Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/116529
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | Training medical image analysis systems like radiologists |
Author: | Maicas Suso, G. Bradley, A. Nascimento, J. Reid, I. Carneiro, G. |
Citation: | Lecture Notes in Artificial Intelligence, 2018 / Frangi, A., Schnabel, J., Davatzikos, C., Alberola-Lopez, C., Fichtinger, G. (ed./s), vol.11070 LNCS, pp.546-554 |
Publisher: | Springer |
Issue Date: | 2018 |
Series/Report no.: | Lecture notes in computer science; 11070 |
ISBN: | 9783030009274 |
ISSN: | 0302-9743 1611-3349 |
Conference Name: | 21st International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2018) (16 Sep 2018 - 20 Sep 2018 : Granada) |
Editor: | Frangi, A. Schnabel, J. Davatzikos, C. Alberola-Lopez, C. Fichtinger, G. |
Statement of Responsibility: | Gabriel Maicas, Andrew P. Bradley, Jacinto C. Nascimento, Ian Reid, and Gustavo Carneiro |
Abstract: | The training of medical image analysis systems using machine learning approaches follows a common script: collect and annotate a large dataset, train the classifier on the training set, and test it on a hold-out test set. This process bears no direct resemblance with radiologist training, which is based on solving a series of tasks of increasing difficulty, where each task involves the use of significantly smaller datasets than those used in machine learning. In this paper, we propose a novel training approach inspired by how radiologists are trained. In particular, we explore the use of meta-training that models a classifier based on a series of tasks. Tasks are selected using teacher-student curriculum learning, where each task consists of simple classification problems containing small training sets. We hypothesize that our proposed meta-training approach can be used to pre-train medical image analysis models. This hypothesis is tested on the automatic breast screening classification from DCE-MRI trained with weakly labeled datasets. The classification performance achieved by our approach is shown to be the best in the field for that application, compared to state of art baseline approaches: DenseNet, multiple instance learning and multi-task learning. |
Keywords: | Meta-learning; curriculum learning; multi-task training; breast image analysis; breast screening; magnetic resonance imaging |
Rights: | © Springer Nature Switzerland AG 2018 |
DOI: | 10.1007/978-3-030-00928-1_62 |
Grant ID: | http://purl.org/au-research/grants/arc/DP180103232 http://purl.org/au-research/grants/arc/CE140100016 http://purl.org/au-research/grants/arc/FL130100102 |
Published version: | http://dx.doi.org/10.1007/978-3-030-00928-1_62 |
Appears in Collections: | Aurora harvest 3 Australian Institute for Machine Learning publications Computer Science publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.