投票掩盖失败：心杂音检测中的聚合选择与噪声鲁棒性
The Vote Hides the Failure: Aggregation Choice and Noise Robustness in Heart Murmur Detection
浏览论文内容
中文总结 AI 辅助
本研究评估两种心杂音检测流水线在噪声下的鲁棒性，发现聚合选择显著影响结论，且稳定聚合分数可能掩盖个体预测退化，强调评估方法的重要性。
中文摘要 AI 辅助
在自动化心音图（PCG）杂音检测中，噪声鲁棒性及其测量方式仍未得到充分研究，尽管人们对低资源筛查的兴趣日益增长。我们在受控的多严重程度噪声条件下，结合噪声增强微调和留出泛化测试，评估了两个独立重新实现的流水线：层次多尺度卷积网络（HMS-Net）--CNN和双向长短期记忆（BiLSTM）--LSTM。在匹配聚合下，完整的BiLSTM流水线在所有条件下的准确率和加权准确率均优于完整的HMS-Net流水线。一个稳定的聚合准确率分数可能误导个体预测所显示的内容：HMS-Net的原生聚合在椒盐噪声下的退化程度远小于相同严重程度下的多数投票（MV）聚合，这一差距反映了其原生规则吸收的窗口级分歧，而BiLSTM的MV准确率在噪声增强训练后上升，尽管其个体预测并未改善。HMS-Net的训练效果在一个准确率指标下显著，但在另一个指标下不显著。噪声鲁棒性结论可能同样依赖于评估选择，而非模型本身。
英文摘要
Noise robustness in automated phonocardiogram (PCG) murmur detection, and how it is measured, remains underexamined despite growing interest in low-resource screening. We evaluate two independently reimplemented pipelines, Hierarchical Multi-Scale Convolutional Network (HMS-Net)--CNN, and Bidirectional Long Short-Term Memory (BiLSTM)--LSTM, under controlled, multi-severity noise with noise-augmented fine-tuning and held-out generalization testing. Under matched aggregation, the complete BiLSTM pipeline outperforms the complete HMS-Net pipeline across all conditions in accuracy and Weighted Accuracy. A stable aggregate accuracy score can misrepresent what individual predictions show: HMS-Net's native aggregation degrades under salt-and-pepper noise far less than majority-vote (MV) aggregation at the same severity, a gap reflecting window-level disagreement its native rule absorbs, while BiLSTM's MV accuracy rises after noise-augmented training even though its individual predictions do not improve. HMS-Net's training effect is significant under one accuracy metric but not another. Noise-robustness conclusions can depend as much on evaluation choices as on the models themselves.
发表机构
- Carnegie Mellon University Africa（卡内基梅隆大学非洲校区）
机构由 AI 辅助整理，请以论文原文为准。
