Abdominal aortic aneurysm monitoring via arterial waveform analysis: towards a convenient point-of-care device

Mohammad Yavarimanesh, Hao Min Cheng, Chen Huan Chen, Shih-Hsien Sung, Aman Mahajan, Rabih A. Chaer, Sanjeev G. Shroff, Jin Oh Hahn, Ramakrishna Mukkamala*

*此作品的通信作者

研究成果: Article同行評審

2 引文 斯高帕斯(Scopus)

摘要

Abdominal aortic aneurysms (AAAs) are lethal but treatable yet substantially under-diagnosed and under-monitored. Hence, new AAA monitoring devices that are convenient in use and cost are needed. Our hypothesis is that analysis of arterial waveforms, which could be obtained with such a device, can provide information about AAA size. We aim to initially test this hypothesis via tonometric waveforms. We study noninvasive carotid and femoral blood pressure (BP) waveforms and reference image-based maximal aortic diameter measurements from 50 AAA patients as well as the two noninvasive BP waveforms from these patients after endovascular repair (EVAR) and from 50 comparable control patients. We develop linear regression models for predicting the maximal aortic diameter from waveform or non-waveform features. We evaluate the models in out-of-training data in terms of predicting the maximal aortic diameter value and changes induced by EVAR. The best model includes the carotid area ratio (diastolic area divided by systolic area) and normalized carotid-femoral pulse transit time ((age·diastolic BP)/(height/PTT)) as input features with positive model coefficients. This model is explainable based on the early, negative wave reflection in AAA and the Moens-Korteweg equation for relating PTT to vessel diameter. The predicted maximal aortic diameters yield receiver operating characteristic area under the curves of 0.83 ± 0.04 in classifying AAA versus control patients and 0.72 ± 0.04 in classifying AAA patients before versus after EVAR. These results are significantly better than a baseline model excluding waveform features as input. Our findings could potentially translate to convenient devices that serve as an adjunct to imaging.

原文English
文章編號168
期刊npj Digital Medicine
5
發行號1
DOIs
出版狀態Published - 12月 2022

指紋

深入研究「Abdominal aortic aneurysm monitoring via arterial waveform analysis: towards a convenient point-of-care device」主題。共同形成了獨特的指紋。

引用此