UGPL：基於不確定性引導的漸進式學習用於計算機斷層掃描中的循證分類

UGPL: Uncertainty-Guided Progressive Learning for Evidence-Based Classification in Computed Tomography

July 18, 2025

作者: Shravan Venkatraman, Pavan Kumar S, Rakesh Raj Madavan, Chandrakala S

cs.AI

摘要

精確分類計算機斷層掃描（CT）影像對於診斷與治療規劃至關重要，然而現有方法往往難以應對病理特徵的細微性與空間多樣性。當前方法通常對影像進行均一化處理，限制了其檢測需重點分析的局部異常的能力。我們提出了UGPL，一種基於不確定性引導的漸進學習框架，該框架通過首先識別診斷模糊區域，隨後對這些關鍵區域進行細緻檢查，實現了從全局到局部的分析。我們的方法採用證據深度學習來量化預測不確定性，通過非極大值抑制機制引導信息豐富的圖塊提取，保持空間多樣性。這一漸進式精煉策略與自適應融合機制相結合，使UGPL能夠整合上下文信息與細粒度細節。在三個CT數據集上的實驗表明，UGPL在腎臟異常、肺癌及COVID-19檢測的準確率上分別提升了3.29%、2.46%和8.08%，持續超越現有最先進方法。我們的分析顯示，不確定性引導組件帶來了顯著效益，當完整實施漸進學習流程時，性能顯著提升。我們的代碼已公開於：https://github.com/shravan-18/UGPL。

English

Accurate classification of computed tomography (CT) images is essential for diagnosis and treatment planning, but existing methods often struggle with the subtle and spatially diverse nature of pathological features. Current approaches typically process images uniformly, limiting their ability to detect localized abnormalities that require focused analysis. We introduce UGPL, an uncertainty-guided progressive learning framework that performs a global-to-local analysis by first identifying regions of diagnostic ambiguity and then conducting detailed examination of these critical areas. Our approach employs evidential deep learning to quantify predictive uncertainty, guiding the extraction of informative patches through a non-maximum suppression mechanism that maintains spatial diversity. This progressive refinement strategy, combined with an adaptive fusion mechanism, enables UGPL to integrate both contextual information and fine-grained details. Experiments across three CT datasets demonstrate that UGPL consistently outperforms state-of-the-art methods, achieving improvements of 3.29%, 2.46%, and 8.08% in accuracy for kidney abnormality, lung cancer, and COVID-19 detection, respectively. Our analysis shows that the uncertainty-guided component provides substantial benefits, with performance dramatically increasing when the full progressive learning pipeline is implemented. Our code is available at: https://github.com/shravan-18/UGPL