arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.04753cs.CV
收藏

HydraMix-Net:一种用于细胞检测和分类的深度多任务半监督学习方法

HydraMix-Net: A Deep Multi-task Semi-supervised Learning Approach for Cell Detection and Classification

University of Warwick(华威大学);Imperial College London(帝国理工学院);The Alan Turing Institute(艾伦·图灵研究所)

R. M. Saad Bashir, Talha Qaiser, Shan E Ahmed Raza, Nasir M. Rajpoot

摘要:

半监督技术通过利用未标记的数据来提高模型的性能,从而消除了大规模标记集的障碍。在本文中,我们提出了一种半监督的深度多任务分类和定位方法HydraMix-Net,用于标记耗时且成本高昂的医学成像领域。首先,使用模型对未标记图像的扩展集的预测来生成伪标签,并进行平均。高熵预测进一步锐化以减少熵,然后与标记集混合进行训练。该模型以多任务学习方式进行训练,具有用于分类定位的噪声容忍联合损失,并且与简单的深度模型相比,在给定有限数据时具有更好的性能。在DLBCL数据上,它实现了80%的准确度,而简单的CNN在仅给出100个标记示例时实现了70%的准确度。

英文摘要:

Semi-supervised techniques have removed the barriers of large scale labelled set by exploiting unlabelled data to improve the performance of a model. In this paper, we propose a semi-supervised deep multi-task classification and localization approach HydraMix-Net in the field of medical imagining where labelling is time consuming and costly. Firstly, the pseudo labels are generated using the model's prediction on the augmented set of unlabelled image with averaging. The high entropy predictions are further sharpened to reduced the entropy and are then mixed with the labelled set for training. The model is trained in multi-task learning manner with noise tolerant joint loss for classification localization and achieves better performance when given limited data in contrast to a simple deep model. On DLBCL data it achieves 80\% accuracy in contrast to simple CNN achieving 70\% accuracy when given only 100 labelled examples.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑