MuRCL: Multi-Instance Reinforcement Contrastive Learning for Whole Slide Image Classification

过度拟合 计算机科学 判别式 人工智能 强化学习 特征(语言学) 特征提取 模式识别(心理学) 机器学习 过程(计算) 特征选择 人工神经网络 语言学 操作系统 哲学
作者
Zhonghang Zhu,Lequan Yu,Wei Wu,Rongshan Yu,Defu Zhang,Liansheng Wang
出处
期刊:IEEE Transactions on Medical Imaging [Institute of Electrical and Electronics Engineers]
卷期号:42 (5): 1337-1348 被引量:19
标识
DOI:10.1109/tmi.2022.3227066
摘要

Multi-instance learning (MIL) is widely adop- ted for automatic whole slide image (WSI) analysis and it usually consists of two stages, i.e., instance feature extraction and feature aggregation. However, due to the "weak supervision" of slide-level labels, the feature aggregation stage would suffer from severe over-fitting in training an effective MIL model. In this case, mining more information from limited slide-level data is pivotal to WSI analysis. Different from previous works on improving instance feature extraction, this paper investigates how to exploit the latent relationship of different instances (patches) to combat overfitting in MIL for more generalizable WSI classification. In particular, we propose a novel Multi-instance Rein- forcement Contrastive Learning framework (MuRCL) to deeply mine the inherent semantic relationships of different patches to advance WSI classification. Specifically, the proposed framework is first trained in a self-supervised manner and then finetuned with WSI slide-level labels. We formulate the first stage as a contrastive learning (CL) process, where positive/negative discriminative feature sets are constructed from the same patch-level feature bags of WSIs. To facilitate the CL training, we design a novel reinforcement learning-based agent to progressively update the selection of discriminative feature sets according to an online reward for slide-level feature aggregation. Then, we further update the model with labeled WSI data to regularize the learned features for the final WSI classification. Experimental results on three public WSI classification datasets (Camelyon16, TCGA-Lung and TCGA-Kidney) demonstrate that the proposed MuRCL outperforms state-of-the-art MIL models. In addition, MuRCL can achieve comparable performance to other state-of-the-art MIL models on TCGA-Esca dataset.
最长约 10秒,即可获得该文献文件

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
刚刚
心灵美从寒完成签到,获得积分10
刚刚
上官若男应助勤奋的真采纳,获得30
1秒前
HighFeng_Lei发布了新的文献求助10
1秒前
大马哈鱼完成签到,获得积分10
1秒前
真的研究牲完成签到,获得积分10
1秒前
青原发布了新的文献求助10
1秒前
汤飞柏发布了新的文献求助10
1秒前
永不止步完成签到,获得积分10
1秒前
2秒前
2秒前
Aten发布了新的文献求助10
2秒前
思源应助三星导弹船采纳,获得10
2秒前
2秒前
老姚完成签到,获得积分10
2秒前
3秒前
l玖发布了新的文献求助30
4秒前
4秒前
5秒前
5秒前
orangel完成签到,获得积分10
6秒前
Logan完成签到,获得积分10
6秒前
7秒前
yar应助科研通管家采纳,获得10
7秒前
安居宝应助科研通管家采纳,获得10
7秒前
完美世界应助科研通管家采纳,获得10
7秒前
科研通AI2S应助科研通管家采纳,获得10
7秒前
ll应助科研通管家采纳,获得10
7秒前
天天快乐应助科研通管家采纳,获得10
7秒前
李健应助冰白采纳,获得10
7秒前
小豆豆应助科研通管家采纳,获得10
8秒前
青原完成签到,获得积分10
8秒前
琉璃应助科研通管家采纳,获得10
8秒前
8秒前
淡定白枫发布了新的文献求助10
8秒前
ll应助科研通管家采纳,获得10
8秒前
qyzhu发布了新的文献求助10
8秒前
典雅碧空应助科研通管家采纳,获得10
8秒前
酷波er应助科研通管家采纳,获得10
8秒前
小马甲应助科研通管家采纳,获得10
8秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
Technical Brochure TB 814: LPIT applications in HV gas insulated switchgear 1000
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Picture Books with Same-sex Parented Families: Unintentional Censorship 500
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 3970572
求助须知:如何正确求助?哪些是违规求助? 3515219
关于积分的说明 11177438
捐赠科研通 3250374
什么是DOI,文献DOI怎么找? 1795265
邀请新用户注册赠送积分活动 875750
科研通“疑难数据库(出版商)”最低求助积分说明 805054