领域自适应方法用于医学影像研究进展
Research progresses of domain adaptive methods for medical imaging
岳珂娟 1伍炯星 2谢东3
作者信息
- 1. 湖南第一师范学院计算机学院,湖南长沙 410205
- 2. 中南大学湘雅三医院急诊科,湖南长沙 410013
- 3. 湖南人文科技学院信息学院,湖南娄底 417000
- 折叠
摘要
人工智能有助于提高医学影像学诊断准确率、提高工作效率,但训练模型的过程中需要对大量图像数据进行标注,且需面临域偏移等问题;利用领域自适应方法可基于少量标注数据训练高效模型.本文就领域自适应方法用于医学影像研究进展进行综述.
Abstract
Artificial intelligence(AI)can help improve the accuracy and efficiency of medical imaging diagnosis,but training models needs a large amount of image data to be annotated,also faces problems such as domain shift.Using domain adaptive methods can train efficient models based on a small amount of annotated data.The research progresses of domain adaptive methods for medical imaging were reviewed in this article.
关键词
机器学习/诊断显像/领域自适应/域偏移Key words
machine learning/diagnostic imaging/domain adaptation/domain shift引用本文复制引用
基金项目
湖南省自然科学基金(2021JJ30173)
国家留学基金资助(留金项[2022]20号)
国家留学基金资助(202208430070)
出版年
2024
NETL
NSTL
万方数据