Effective pseudo-labeling based on heatmap for unsupervised domain adaptation in cell detection
Effective pseudo-labeling based on heatmap for unsupervised domain adaptation in cell detection
复制标题
DOI：
10.1016/j.media.2022.102436
复制
发表时间：
2022-04
影响因子：
10.9
通讯作者：
Hyeonwoo Cho;Kazuya Nishimura;Kazuhide Watanabe;Ryoma Bise
中科院分区：
文献类型：
--
作者：
Hyeonwoo Cho;Kazuya Nishimura;Kazuhide Watanabe;Ryoma Bise
细胞检测是生物医学研究中的一个重要课题。最近，深度学习方法使提高细胞检测性能成为可能。然而，用特定条件（源域）下的训练数据训练的检测网络可能无法很好地处理其他条件（目标域）下的数据，这就是所谓的域转移问题。具体而言，根据研究目的，在不同条件下培养细胞。特征，例如，单元的形状和密度根据条件而改变，并且这种改变可能导致畴移问题。在这里，我们提出了一种使用伪细胞位置热图进行细胞检测的无监督域自适应方法，其中细胞质心位于地图和选择性伪标记中高斯分布的峰值处。在目标域的预测结果中，即使峰值位置正确，峰值周围的信号分布也往往具有非高斯形状。因此，使用预测热图中的峰值位置重新生成伪单元位置热图以具有清晰的高斯形状。我们的方法基于不确定性和课程学习选择自信的伪单元位置热图。我们进行了大量的实验表明，与现有的方法相比，我们的方法提高了检测性能在不同的条件下。
Cell detection is an important task in biomedical research. Recently, deep learning methods have made it possible to improve the performance of cell detection. However, a detection network trained with training data under a specific condition (source domain) may not work well on data under other conditions (target domains), which is called the domain shift problem. In particular, cells are cultured under different conditions depending on the purpose of the research. Characteristics,e.g., the shapes and density of the cells, change depending on the conditions, and such changes may cause domain shift problems. Here, we propose an unsupervised domain adaptation method for cell detection using a pseudo-cell-position heatmap, where the cell centroid is at the peak of a Gaussian distribution in the map and selective pseudo-labeling. In the prediction result for the target domain, even if the peak location is correct, the signal distribution around the peak often has a non-Gaussian shape. The pseudo-cell-position heatmap is thus re-generated using the peak positions in the predicted heatmap to have a clear Gaussian shape. Our method selects confident pseudo-cell-position heatmaps based on uncertainty and curriculum learning. We conducted numerous experiments showing that, compared with the existing methods, our method improved detection performance under different conditions.