IVAICVLGSep 25, 2024

TSBP: Improving Object Detection in Histology Images via Test-time Self-guided Bounding-box Propagation

arXiv:2409.16678v11 citationsh-index: 3Has Code
Originality Incremental advance
AI Analysis

This work addresses the problem of improving object detection accuracy in histology images for medical diagnosis, though it is incremental as it builds on existing deep learning detection networks.

The paper tackles the suboptimal performance of using a preset global threshold in object detection for histology images by proposing TSBP, a test-time self-guided bounding-box propagation method that leverages Earth Mover's Distance to enhance detection without needing extra labeled data, resulting in significant improvements in gland and cell detection tasks.

A global threshold (e.g., 0.5) is often applied to determine which bounding boxes should be included in the final results for an object detection task. A higher threshold reduces false positives but may result in missing a significant portion of true positives. A lower threshold can increase detection recall but may also result in more false positives. Because of this, using a preset global threshold (e.g., 0.5) applied to all the bounding box candidates may lead to suboptimal solutions. In this paper, we propose a Test-time Self-guided Bounding-box Propagation (TSBP) method, leveraging Earth Mover's Distance (EMD) to enhance object detection in histology images. TSBP utilizes bounding boxes with high confidence to influence those with low confidence, leveraging visual similarities between them. This propagation mechanism enables bounding boxes to be selected in a controllable, explainable, and robust manner, which surpasses the effectiveness of using simple thresholds and uncertainty calibration methods. Importantly, TSBP does not necessitate additional labeled samples for model training or parameter estimation, unlike calibration methods. We conduct experiments on gland detection and cell detection tasks in histology images. The results show that our proposed TSBP significantly improves detection outcomes when working in conjunction with state-of-the-art deep learning-based detection networks. Compared to other methods such as uncertainty calibration, TSBP yields more robust and accurate object detection predictions while using no additional labeled samples. The code is available at https://github.com/jwhgdeu/TSBP.

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