Weakly Supervised Deep Nuclei Segmentation Using Partial Points Annotation in Histopathology Images

Abstract

Nuclei segmentation is a fundamental task in histopathology image analysis. Typically, such segmentation tasks require significant effort to manually generate accurate pixel-wise annotations for fully supervised training. To alleviate such tedious and manual effort, in this paper we propose a novel weakly supervised segmentation framework based on partial points annotation, i.e., only a small portion of nuclei locations in each image are labeled. The framework consists of two learning stages. In the first stage, we design a semi-supervised strategy to learn a detection model from partially labeled nuclei locations. Specifically, an extended Gaussian mask is designed to train an initial model with partially labeled data. Then, selftraining with background propagation is proposed to make use of the unlabeled regions to boost nuclei detection and suppress false positives. In the second stage, a segmentation model is trained from the detected nuclei locations in a weakly-supervised fashion. Two types of coarse labels with complementary information are derived from the detected points and are then utilized to train a deep neural network. The fully-connected conditional random field loss is utilized in training to further refine the model without introducing extra computational complexity during inference. The proposed method is extensively evaluated on two nuclei segmentation datasets. The experimental results demonstrate that our method can achieve competitive performance compared to the fully supervised counterpart and the state-of-the-art methods while requiring significantly less annotation effort.

Peri-Implant Bone Loss Measurement Using a Region-Based Convolutional Neural Network on Dental Periapical Radiographs

“…Meanwhile, in the field of segmenting biological images, such as cell segmentation, the average IoU [29][30][31], or the mean Dice coefficient [32,33] are frequently used. These metrics can also be applied to humans because they do not require a confidence score.…”

Medical Image Segmentation With Limited Supervision: A Review of Deep Network Models

“…They introduced a concentric loss to make the segmentation network be trained only with the estimated labels in the inside circle and outside the outer circle. For nuclei segmentation, Qu et al [256], [265] addressed a more challenging case, where only sparse points annotation, i.e., only a small portion of nuclei locations in each image, were annotated with center points. Their method consists of two stages, the first stage conducts nuclei detection with a self-training strategy, and the second stage performs semi-supervised segmentation with pseudo-labels generated with Voronoi diagram and k-means clustering.…”

ImPartial: Partial Annotations for Cell Instance Segmentation

“…Using partial annotations instead of fully annotated images has been explored in the context of natural images (13, 20-22, 24, 25). Many of the weakly supervised techniques for nuclei and gland segmentation (16)(17)(18)(19) require patches where partial labels are available for all the instances in order to boost the segmentation loss with coarse labels. Since in our approach we rely on having scribbles of a subset of the available instances in the image (e.g., nuclei), our work most closely relates to the recently proposed Scribble2Label (23).…”

“…The limitations of these methods lie in the assumption that the chosen augmentation family can adequately cover the full range of variation of the tissue to be segmented. Recent weakly supervised techniques, where only nuclei point annotations are provided, have been proposed to avoid reliance on data augmentation (16)(17)(18). These methodologies require fully annotated nuclei patches to generate coarse labels used to regularize a segmentation objective.…”