arXiv cs.AI· Yi Tian, Xiaohuan Chen, Yufei tang, Xinyu Lin, Hongliang Li, Chengbin Hou, Degen Huang, Rongshan Yu·· 3 天前AI 评分22
RespFocus：用多实例学习为弱监督呼吸音分类提供可检验的时间证据
RespFocus: Learning Temporal Evidence for Weakly Supervised Respiratory Sound Classification
AI 导读
RespFocus 是一个基于多实例学习（MIL）的弱监督框架，从周期级标签中学习可检验的时间声学证据并融入分类，在公开 ICBHI 2017 上取得四分类 Score 69.55%，在新建的多中心儿科呼吸音队列 PedResp2026 上取得 72.91%。PedResp2026 采集自常规临床环境，带有经裁定的周期级标注。证据诊断与匹配的时间干预实验支持所学证据对模型预测的功能性贡献。
正文 · 原文
Abstract:Respiratory auscultation provides a low-cost means of assessing pulmonary function, yet transient crackles and sustained wheezes can be difficult to recognize in noisy clinical recordings. Conventional deep learning classifiers often operate as black boxes, offering limited insight into the temporal acoustic evidence supporting their predictions. We present RespFocus, a weakly supervised framework that uses multiple instance learning (MIL) to learn inspectable temporal evidence from cycle-level labels and integrate it into classification. We also introduce PedResp2026, a multicenter pediatric respiratory sound cohort collected in routine clinical environments with adjudicated cycle-level annotations. In classification experiments, RespFocus achieves four-class Scores of 69.55\% on public ICBHI 2017 and 72.91\% on PedResp2026. Evidence diagnostics and matched temporal interventions support the functional contribution of the learned evidence to model predictions. Together, these findings highlight RespFocus's potential as a tool for automated respiratory sound diagnosis.
| Comments: | 9 pages, 5 figures,2 tables |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.15702 [cs.LG] |
| (or arXiv:2610.15702v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.15702 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hongliang Li [view email]
[v1]
Sun, 11 Oct 2026 15:37:56 UTC (3,947 KB)
来源：arXiv cs.AI · arxiv.org