A digital biomarker of diabetes from smartphone-based vascular signals.
A digital biomarker of diabetes from smartphone-based vascular signals.
复制标题
基于智能手机血管信号的糖尿病数字生物标志物。
DOI：
10.1038/s41591-020-1010-5
复制
发表时间：
2020-10
期刊：
影响因子：
82.9
通讯作者：
Tison GH
中科院分区：
文献类型：
--
作者：
Avram R;Olgin JE;Kuhar P;Hughes JW;Marcus GM;Pletcher MJ;Aschbacher K;Tison GH
全球糖尿病负担正在迅速增加，从2019年的4.51亿人增加到2045年的6.93亿人。2型糖尿病的隐匿性发作推迟了诊断，增加了发病率。考虑到糖尿病的多因素血管效应，我们假设基于智能手机的光体积描记(PPG)可以为糖尿病提供一个广泛可用的数字生物标记物。在这里，我们开发了一个深度神经网络(DNN)来使用基于智能手机的PPG来检测流行的糖尿病，最初的队列为53,870人(“主要队列”)，然后在另一个由7,806人组成的单独队列(“当代队列”)和来自三个诊所的181名预期登记的个人(“临床队列”)中进行了验证。DNN的曲线下面积(AUC)在初级队列中为0.766(95%可信区间：0.750-0.782；敏感性75%，特异性65%)，在当代队列中为0.740(95%可信区间：0.723-0.758；敏感性81%，特异性54%)。当将DNN的输出称为DNN评分与年龄、性别、种族/民族和体重指数一起纳入回归分析时，AUC值为0.830，DNN评分仍然是糖尿病的独立预测因素。DNN在临床队列中的表现与在其他验证数据集中的表现相似。在HbA1c患者中，连续DNN评分与HbA1c(HbA1c)呈显著正相关(p≤0.001)。这些发现表明，基于智能手机的PPG提供了一种容易获得的、非侵入性的流行糖尿病的数字生物标记物。
The global burden of diabetes is rapidly increasing, from 451 million people in 2019 to 693 million by 2045. The insidious onset of type 2 diabetes delays diagnosis and increases morbidity. Given the multifactorial vascular effects of diabetes, we hypothesized that smartphone-based photoplethysmography (PPG) could provide a widely-accessible digital biomarker for diabetes. Here, we developed a deep neural network (DNN) to detect prevalent diabetes using smartphone-based PPG from an initial cohort of 53,870 individuals (the “Primary Cohort”), which was then validated in a separate cohort of 7,806 individuals (the “Contemporary Cohort”), and a cohort of 181 prospectively-enrolled individuals from three clinics (the “Clinic Cohort”). The DNN achieved an area under the curve (AUC) for prevalent diabetes of 0.766 in the Primary Cohort (95% confidence interval (CI): 0.750–0.782; sensitivity 75%, specificity 65%) and 0.740 in the Contemporary Cohort (95% CI: 0.723–0.758; sensitivity 81%, specificity 54%). When the output of the DNN, called the DNN Score, was included in a regression analysis alongside age, gender, race/ethnicity, and body mass index, the AUC was 0.830 and the DNN Score remained independently predictive of diabetes. The performance of the DNN in the Clinic Cohort was similar to that in other validation datasets. There was a significant and positive association between the continuous DNN Score and hemoglobin A1c (HbA1c) (p≤0.001) among those with HbA1c. These findings demonstrate that smartphone-based PPG provides a readily attainable, noninvasive digital biomarker of prevalent diabetes.
登录
查看更多内容
二手烟和心房颤动：健康EHEART研究的数据。
影响因子：
5.5
作者：
Dixit S;Pletcher MJ;Vittinghoff E;Imburgia K;Maguire C;Whitman IR;Glantz SA;Olgin JE;Marcus GM
通讯作者：
Marcus GM
影响因子：
16.2
作者：
Bertoni, AG;Anderson, GF;Brancati, FL
通讯作者：
Brancati, FL
影响因子：
5.1
作者：
Cho, N. H.;Shaw, J. E.;Malanda, B.
通讯作者：
Malanda, B.
影响因子：
1.9
作者：
Elgendi M
通讯作者：
Elgendi M
影响因子：
15.2
作者：
Avram, Robert;Tison, Geoffrey H.;Olgin, Jeffrey
通讯作者：
Olgin, Jeffrey