Cluster-Guided Semi-Supervised Domain Adaptation for Imbalanced Medical Image Classification
Cluster-Guided Semi-Supervised Domain Adaptation for Imbalanced Medical Image Classification
复制标题
DOI：
10.1109/isbi53787.2023.10230451
复制
发表时间：
2023-03
期刊：
影响因子：
--
通讯作者：
S. Harada;Ryoma Bise;Kengo Araki;A. Yoshizawa;K. Terada;Mariyo Rokutan-Kurata;N. Nakajima;Hiroyuki Abe;T. Ushiku;Seiichi Uchida
中科院分区：
文献类型：
--
作者：
S. Harada;Ryoma Bise;Kengo Araki;A. Yoshizawa;K. Terada;Mariyo Rokutan-Kurata;N. Nakajima;Hiroyuki Abe;T. Ushiku;Seiichi Uchida
半监督域适应是一种通过使用来自目标域的许多未标记样本和少量标记样本修改另一个（源）域中的分类器来构建目标域分类器的技术。在本文中，我们开发了一种半监督域适应方法，该方法对医学图像分类任务中常见的类不平衡情况具有鲁棒性。为了鲁棒性，我们提出了一种弱监督的聚类管道来获得高纯度的聚类，并在表示学习中利用这些聚类来进行领域适应。所提出的方法在使用严重类别不平衡的病理图像块的实验中显示出最先进的性能。
Semi-supervised domain adaptation is a technique to build a classifier for a target domain by modifying a classifier in another (source) domain using many unlabeled samples and a small number of labeled samples from the target domain. In this paper, we develop a semi-supervised domain adaptation method, which has robustness to class-imbalanced situations, which are common in medical image classification tasks. For robustness, we propose a weakly-supervised clustering pipeline to obtain high-purity clusters and utilize the clusters in representation learning for domain adaptation. The proposed method showed state-of-the-art performance in the experiment using severely class-imbalanced pathological image patches.