文献类型：期刊文献

英文题名：PFEMed: Few-shot medical image classification using prior guided feature enhancement

作者：Dai, Zhiyong[1];Yi, Jianjun[2];Yan, Lei[3];Xu, Qingwen[4];Hu, Liang[1,5];Zhang, Qi[1,6];Li, Jiahui[7];Wang, Guoqiang[8]

机构：[1]DeepBlue Acad Sci, Shanghai 200240, Peoples R China;[2]East China Univ Sci & Technol, Sch Mech & Power Engn, Shanghai 200237, Peoples R China;[3]Wenyao Technol, Hainan 571924, Peoples R China;[4]ShanghaiTech Univ, Sch Informat Sci Technol, Shanghai 201210, Peoples R China;[5]Tongji Univ, Shanghai 201804, Peoples R China;[6]Univ Technol Sydney, Sydney, Australia;[7]Univ Shanghai Sci & Technol, Shanghai, Peoples R China;[8]Shanghai Univ Engn Sci, Sch Math Phys & Stat, Shanghai 201620, Peoples R China

年份：2023

卷号：134

外文期刊名：PATTERN RECOGNITION

收录：;EI(收录号：20224312995732);WOS:【SCI-EXPANDED(收录号:WOS:000937988300008)】；

基金：Acknowledgment This paper was supported by the Natural Science Fund of China (NSFC) under Grant No. 51575186, Shanghai Science and Tech-nology Action Plan under Grant No. 21JM0010300, 18DZ1204000, 19510730600, 18510750100, 18510730600, Shanghai Aerospace Science and Technology Innovation Fund (SAST) under Grant No. 2020-59 and Shanghai Sailing Program under Grant No. 19YF1420200.

语种：英文

外文关键词：Deep learning; Domain adaption; Few-shot learning; Medical image classification; Variational autoencoder

摘要：Deep learning-based methods have recently demonstrated outstanding performance on general image classification tasks. As optimization of these methods is dependent on a large amount of labeled data, their application in medical image classification is limited. To address this issue, we propose PFEMed, a novel few-shot classification method for medical images. To extract general and specific features from medical images, this method employs a dual-encoder structure, that is, one encoder with fixed weights pre-trained on public image classification datasets and another encoder trained on the target medical dataset. In addition, we introduce a novel prior-guided Variational Autoencoder (VAE) module to enhance the robustness of the target feature, which is the concatenation of the general and specific features. Then, we match the target features extracted from both the support and query medical image samples and pre-dict the category attribution of the query examples. Extensive experiments on several publicly available medical image datasets show that our method outperforms current state-of-the-art few-shot methods by a wide margin, particularly outperforming MetaMed on the Pap smear dataset by over 2.63%.(c) 2022 Elsevier Ltd. All rights reserved.