Diagnosis of Peripheral Artery Disease Using Backflow Abnormalities in Proximal Recordings of Accelerometer Contact Microphone (ACM)
Diagnosis of Peripheral Artery Disease Using Backflow Abnormalities in Proximal Recordings of Accelerometer Contact Microphone (ACM)
复制标题
DOI：
10.1109/jbhi.2022.3218595
复制
发表时间：
2022-11
影响因子：
7.7
通讯作者：
Arash Shokouhmand;H. Wen;Samiha Khan;J. Puma;Amisha Patel;Philip Green;Farrokh Ayazi;Negar Tavassolian
中科院分区：
文献类型：
--
作者：
Arash Shokouhmand;H. Wen;Samiha Khan;J. Puma;Amisha Patel;Philip Green;Farrokh Ayazi;Negar Tavassolian
目的：开发一种准确的、非侵入性的方法，用于从心脏系统的加速度计接触式麦克风（ACM）记录诊断外周动脉疾病（PAD）。方法：首先从ACM记录中提取Mel频率倒谱系数（MFCC）。然后，提取的MFCC用于微调预训练的ResNet50网络，其中间层提供高抽象级别系数（HLACs）的流，这些系数可以提供关于PAD患者动脉阻塞引起的血压回流的信息。最后，将视觉转换器Transformer与特征提取层集成以检测PAD，并对严重程度进行分层。这种架构被称为多流驱动的视觉Transformer（MSPViT）。在74名PAD和21名健康受试者上评价了MSPViT的性能。结果如下：报告的二元分类的灵敏度、特异性、F1评分和曲线下面积（AUC）分别为99.45%、98.21%、99.37%和0.99，确保了PAD的准确检测。此外，MSPViT表明，将受试者分为健康、轻度PAD和重度PAD类别的平均灵敏度、特异性、F1评分和AUC分别为96.66%、97.34%、96.29%和0.96。计算轮廓分数以评估针对MSPViT的倒数第二层中的类别形成的聚类的可分性。0.66和0.81的平均轮廓得分分别在PAD检测和严重程度分类中表现出优异的聚类可分离性。结论：所实现的性能表明，近端ACM驱动框架可以取代最先进的PAD检测技术。意义：本研究为快速准确诊断PAD及其严重程度分层迈出了重要的一步。
Objective: The development of an accurate, non-invasive method for the diagnosis of peripheral artery disease (PAD) from accelerometer contact microphone (ACM) recordings of the cardiac system. Methods: Mel frequency cepstral coefficients (MFCCs) are initially extracted from ACM recordings. The extracted MFCCs are then used to fine-tune a pre-trained ResNet50 network whose middle layers provide streams of high-level-of-abstraction coefficients (HLACs) which could provide information on blood pressure backflow caused by arterial obstructions in PAD patients. A vision transformer is finally integrated with the feature extraction layer to detect PAD, and stratify the severity level. This architecture is coined multi-stream-powered vision transformer (MSPViT). The performance of MSPViT is evaluated on 74 PAD and 21 healthy subjects. Results: Sensitivity, specificity, F1 score, and area under the curve (AUC) of 99.45%, 98.21%, 99.37%, and 0.99, respectively, are reported for the binary classification which ensures accurate detection of PAD. Furthermore, MSPViT suggests average sensitivity, specificity, F1 score, and AUC of 96.66%, 97.34%, 96.29%, and 0.96, respectively, for the classification of subjects into healthy, mild-PAD, and severe-PAD classes. The silhouette score is calculated to assess the separability of clusters formed for classes in the penultimate layer of MSPViT. An average silhouette score of 0.66 and 0.81 demonstrate excellent cluster separability in PAD detection and severity classification, respectively. Conclusion: The achieved performance suggests that the proximal ACM-driven framework can replace state-of-the-art techniques for PAD detection. Significance: This study presents a fundamental step towards prompt and accurate diagnosis of PAD and stratification of its severity level.