Scholarlink

Abstract <p>Early concurrent screening of multiple retinal pathologies is vital for preventing vision loss but poses substantial challenges for deep learning models. These include complex disease co-occurrences, severe class imbalance (long-tailed distributions), and the imperative for diagnostic logic (e.g., avoiding conflicting “Normal” and “Disease” predictions. This paper presents C<InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(^2\)</EquationSource> </InlineEquation>Net, a unified Co-Occurrence and Consistency-Aware Network designed to address these coupled challenges simultaneously. Specifically, C<InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(^2\)</EquationSource> </InlineEquation>Net integrates three complementary components: (1) a Co-Occurrence-Aware Supervised Contrastive Loss (CoaSupCon) that explicitly encodes positive label dependencies to capture disease correlations; (2) an Auto-Adaptive Multi-Label Focal Loss (Auto-MLFocal) that automatically learns per-class balancing and focusing parameters to mitigate long-tail imbalance; and (3) a Diagnostic Consistency Mechanism (DCM) that enforces negative dependencies and clinical logic via conflict suppression and default-to-other constraints. By unifying positive and negative dependency modeling, C<InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(^2\)</EquationSource> </InlineEquation>Net promotes robust and clinically reliable predictions. Extensive evaluations on the ODIR-5K and MuReD benchmarks under a rigorous 5-fold cross-validation protocol demonstrate that C<InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(^2\)</EquationSource> </InlineEquation>Net achieves consistently better performance than representative state-of-the-art methods. On ODIR-5K, it achieves a macro-average F1 (AF1) of 72.1% and a macro-average Precision (mAP) of 74.4%. On MuReD, it reaches an AF1 of 77.1% and a mAP of 80.5%. Results confirm that C<InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(^2\)</EquationSource> </InlineEquation>Net not only improves diagnostic accuracy but also improves logical reliability, offering a unified framework for multi-label medical imaging.</p> Graphical abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

C\(^2\)Net: A co-occurrence and consistency-aware framework for structured multi-label fundus diagnosis

  • 期刊论文

  • 发表日期：

    2026年4月28日

查看全文
  • Tianqi Wang,
  • Qingshan Hou,
  • Peng Cao,
  • Jinzhu Yang,
  • Osmar R. Zaiane

摘要

Abstract

Early concurrent screening of multiple retinal pathologies is vital for preventing vision loss but poses substantial challenges for deep learning models. These include complex disease co-occurrences, severe class imbalance (long-tailed distributions), and the imperative for diagnostic logic (e.g., avoiding conflicting “Normal” and “Disease” predictions. This paper presents C \(^2\) Net, a unified Co-Occurrence and Consistency-Aware Network designed to address these coupled challenges simultaneously. Specifically, C \(^2\) Net integrates three complementary components: (1) a Co-Occurrence-Aware Supervised Contrastive Loss (CoaSupCon) that explicitly encodes positive label dependencies to capture disease correlations; (2) an Auto-Adaptive Multi-Label Focal Loss (Auto-MLFocal) that automatically learns per-class balancing and focusing parameters to mitigate long-tail imbalance; and (3) a Diagnostic Consistency Mechanism (DCM) that enforces negative dependencies and clinical logic via conflict suppression and default-to-other constraints. By unifying positive and negative dependency modeling, C \(^2\) Net promotes robust and clinically reliable predictions. Extensive evaluations on the ODIR-5K and MuReD benchmarks under a rigorous 5-fold cross-validation protocol demonstrate that C \(^2\) Net achieves consistently better performance than representative state-of-the-art methods. On ODIR-5K, it achieves a macro-average F1 (AF1) of 72.1% and a macro-average Precision (mAP) of 74.4%. On MuReD, it reaches an AF1 of 77.1% and a mAP of 80.5%. Results confirm that C \(^2\) Net not only improves diagnostic accuracy but also improves logical reliability, offering a unified framework for multi-label medical imaging.

Graphical abstract