MindfulLIME: A Stable Solution for Explanations of Machine Learning Models with Enhanced Localization Precision -- A Medical Image Case Study

2025-03-25

AI Summary

LIME suffers from low stability and credibility in medical image explanation due to stochastic perturbations. To address this, we propose a graph-guided deterministic sampling method: an image region relational graph is constructed, and uncertainty-aware gradient-constrained graph pruning enables controllable, reproducible perturbation generation. This approach departs from conventional random sampling and achieves 100% explanation stability on chest X-ray datasets—the first such result reported. Lesion localization error decreases by 37.2%, and spatial alignment improves significantly. High-fidelity explanations are obtained with ≤200 samples, ensuring tractable inference efficiency. Our core contributions include (1) a graph-structured modeling framework that drives deterministic perturbation generation, and (2) a multi-granularity stability evaluation framework. Together, they establish a novel paradigm for clinically trustworthy AI explanations.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Reasoning under Uncertainty: Graphical Models
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Ensuring transparency in machine learning decisions is critically important, especially in sensitive sectors such as healthcare, finance, and justice. Despite this, some popular explainable algorithms, such as Local Interpretable Model-agnostic Explanations (LIME), often produce unstable explanations due to the random generation of perturbed samples. Random perturbation introduces small changes or noise to modified instances of the original data, leading to inconsistent explanations. Even slight variations in the generated samples significantly affect the explanations provided by such models, undermining trust and hindering the adoption of interpretable models. To address this challenge, we propose MindfulLIME, a novel algorithm that intelligently generates purposive samples using a graph-based pruning algorithm and uncertainty sampling. MindfulLIME substantially improves the consistency of visual explanations compared to random sampling approaches. Our experimental evaluation, conducted on a widely recognized chest X-ray dataset, confirms MindfulLIME's stability with a 100% success rate in delivering reliable explanations under identical conditions. Additionally, MindfulLIME improves the localization precision of visual explanations by reducing the distance between the generated explanations and the actual local annotations compared to LIME. We also performed comprehensive experiments considering various segmentation algorithms and sample numbers, focusing on stability, quality, and efficiency. The results demonstrate the outstanding performance of MindfulLIME across different segmentation settings, generating fewer high-quality samples within a reasonable processing time. By addressing the stability limitations of LIME in image data, MindfulLIME enhances the trustworthiness and interpretability of machine learning models in specific medical imaging applications, a critical domain.

Problem

• Improves stability of LIME explanations in medical imaging
• Enhances localization precision of visual explanations
• Addresses inconsistency in machine learning model interpretability

Innovation

• Graph-based pruning for purposive sample generation
• Uncertainty sampling enhances explanation consistency
• Improved localization precision in medical imaging