Mitosis Detection from Partial Annotation by Dataset Generation via Frame-Order Flipping
This work addresses the time-consuming labor of collecting fully annotated data for mitosis detection in biomedical research, offering an incremental improvement by enabling training with partial annotations.
The paper tackles the problem of mitosis detection in biomedical imaging by proposing a method that can be trained with partially annotated sequences, generating a fully labeled dataset via frame-order flipping and alpha-blending pasting, and demonstrates performance improvements over other methods using partial labels on four datasets.
Detection of mitosis events plays an important role in biomedical research. Deep-learning-based mitosis detection methods have achieved outstanding performance with a certain amount of labeled data. However, these methods require annotations for each imaging condition. Collecting labeled data involves time-consuming human labor. In this paper, we propose a mitosis detection method that can be trained with partially annotated sequences. The base idea is to generate a fully labeled dataset from the partial labels and train a mitosis detection model with the generated dataset. First, we generate an image pair not containing mitosis events by frame-order flipping. Then, we paste mitosis events to the image pair by alpha-blending pasting and generate a fully labeled dataset. We demonstrate the performance of our method on four datasets, and we confirm that our method outperforms other comparisons which use partially labeled sequences.