Self-supervised Nuclei Segmentation in Histopathological Images Using Attention

Abstract

Segmentation and accurate localization of nuclei in histopathological images is a very challenging problem, with most existing approaches adopting a supervised strategy. These methods usually rely on manual annotations that require a lot of time and effort from medical experts. In this study, we present a self-supervised approach for segmentation of nuclei for whole slide histopathology images. Our method works on the assumption that the size and texture of nuclei can determine the magnification at which a patch is extracted. We show that the identification of the magnification level for tiles can generate a preliminary self-supervision signal to locate nuclei. We further show that by appropriately constraining our model it is possible to retrieve meaningful segmentation maps as an auxiliary output to the primary magnification identification task. Our experiments show that with standard post-processing, our method can outperform other unsupervised nuclei segmentation approaches and report similar performance with supervised ones on the publicly available MoNuSeg dataset. Our code and models are available online to facilitate further research.

Unsupervised Data-Driven Nuclei Segmentation For Histology Images

“…As shown in the table, our method outperforms all other unsupervised ones by a large margin in either testing scenarios. Notably, the gap is large even for the DL-based unsupervised approach denoted by "self-supervised" [14] in the table. As compared with supervised methods, CBM stands at a competitive position.…”

“…For example, Qu et al [13] proposed a two-stage learning framework using coarse labels. Other researchers [14,15] investigated the self-supervised DL method to reduce the number of required labeled data by exploiting the observation that nuclei size and texture can determine the magnification scale. Besides, there is a domain shift problem [16] arising from stain and nuclei variations in histology images of different organs, patients and acquisition protocols.…”

A 2021 update on cancer image analytics with deep learning

“…Empirical evidence suggests that solving the auxiliary task (e.g., solving a jigsaw) serves as domain‐specific pre‐training by teaching the CNN to extract features that are useful for the main task (e.g., recognizing cancer) as well. Specifically, in cancer image analysis, Self‐Path (Koohbanani, Unnikrishnan, Khurram, Krishnaswamy, & Rajpoot, 2020) used domain specific self‐supervision tasks for effective learning and domain adaptation on histopathology images, while (Sahasrabudhe et al, 2020) used learning to detect patch magnification level as a pretext task for nucleus localization and segmentation.…”

Weakly supervised segmentation with cross-modality equivariant constraints

“…Classical pre-text tasks include predicting image orientation [35] or relative position prediction [36], solving jigsaw puzzles [37], image inpainting [38], colorization [39] and many others [40]- [43]. More recently, equivariance has been employed to impose semantic consistency, either at keypoints [44], class activations [45], feature representations [46] or the network outputs [47]. Nevertheless, a main limitation of these works is that equivariance is enforced across affine transformations of the same image [17], [46], [48] or between virtually generated versions [47].…”

“…More recently, equivariance has been employed to impose semantic consistency, either at keypoints [44], class activations [45], feature representations [46] or the network outputs [47]. Nevertheless, a main limitation of these works is that equivariance is enforced across affine transformations of the same image [17], [46], [48] or between virtually generated versions [47]. For example, Hung et al [47] further apply an appearance perturbation (e.g., color jittering) to the input image before the affine transformation.…”