Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Location embedding based pairwise distance learning for fine-grained diagnosis of urinary stones

Qiangguo Jin, Jiapeng Huang, Changming Sun, Hui Cui, Ping Xuan, Ran Su, Leyi Wei, Yu-Jie Wu, Chia-An Wu, Henry B. L. Duh, Yueh-Hsun Lu

TL;DR

Urinary stone diagnosis from low-dose KUB radiographs is challenging due to low contrast and variable stone locations. The authors propose LEPD-Net, a three-module network combining context-aware region enhancement, stone location embedding, and fine-grained pairwise distance learning to fuse location priors with image features and promote discriminative, sample-efficient learning. An in-house dataset of 414 patients with 974 stone patches is introduced to validate the approach; LEPD-Net outperforms multiple state-of-the-art methods, with ablations confirming the contributions of CRE, SLE, and FPD. The work demonstrates that incorporating precise stone location information and pairwise discrimination significantly improves fine-grained urinary stone diagnosis on cost-effective KUB imaging, with potential applicability to other medical imaging tasks.

Abstract

The precise diagnosis of urinary stones is crucial for devising effective treatment strategies. The diagnostic process, however, is often complicated by the low contrast between stones and surrounding tissues, as well as the variability in stone locations across different patients. To address this issue, we propose a novel location embedding based pairwise distance learning network (LEPD-Net) that leverages low-dose abdominal X-ray imaging combined with location information for the fine-grained diagnosis of urinary stones. LEPD-Net enhances the representation of stone-related features through context-aware region enhancement, incorporates critical location knowledge via stone location embedding, and achieves recognition of fine-grained objects with our innovative fine-grained pairwise distance learning. Additionally, we have established an in-house dataset on urinary tract stones to demonstrate the effectiveness of our proposed approach. Comprehensive experiments conducted on this dataset reveal that our framework significantly surpasses existing state-of-the-art methods.

Location embedding based pairwise distance learning for fine-grained diagnosis of urinary stones

TL;DR

Urinary stone diagnosis from low-dose KUB radiographs is challenging due to low contrast and variable stone locations. The authors propose LEPD-Net, a three-module network combining context-aware region enhancement, stone location embedding, and fine-grained pairwise distance learning to fuse location priors with image features and promote discriminative, sample-efficient learning. An in-house dataset of 414 patients with 974 stone patches is introduced to validate the approach; LEPD-Net outperforms multiple state-of-the-art methods, with ablations confirming the contributions of CRE, SLE, and FPD. The work demonstrates that incorporating precise stone location information and pairwise discrimination significantly improves fine-grained urinary stone diagnosis on cost-effective KUB imaging, with potential applicability to other medical imaging tasks.

Abstract

The precise diagnosis of urinary stones is crucial for devising effective treatment strategies. The diagnostic process, however, is often complicated by the low contrast between stones and surrounding tissues, as well as the variability in stone locations across different patients. To address this issue, we propose a novel location embedding based pairwise distance learning network (LEPD-Net) that leverages low-dose abdominal X-ray imaging combined with location information for the fine-grained diagnosis of urinary stones. LEPD-Net enhances the representation of stone-related features through context-aware region enhancement, incorporates critical location knowledge via stone location embedding, and achieves recognition of fine-grained objects with our innovative fine-grained pairwise distance learning. Additionally, we have established an in-house dataset on urinary tract stones to demonstrate the effectiveness of our proposed approach. Comprehensive experiments conducted on this dataset reveal that our framework significantly surpasses existing state-of-the-art methods.
Paper Structure (13 sections, 5 equations, 3 figures, 2 tables)

This paper contains 13 sections, 5 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Dataset and preprocessing
  3. Methodology
  4. Context-aware region enhancement (CRE) module
  5. Stone location embedding (SLE) module
  6. Fine-grained pairwise distance learning (FPD) module
  7. Loss function
  8. Experiments and results
  9. Implementation details and evaluation measures
  10. Performance comparison
  11. Ablation study
  12. Conclusions
  13. Acknowledgements

Figures (3)

  • Figure 1: The overall architecture of LEPD-Net. (a) Context-aware region enhancement module. (b) Stone location embedding module. (c) Fine-grained pairwise distance learning module. It is noted that the fine-grained pairwise distance learning module will be removed during inference.
  • Figure 2: (a) Visualization of location distributions of 974 stones with four different types. (b) An example of typical stone patch and the location information.
  • Figure 3: (a) AUC-ROC and Precision-Recall curves for our LEPF-Net and other comparing methods. (b) Visual saliency maps for challenging-to-classify stones.