SAM-Sode: Towards Faithful Explanations for Tiny Bacteria Detection

2026-05-20

AI Summary

This work addresses the challenges of boundary ambiguity and attribution dispersion in traditional explainable methods for detecting minute bacteria, which arise from sparse target morphology and complex backgrounds. To overcome these limitations, the authors propose the SAM-Sode framework, which transforms initial feature attribution maps into geometry-aware prompts and leverages the prior knowledge of the SAM3 foundation model for spatial refinement and morphological reconstruction. A dual-constraint mechanism incorporating both physical plausibility and geometric alignment ensures high-quality, instance-level denoising. Evaluated on a newly curated dataset of 2,524 bacterial images with intricate circuit-board backgrounds as well as multiple public benchmarks, the method significantly suppresses background interference and produces more coherent, faithful, and expert-aligned explanations, thereby substantially enhancing decision transparency in microscopic object detection.

Abstract

Interpretability in object detection provides crucial confidence support for clinical auxiliary diagnosis. However, in tiny bacteria detection, traditional explanation methods often suffer from blurred foreground boundaries and diffuse feature attribution due to the extreme sparsity of target morphological features and severe interference from complex backgrounds. Such limitations hinder the provision of logically coherent morphological evidence. To bridge this gap, we propose a novel eXplainable AI (XAI) framework, SAM-Sode. The framework innovatively transforms initial feature attribution maps into geometry-aware prompts, leveraging the prior knowledge of the foundation model (SAM3) to achieve spatial refinement and morphological reconstruction of the explanatory mappings. Furthermore, we introduce a dual-constraint mechanism based on physical significance and geometric alignment to perform instance-level denoising, generating coherent explanations that better align with human expert intuition. Experimental results on our self-constructed bacteria dataset with complex circuit backgrounds (containing 2,524 images) and other public datasets demonstrate that the proposed method effectively suppresses background redundancy and significantly enhances the decision-making transparency of tiny object detection.

Problem

• tiny bacteria detection
• interpretability
• feature attribution
• morphological evidence
• object detection

Innovation

• explainable AI
• tiny object detection
• geometry-aware prompting
• feature attribution refinement
• dual-constraint denoising