Grangeat Pierre, Gharbi Sadok, Koenig Anne, Comsa Maria-Paula, Accensi Marc, Grateau Henri, Ghaith Abdallah, Chacaroun Samarmar, Doutreleau Stephane, Verges Samuel
Annu Int Conf IEEE Eng Med Biol Soc. 2020 Jul;2020:4640-4643. doi: 10.1109/EMBC44109.2020.9175876.
The development of wearable devices for healthcare monitoring is of primary interest, in particular for homecare applications. But it is challenging to develop an evaluation framework to test and optimize such a device by following a non-invasive protocol. As well established reference devices do exist for capnometry, we propose a protocol to evaluate and compare the performance of the transcutaneous carbon dioxide monitoring wristband that we develop. We present here this protocol, the signal processing pipeline and the data analysis based on signal alignment and intercorrelation study, and the first results on a cohort of 13 healthy subjects. This test allows demonstrating the influence of the device response time and of the carbon dioxide content in the ambient air.Clinical Relevance-The protocol described here allows to test and optimize the new device in clinical conditions simulating hypo and hypercapnia variations on a subject at rest, as it would be the case at home to monitor the health status of chronic respiratory patients, and to compare the performances with reference devices. A strong intercorrelation greater than 0.8 has been observed in 5 healthy subjects out of 13 and factors influencing the intercorrelation are suggested.
用于医疗监测的可穿戴设备的开发备受关注,尤其是在家庭护理应用方面。但是,按照非侵入性方案开发一个评估框架来测试和优化此类设备具有挑战性。由于已经存在用于二氧化碳监测的成熟参考设备,我们提出了一种方案来评估和比较我们开发的经皮二氧化碳监测腕带的性能。我们在此展示该方案、信号处理流程以及基于信号对齐和互相关研究的数据分析,以及对13名健康受试者队列的初步结果。该测试能够证明设备响应时间和环境空气中二氧化碳含量的影响。临床意义——这里描述的方案允许在模拟静息受试者低碳酸血症和高碳酸血症变化的临床条件下测试和优化新设备,就像在家中监测慢性呼吸疾病患者健康状况那样,并与参考设备比较性能。在13名健康受试者中有5名观察到大于0.8的强互相关性,并提出了影响互相关性的因素。