一个深度学习模型,用于多类音频分类的声病理在办公室强光镜
Yeo E Kim1, Maria Dobko1, Haomiao Li2
1Sean Parker Institute for the Voice, Department of Otolaryngology-Head and Neck Surgery, Weill Cornell Medicine, New York, New York, U.S.A.
The Laryngoscope
|February 5, 2025
概括
使用声学视频中的语音数据的深度学习模型在区分健康的声 (VF) 与单边 (UVFP) 和VF病变方面表现出适度的准确性,但在多类分类和外部验证方面遇到了困难.
科学领域:
- 耳鼻喉科 耳鼻喉科 耳鼻喉科
- 人工智能在医学中的应用
- 语音科学 语言科学
背景情况:
- 视频喉镜是可视化声 (VF) 动态的一个关键工具.
- 在诊断和治疗中,区分健康的VF,单侧 (UVFP) 和VF病变至关重要.
- 使用语音分析自动分类VF条件具有提高诊断准确性的潜力.
研究的目的:
- 开发和验证深度学习分类器,以区分三个VF状态:健康 (HVF),UVFP和VF病变.
- 评估二元 (HVF与病理) 和多类 (HVF,UVFP,病变) 分类模型的性能.
- 在独立的外部数据集上评估模型的概括性.
主要方法:
- 从105名UVFP,63名VF病变和41名HVF患者的视频中提取了语音数据.
- 音频样本被转换为Mel谱图,并用于训练ResNet18深度学习模型.
- 模型被训练为二进制和多类分类,并被内部和外部验证.
主要成果:
- 二元分类器在测试组中取得了更高的性能 (精度83%,F1得分0.90) 比多类分类器 (精度40%,F1得分0.36).
- 外部验证显示,这两种模型的性能都下降了,二进制分类器达到63%的准确性和0.48 F1-score.
- 多类分类器在外部数据集上表现不佳 (准确率为35%,F1得分为0.25).
结论:
- 深度学习模型可以将健康和病态的声状况与光视频语音数据区分开来,准确度中等.
- 多类分类和外部验证显著降低了模型性能,这表明在概括方面存在挑战.
- 压缩镜录音中的语音数据可能对复杂的VF条件具有有限的独立诊断价值;需要进一步研究.
相关概念视频
Classification of Illness
7.3K
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...
7.3K
Ultrasound II: Endoscopic Ultrasound and FibroScan
66
Endoscopic Ultrasound (EUS) and FibroScan are valuable diagnostic tools in gastroenterology and hepatology, each with specific applications and techniques.
Endoscopic Ultrasound (EUS):
Endoscopic Ultrasound (EUS):
66
Classification of Epithelial Tissues: Stratified Epithelium
8.7K
Stratified epithelium consists of several stacked layers of cells. They provide the durability to withstand constant physical and chemical attacks. Stratified epithelium is named after the shape of the most apical layer of cells. Stratified squamous epithelium is the most common type found in the human body. In this tissue, the apical cells are squamous, whereas the basal layer contains either columnar or cuboidal cells. The basal cells divide to form new daughter cells, which gradually become...
8.7K
Endoscopic Studies I: Bronchoscopy and Thoracoscopy
172
Endoscopy is a non-surgical medical technique used to examine a person's internal organs and vessels. This lesson will focus on two types of endoscopic studies: bronchoscopy and thoracoscopy.
Bronchoscopy
Description
Bronchoscopy is a procedure that involves direct visualization of the larynx, trachea, and bronchi for diagnostic and therapeutic purposes. A flexible fiber optic or rigid bronchoscope is used to carry out the procedure. The fiber-optic bronchoscope is more frequently used due...
Bronchoscopy
Description
Bronchoscopy is a procedure that involves direct visualization of the larynx, trachea, and bronchi for diagnostic and therapeutic purposes. A flexible fiber optic or rigid bronchoscope is used to carry out the procedure. The fiber-optic bronchoscope is more frequently used due...
172
Respiratory System Abnormal Finding II: Palpation and Auscultation
258
In assessing respiratory abnormalities, palpation and auscultation are critical tools for detecting and interpreting various pathophysiological changes. These techniques provide insight into underlying disorders by evaluating tactile sensations and sounds produced by the respiratory system.
Palpation Findings
During a respiratory assessment, palpation can reveal several vital abnormalities:
Palpation Findings
During a respiratory assessment, palpation can reveal several vital abnormalities:
258
Auditory Pathway
4.6K
Auditory pathways constitute the complex neural circuits responsible for transmitting and interpreting auditory information from the peripheral auditory system to the brain. Sound waves are initially captured by the outer ear, funneled through the ear canal, and reach the tympanic membrane (eardrum). These vibrations are transmitted via the middle ear's ossicles to the inner ear's cochlea.
When viewed cross-sectionally, the cochlea reveals the scala vestibuli and scala tympani flanking...
When viewed cross-sectionally, the cochlea reveals the scala vestibuli and scala tympani flanking...
4.6K


