[论文解读] Towards Objective Gastrointestinal Auscultation: Automated Segmentation and Annotation of Bowel Sound Patterns

Zahra Mansour, Verena Uslar+3 more|arXiv (Cornell University)|Mar 7, 2026

Phonocardiography and Auscultation Techniques被引用 1

一句话总结

论文提出一个自动化流水线，检测肠鸣音事件，将其分类为四种模式，使用预训练模型（AST 和 Wav2Vec 2.0），并对队列特定的性能进行评价，同时显著减少手动标注工作量。

ABSTRACT

Bowel sounds (BS) are typically momentary and have low amplitude, making them difficult to detect accurately through manual auscultation. This leads to significant variability in clinical assessment. Digital acoustic sensors allow the acquisition of high-quality BS and enable automated signal analysis, offering the potential to provide clinicians with both objective and quantitative feedback on bowel activity. This study presents an automated pipeline for bowel sound segmentation and classification using a wearable acoustic SonicGuard sensor. BS signals from 83 subjects were recorded using a SonicGuard sensor. Data from 40 subjects were manually annotated by clinical experts and used to train an automatic annotation algorithm, while the remaining subjects were used for further model evaluation. An energy-based event detection algorithm was developed to detect BS events. Detected sound segments were then classified into BS patterns using a pretrained Audio Spectrogram Transformer (AST) model. Model performance was evaluated separately for healthy individuals and patients. The best configuration used two specialized models, one trained on healthy subjects and one on patients, achieving (accuracy: 0.97, AUROC: 0.98) for healthy group and (accuracy: 0.96, AUROC: 0.98) for patient group. The auto-annotation method reduced manual labeling time by approximately 70%, and expert review showed that less than 12% of automatically detected segments required correction. The proposed automated segmentation and classification system enables quantitative assessment of bowel activity, providing clinicians with an objective diagnostic tool that may improve the diagnostic of gastrointestinal function and support the annotation of large-scale datasets.

研究动机与目标

• 提供对来自可穿戴传感器的肠鸣音的自动、客观分析。
• 在多样化模式下鲁棒地分段肠鸣音事件。
• 将分段事件分类为四种临床相关的肠鸣音模式。
• 在保持专家级标注质量的前提下，降低手动标注工作量。

提出的方法

• 开发一个基于 1 ms 帧的能量与 RMS–能量联合检测器用于肠鸣音事件。
• 将能量转换为 dB 并使用基线统计进行归一化以检测起始/结束。
• 使用预训练的 AST 和 Wav2Vec 2.0 模型并在队列特异性微调，将检测到的片段分类为四种模式。
• 在 Healthy、Patient 和 Mixed 测试集上评估分类器，以评估同领域与跨领域的性能。
• 执行后处理以合并相邻片段并可选填补空隙以生成连贯的 BS 序列。
• 提供带专家参与验证的自动标注工作流，以量化标注效率。

实验结果

研究问题

• RQ1 自动化流水线是否能在健康、患者等多样模式（SB、MB、CRS、HS）下可靠检测肠鸣音事件，且基于能量与 RMS 的特征？
• RQ2 在对健康与患者数据进行微调后，预训练模型（AST 与 Wav2Vec 2.0）在分类肠鸣音模式上的表现如何？
• RQ3 队列特异性训练是否提升健康受试者与胃肠疾病患者的分类性能与泛化能力？
• RQ4 在半自动化工作流中，使用带专家纠正的 Auto-Annotation 能否显著节省时间并提升标注质量？

主要发现

• AST 在训练/测试子组中始终优于 Wav2Vec 2.0。
• 仅健康的 AST 在健康测试数据上达到 ACC 0.97 和 AUROC 0.98。
• 仅患者的 AST 在患者测试数据上达到 ACC 0.96 和 AUROC 0.98。
• 健康+患者组合的 AST 在混合测试数据上达到 ACC 0.94 和 AUROC 0.98。
• Auto-annotation 将人工标注时间减少约 70%，仅对 <12% 的片段需要专家纠正。
• Auto-annotation 保留了类的流行顺序并捕捉了 SB/CRS/HS 的典型时长范围，对 MB 时长有少量低估。