Medical image classification using self-supervised learning-based masked autoencoder

人工智能 计算机科学 可解释性 自编码 模式识别(心理学) 机器学习 深度学习 上下文图像分类 特征学习 特征(语言学) 遮罩(插图) 特征提取 图像(数学) 艺术 语言学 哲学 视觉艺术
作者
Zong Fan,Zhimin Wang,Ping Gong,Christine U. Lee,Shanshan Tang,Xiaohui Zhang,Yao Hao,Zhongwei Zhang,Pengfei Song,Shigao Chen,Li Hua
标识
DOI:10.1117/12.3006938
摘要

Accurate classification of medical images is crucial for disease diagnosis and treatment planning. Deep learning (DL) methods have gained increasing attention in this domain. However, DL-based classification methods encounter challenges due to the unique characteristics of medical image datasets, including limited amounts of labeled images and large image variations. Self-supervised learning (SSL) has emerged as a solution that learns informative representations from unlabeled data to alleviate the scarcity of labeled images and improve model performance. A recently proposed generative SSL method, masked autoencoder (MAE), has shown excellent capability in feature representation learning. The MAE model trained on unlabeled data can be easily tuned to improve the performance of various downstream classification models. In this paper, we performed a preliminary study to integrate MAE with the self-attention mechanism for tumor classification on breast ultrasound (BUS) data. Considering the speckle noise, image quality variations of BUS images, and varying tumor shapes and sizes, two revisions were adopted in using MAE for tumor classification. First, MAE's patch size and masking ratio were adjusted to avoid missing information embedded in small lesions on BUS images. Second, attention maps were extracted to improve the interpretability of the model's decision-making process. Experiments demonstrated the effectiveness and potential of the MAE-based classification model on small labeled datasets.

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
乐乐应助轩然采纳,获得10
3秒前
科研通AI5应助简单小土豆采纳,获得10
4秒前
kk发布了新的文献求助10
5秒前
6秒前
7秒前
鄢懋卿完成签到,获得积分10
7秒前
小代完成签到,获得积分10
7秒前
7秒前
9秒前
刘昌虎发布了新的文献求助10
9秒前
10秒前
xyy9919完成签到,获得积分10
10秒前
cassie发布了新的文献求助30
11秒前
乐天发布了新的文献求助10
11秒前
没问题完成签到,获得积分20
12秒前
为之发布了新的文献求助10
13秒前
14秒前
轩然发布了新的文献求助10
14秒前
晴空万里完成签到 ,获得积分10
14秒前
Koi发布了新的文献求助30
15秒前
轩然完成签到,获得积分10
18秒前
徐小发布了新的文献求助10
20秒前
刘昌虎完成签到,获得积分20
21秒前
为之完成签到,获得积分20
22秒前
VitoLi发布了新的文献求助10
23秒前
23秒前
上官若男应助小鞠佩奇采纳,获得10
25秒前
现代绮玉完成签到,获得积分10
25秒前
qiu发布了新的文献求助10
27秒前
枫1538完成签到,获得积分10
28秒前
疯狂的炳发布了新的文献求助10
28秒前
NexusExplorer应助U9A采纳,获得10
31秒前
wanci应助喽喽采纳,获得10
33秒前
33秒前
Newt应助明天会更好采纳,获得10
34秒前
34秒前
英俊的如霜完成签到,获得积分10
36秒前
37秒前
37秒前
38秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
Cognitive Neuroscience: The Biology of the Mind 1000
Technical Brochure TB 814: LPIT applications in HV gas insulated switchgear 1000
Immigrant Incorporation in East Asian Democracies 600
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
A Preliminary Study on Correlation Between Independent Components of Facial Thermal Images and Subjective Assessment of Chronic Stress 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3967482
求助须知:如何正确求助?哪些是违规求助? 3512759
关于积分的说明 11164944
捐赠科研通 3247740
什么是DOI,文献DOI怎么找? 1794021
邀请新用户注册赠送积分活动 874785
科研通“疑难数据库(出版商)”最低求助积分说明 804517