CVLGSep 22, 2021

Label Cleaning Multiple Instance Learning: Refining Coarse Annotations on Single Whole-Slide Images

arXiv:2109.10778v237 citations
Originality Incremental advance
AI Analysis

This addresses the labor-intensive and costly challenge of generating accurate annotations in digital pathology for clinical diagnosis and research, offering a lightweight tool for pathologists.

The paper tackles the problem of refining coarse annotations in whole-slide images for pathology, presenting LC-MIL, a method that significantly improves annotation accuracy using only a single slide without external training data, outperforming state-of-the-art alternatives.

Annotating cancerous regions in whole-slide images (WSIs) of pathology samples plays a critical role in clinical diagnosis, biomedical research, and machine learning algorithms development. However, generating exhaustive and accurate annotations is labor-intensive, challenging, and costly. Drawing only coarse and approximate annotations is a much easier task, less costly, and it alleviates pathologists' workload. In this paper, we study the problem of refining these approximate annotations in digital pathology to obtain more accurate ones. Some previous works have explored obtaining machine learning models from these inaccurate annotations, but few of them tackle the refinement problem where the mislabeled regions should be explicitly identified and corrected, and all of them require a -- often very large -- number of training samples. We present a method, named Label Cleaning Multiple Instance Learning (LC-MIL), to refine coarse annotations on a single WSI without the need of external training data. Patches cropped from a WSI with inaccurate labels are processed jointly within a multiple instance learning framework, mitigating their impact on the predictive model and refining the segmentation. Our experiments on a heterogeneous WSI set with breast cancer lymph node metastasis, liver cancer, and colorectal cancer samples show that LC-MIL significantly refines the coarse annotations, outperforming state-of-the-art alternatives, even while learning from a single slide. Moreover, we demonstrate how real annotations drawn by pathologists can be efficiently refined and improved by the proposed approach. All these results demonstrate that LC-MIL is a promising, light-weight tool to provide fine-grained annotations from coarsely annotated pathology sets.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes