attend. The experimental results show that our proposed method achieves comparable performance to previous methods that rely on extensive patch sampling techniques.
URI
http://hdl.handle.net/20.500.11750/14543
Publisher
한국 방송・미디어 공학회
Related Researcher
  • Author Park, Sang Hyun Medical Image & Signal Processing Lab
  • Research Interests 컴퓨터비전, 인공지능, 의료영상처리
Files:
There are no files associated with this item.
Collection:
Department of Robotics and Mechatronics EngineeringMedical Image & Signal Processing Lab2. Conference Papers


qrcode mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE