Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/129003
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | Unsupervised task design to meta-train medical image classifiers |
Author: | Maicas, G. Nguyen, C. Motlagh, F. Nascimento, J.C. Carneiro, G. |
Citation: | Proceedings / IEEE International Symposium on Biomedical Imaging: from nano to macro. IEEE International Symposium on Biomedical Imaging, 2020, vol.2020-April, pp.1339-1342 |
Publisher: | IEEE |
Issue Date: | 2020 |
Series/Report no.: | IEEE International Symposium on Biomedical Imaging |
ISBN: | 9781538693308 |
ISSN: | 1945-7928 1945-8452 |
Conference Name: | IEEE International Symposium on Biomedical Imaging (ISBI) (3 Apr 2020 - 7 Apr 2020 : Iowa City, Iowa, USA) |
Statement of Responsibility: | Gabriel Maicasy, Cuong Nguyeny, Farbod Motlaghy, Jacinto C. Nascimentozz, Gustavo Carneiro |
Abstract: | Meta-training has been empirically demonstrated to be the most effective pre-training method for few-shot learning of medical image classifiers (i.e., classifiers modeled with small training sets). However, the effectiveness of meta-training relies on the availability of a reasonable number of hand-designed classification tasks, which are costly to obtain, and consequently rarely available. In this paper, we propose a new method to unsupervisedly design a large number of classification tasks to meta-train medical image classifiers. We evaluate our method on a breast dynamically contrast enhanced magnetic resonance imaging (DCE-MRI) data set that has been used to benchmark few-shot training methods of medical image classifiers. Our results show that the proposed unsupervised task design to meta-train medical image classifiers builds a pre-trained model that, after fine-tuning, produces better classification results than other unsupervised and supervised pre-training methods, and competitive results with respect to meta-training that relies on hand-designed classification tasks. |
Keywords: | meta-training; unsupervised learning; unsupervised task design; breast image analysis; magnetic resonance imaging; few-shot classification; pre-training; clustering |
Rights: | ©2020 IEEE |
DOI: | 10.1109/ISBI45749.2020.9098470 |
Grant ID: | http://purl.org/au-research/grants/arc/DP180103232 |
Published version: | https://ieeexplore.ieee.org/xpl/conhome/9091448/proceeding |
Appears in Collections: | Aurora harvest 4 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.