一个半监督的算法来提高众包数据集的一致性:COVID-19关于呼吸系统疾病分类的案例研究
Lara Orlandic1, Tomas Teijeiro2, David Atienza1
1Embedded Systems Laboratory (ESL), EPFL, Lausanne, Switzerland.
Computer methods and programs in biomedicine
|August 20, 2023
概括
半监督学习 (SSL) 通过提高COVID-19检测和咳特征的数据一致性来增强咳音频分类. 这种方法汇总了专家知识,创造了一个更可靠的数据集,用于训练诊断模型.
科学领域:
- 医疗声学和信号处理
- 机器学习用于医疗保健
- 呼吸系统疾病诊断诊断 呼吸系统疾病诊断
背景情况:
- 咳音频分类显示了查COVID-19等呼吸系统疾病的前景.
- 众包咳数据,就像COUGHVID数据集一样,由于从传染病患者收集数据的风险至关重要.
- 数据集中的专家注释可能会遭受错误标签和专家间的分歧.
研究的目的:
- 使用半监督学习 (SSL) 提高COUGHVID数据集的标签一致性.
- 为了提高COVID-19与健康咳的分类准确性,咳类型 (湿/干) 和严重程度.
- 为培训咳分类器生成可靠的增强数据集.
主要方法:
- 应用SSL专家知识聚合技术,以解决COUGHVID数据集中的标签不一致性和稀疏性.
- 利用了音频信号处理和可解释的机器学习模型.
- 确定了重新标记的音频样本的子样本,用于训练或增强咳分类器.
主要成果:
- 重新标记的数据显示,类间特征分离性显著更高 (COVID-19的3倍与健康的11.3倍,类型的11.3倍,严重程度的5.1倍).
- 重新标记的数据中的扩大光谱差异导致健康和COVID-19咳之间的不同功率光谱密度 (1-1.5kHz范围).
- 在重新标记的数据集上训练的COVID-19分类器获得了0.797.7的AUC.
结论:
- 引入了一种新的SSL专家知识聚合技术,用于咳声音分类.
- 展示了一种可解释的方法来结合多专家的医学知识,产生一致和丰富的数据.
- 重新标记的数据集为改进咳分类任务提供了坚实的基础.
相关概念视频
01:17Classification of Illness
7.6K
The meaning of illness is individualized to each person who experiences an alteration in health. In contrast, disease is a medical term indicating a pathological change in the structure and function of the body or mind. It is a condition that has specific symptoms and boundaries.
An illness is a response to a disease in which the person's level of functioning is changed compared with a previous level. The general classification of illness includes acute and chronic.
Acute illness is severe...
An illness is a response to a disease in which the person's level of functioning is changed compared with a previous level. The general classification of illness includes acute and chronic.
Acute illness is severe...
7.6K
01:31Common Respiratory Disorders
667
Respiratory disorders, a prevalent health concern globally, are generally divided into two primary categories: upper and lower respiratory tract disorders. The categorization is based on the area of the respiratory system they affect.
Upper respiratory disorders impact the airways above the vocal cords, encompassing areas like the nose, sinuses, and throat. Various conditions fall under this category, including the common cold and allergic rhinitis. These disorders can stem from several causes,...
Upper respiratory disorders impact the airways above the vocal cords, encompassing areas like the nose, sinuses, and throat. Various conditions fall under this category, including the common cold and allergic rhinitis. These disorders can stem from several causes,...
667
01:11Residuals and Least-Squares Property
7.4K
The vertical distance between the actual value of y and the estimated value of y. In other words, it measures the vertical distance between the actual data point and the predicted point on the line
If the observed data point lies above the line, the residual is positive, and the line underestimates the actual data value for y. If the observed data point lies below the line, the residual is negative, and the line overestimates the actual data value for y.
The process of fitting the best-fit...
If the observed data point lies above the line, the residual is positive, and the line underestimates the actual data value for y. If the observed data point lies below the line, the residual is negative, and the line overestimates the actual data value for y.
The process of fitting the best-fit...
7.4K