Weakly Supervised Lymph Nodes Segmentation Based on Partial Instance Annotations with Pre-trained Dual-branch Network and Pseudo Label Learning

2024-08-05

Machine Learning for Biomedical Imaging

AI Summary

To address the high annotation cost for lymph node segmentation in medical imaging, this work investigates weakly supervised learning under partial instance annotation—where only a subset of lymph nodes is labeled. We propose a dynamic hybrid pseudo-labeling mechanism coupled with a dual-branch decoder architecture, integrated with self-supervised pretraining, enabling, for the first time under this setting, robust generation of high-confidence pseudo-labels and reliable fusion of multi-source supervision signals. On the LNQ dataset, our method achieves a Dice coefficient of 54.10% (a relative improvement of 43.06%) and an ASSD of 8.72 mm (a reduction of 12.11 mm), significantly outperforming existing weakly supervised baselines. Our core contributions are: (1) establishing the first weakly supervised framework tailored to partial instance annotation for lymph node segmentation; and (2) mitigating supervision scarcity caused by sparse annotations via dynamic pseudo-label weighting and collaborative optimization of dual decoders.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Computer Vision: Segmentation
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Assessing the presence of potentially malignant lymph nodes aids in estimating cancer progression, and identifying surrounding benign lymph nodes can assist in determining potential metastatic pathways for cancer. For quantitative analysis, automatic segmentation of lymph nodes is crucial. However, due to the labor-intensive and time-consuming manual annotation process required for a large number of lymph nodes, it is more practical to annotate only a subset of the lymph node instances to reduce annotation costs. In this study, we propose a pre-trained Dual-Branch network with Dynamically Mixed Pseudo label (DBDMP) to learn from partial instance annotations for lymph nodes segmentation. To obtain reliable pseudo labels for lymph nodes that are not annotated, we employ a dual-decoder network to generate different outputs that are then dynamically mixed. We integrate the original weak partial annotations with the mixed pseudo labels to supervise the network. To further leverage the extensive amount of unannotated voxels, we apply a self-supervised pre-training strategy to enhance the model’s feature extraction capability. Experiments on the mediastinal Lymph Node Quantification (LNQ) dataset demonstrate that our method, compared to directly learning from partial instance annotations, significantly improves the Dice Similarity Coefficient (DSC) from 11.04% to 54.10% and reduces the Average Symmetric Surface Distance (ASSD) from 20.83 mm to 8.72 mm. The code is available at https://github.com/WltyBY/LNQ2023_training_code

Problem

• Segment lymph nodes using partial annotations
• Reduce annotation costs with pseudo labels
• Improve segmentation accuracy via pre-training

Innovation

• Pre-trained Dual-Branch network for segmentation
• Dynamically Mixed Pseudo label learning
• Self-supervised pre-training strategy