PDF(Pubmed)
Abstract：
UNASSIGNED: Advancements in medical technologies have led to the development of contact-free methods of haemodynamic monitoring such as remote photoplethysmography (rPPG). rPPG uses video cameras to interpret variations in skin colour related to blood flow, which are analysed to generate vital signs readings. rPPG potentially ameliorates problems like fretfulness and fragile skin contact associated with conventional probes in children. While rPPG has been validated in adults, no prior validation has been performed in children.
UNASSIGNED: A two-phased prospective cross-sectional single-centre study was conducted from January to April 2023 to evaluate the feasibility, acceptability, and accuracy of obtaining heart rate (HR), respiratory rate (RR) and oxygen saturation (SpO2) using rPPG in children, compared to the current standard of care. In Phase 1, we recruited patients ≤16 years from the neonatal and paediatric wards. We excluded preterm neonates with gestational age <35 weeks and newborns <24 hours old. The rPPG webcam was positioned 30 cm from the face. After 1 minute of facial scanning, readings generated were compared with pulse oximetry for HR and SpO2, and manual counting for RR. Correlation and Bland-Altman analyses were performed. In Phase 2, we focused on the population in whom there was potential correlation between rPPG and the actual vital signs.
UNASSIGNED: Ten neonates and 28 children aged 5 to 16 years were recruited for Phase 1 (765 datapoints). All patients were haemodynamically stable and normothermic. Patients and caregivers showed high acceptability to rPPG. rPPG values were clinically discrepant for children <10 years. For those ≥10 years, moderate correlation was observed for HR, with Spearman\'s correlation coefficient (Rs) of 0.50 [95% confidence intervals (CI): 0.42, 0.57]. We performed Phase 2 on 23 patients aged 12 to 16 years (559 datapoints). Strong correlation was observed for HR with Rs=0.82 (95% CI: 0.78, 0.85). There was weak correlation for SpO2 and RR (Rs=-0.25 and -0.02, respectively).
UNASSIGNED: Our study showed that rPPG is acceptable and feasible for neonates and children aged 5 to 16 years, and HR values in older children aged 12 to 16 years correlated well with the current standard. The rPPG algorithms need to be further refined for younger children, and for obtaining RR and SpO2 in all children. If successful, rPPG will provide a viable contact-free alternative for assessing paediatric vital signs, with potential use in remote monitoring and telemedicine.
UNASSIGNED: A two-phased prospective cross-sectional single-centre study was conducted from January to April 2023 to evaluate the feasibility, acceptability, and accuracy of obtaining heart rate (HR), respiratory rate (RR) and oxygen saturation (SpO2) using rPPG in children, compared to the current standard of care. In Phase 1, we recruited patients ≤16 years from the neonatal and paediatric wards. We excluded preterm neonates with gestational age <35 weeks and newborns <24 hours old. The rPPG webcam was positioned 30 cm from the face. After 1 minute of facial scanning, readings generated were compared with pulse oximetry for HR and SpO2, and manual counting for RR. Correlation and Bland-Altman analyses were performed. In Phase 2, we focused on the population in whom there was potential correlation between rPPG and the actual vital signs.
UNASSIGNED: Ten neonates and 28 children aged 5 to 16 years were recruited for Phase 1 (765 datapoints). All patients were haemodynamically stable and normothermic. Patients and caregivers showed high acceptability to rPPG. rPPG values were clinically discrepant for children <10 years. For those ≥10 years, moderate correlation was observed for HR, with Spearman\'s correlation coefficient (Rs) of 0.50 [95% confidence intervals (CI): 0.42, 0.57]. We performed Phase 2 on 23 patients aged 12 to 16 years (559 datapoints). Strong correlation was observed for HR with Rs=0.82 (95% CI: 0.78, 0.85). There was weak correlation for SpO2 and RR (Rs=-0.25 and -0.02, respectively).
UNASSIGNED: Our study showed that rPPG is acceptable and feasible for neonates and children aged 5 to 16 years, and HR values in older children aged 12 to 16 years correlated well with the current standard. The rPPG algorithms need to be further refined for younger children, and for obtaining RR and SpO2 in all children. If successful, rPPG will provide a viable contact-free alternative for assessing paediatric vital signs, with potential use in remote monitoring and telemedicine.
摘要：
■医疗技术的进步导致了非接触式血流动力学监测方法的发展，例如远程光电体积描记术（rPPG）。rPPG使用摄像机来解释与血流相关的肤色变化，对其进行分析以生成生命体征读数。rPPG可能会改善与儿童常规探针相关的烦躁和脆弱的皮肤接触等问题。虽然rPPG已经在成年人中得到验证，之前未对儿童进行过验证.
■在2023年1月至4月进行了两阶段的前瞻性横截面单中心研究，以评估可行性，可接受性,和获得心率（HR）的准确性，使用rPPG的儿童呼吸频率（RR）和氧饱和度（SpO2），与目前的护理标准相比。在第1阶段，我们从新生儿和儿科病房招募了≤16岁的患者。我们排除了胎龄<35周的早产儿和<24小时的新生儿。rPPG网络摄像头位于距离面部30cm处。经过1分钟的面部扫描,将产生的读数与脉搏血氧饱和度测定HR和SpO2以及手动计数RR进行比较.进行相关性和Bland-Altman分析。在第2阶段，我们专注于rPPG与实际生命体征之间存在潜在相关性的人群。
■招募了10名新生儿和28名5至16岁的儿童进行第一阶段（765个数据点）。所有患者血流动力学稳定，体温正常。患者和护理人员对rPPG表现出很高的可接受性。<10岁儿童的rPPG值在临床上存在差异。对于那些≥10年的人，观察到HR的中度相关性，Spearman相关系数（Rs）为0.50[95%置信区间（CI）：0.42，0.57]。我们对23名年龄在12至16岁(559个数据点)的患者进行了第二阶段。观察到HR的强相关性,Rs=0.82(95%CI:0.78,0.85)。SpO2和RR之间存在弱相关性（Rs分别为-0.25和-0.02）。
■我们的研究表明，rPPG对于5至16岁的新生儿和儿童是可以接受和可行的，年龄在12至16岁的大龄儿童的HR值与现行标准有很好的相关性。rPPG算法需要针对年幼的孩子进一步完善，并在所有儿童中获得RR和SpO2。如果成功,rPPG将为评估儿科生命体征提供可行的非接触替代方案，具有远程监控和远程医疗的潜在用途。
■在2023年1月至4月进行了两阶段的前瞻性横截面单中心研究，以评估可行性，可接受性,和获得心率（HR）的准确性，使用rPPG的儿童呼吸频率（RR）和氧饱和度（SpO2），与目前的护理标准相比。在第1阶段，我们从新生儿和儿科病房招募了≤16岁的患者。我们排除了胎龄<35周的早产儿和<24小时的新生儿。rPPG网络摄像头位于距离面部30cm处。经过1分钟的面部扫描,将产生的读数与脉搏血氧饱和度测定HR和SpO2以及手动计数RR进行比较.进行相关性和Bland-Altman分析。在第2阶段，我们专注于rPPG与实际生命体征之间存在潜在相关性的人群。
■招募了10名新生儿和28名5至16岁的儿童进行第一阶段（765个数据点）。所有患者血流动力学稳定，体温正常。患者和护理人员对rPPG表现出很高的可接受性。<10岁儿童的rPPG值在临床上存在差异。对于那些≥10年的人，观察到HR的中度相关性，Spearman相关系数（Rs）为0.50[95%置信区间（CI）：0.42，0.57]。我们对23名年龄在12至16岁(559个数据点)的患者进行了第二阶段。观察到HR的强相关性,Rs=0.82(95%CI:0.78,0.85)。SpO2和RR之间存在弱相关性（Rs分别为-0.25和-0.02）。
■我们的研究表明，rPPG对于5至16岁的新生儿和儿童是可以接受和可行的，年龄在12至16岁的大龄儿童的HR值与现行标准有很好的相关性。rPPG算法需要针对年幼的孩子进一步完善，并在所有儿童中获得RR和SpO2。如果成功,rPPG将为评估儿科生命体征提供可行的非接触替代方案，具有远程监控和远程医疗的潜在用途。
