特征缩放比架构选择更重要：低资源诊所离线呼吸音筛查研究
Feature Scaling Beats Architecture Choice in Offline Respiratory Screening for Low-Resource Clinics
AI 导读
一项针对低资源诊所的离线呼吸音分类研究显示，在包含 1,211 段音频、201 名患者、五类呼吸音（哮喘、COPD、肺炎、支气管音、健康）的数据集上，基于 30 个 MFCC、ZCR、RMS 统计特征的 SVM 达到宏召回 0.78，1.13M 参数的卷积循环网络为 0.75 到 0.77，模型家族之间无显著差异。
正文 · 原文
Abstract:Deep models are the usual choice for respiratory sound classification, but in clinics without GPUs or connectivity that choice carries a cost. We compare classical and deep classifiers on a five-class respiratory-sound corpus of 1,211 clips (asthma, COPD, pneumonia, Bronchial, healthy) from 201 patient groups, with a 3.86:1 class imbalance, using pre-registered patient-grouped cross-validation (25 runs per configuration). An SVM over 30 summary features (MFCC, ZCR and RMS statistics) reaches macro recall 0.78 and 0.71 recall on the rarest class. A 1.13M-parameter convolutional recurrent network over log-Mel spectrograms reaches 0.75 to 0.77, and no difference between model families is resolved. The largest single effect we measure is a preprocessing effect rather than an architectural one. Standardizing the SVM's features, a routine step for RBF kernels, moves its macro recall from 0.51 to 0.78, a 27-point swing wider than any gap between model families here. Augmentation did not help: SMOTE, waveform augmentation and SpecAugment have no measurable effect. Latency does not separate the models (51 to 74 ms per clip on a CPU-only laptop), but the classical models avoid a 449 MB PyTorch runtime. Clip-level splits, which let a patient's recordings fall on both sides, inflate macro recall by 7.5 to 9.8 points. Recording source is strongly tied to label (Cramer's V = 0.65 between native sampling rate and class), and in an exploratory analysis clips recorded at a class's minority rate are mostly assigned to classes typical of that rate; we state this confound as the paper's main limitation.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Sound (cs.SD) |
| Cite as: | arXiv:2610.33030 [cs.LG] |
| (or arXiv:2610.33030v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33030 arXiv-issued DOI via DataCite |
Submission history
From: Abubakar Chilala [view email]
[v1]
Mon, 12 Oct 2026 17:19:47 UTC (455 KB)
来源：arXiv cs.AI · arxiv.org