Scholarlink

<p>Cardiovascular diseases, particularly hypertension, remain a major global health burden, highlighting the need for accurate and accessible blood pressure (BP) monitoring. Cuffless BP measurement (BPM) based on pulse wave propagation methods (PWPM), including pulse arrival time (PAT), pulse transit time (PTT), and pulse wave velocity (PWV), has attracted increasing research interest. This review comprises two components. First, a narrative review of studies published up to June 2025 examines sensing technologies, mathematical models, and validation protocols used in PWPM-based BPM. Second, a statistical re-evaluation of 22 studies published between 2015 and 2025 was conducted using the Credence of Device Acceptability (CDA) and the Probability of Tolerable Error (PTE), grounded in the statistical principles of ISO 81060-2. Accuracy varied widely across physiological conditions, sensing technologies, and study designs, with no single approach demonstrating consistent superiority. The re-evaluation provided a more stringent assessment of performance: only five studies achieved CDA values exceeding 0.95 for both systolic and diastolic BP. Overall, diastolic BP estimation demonstrated superior accuracy compared with systolic BP. Incorporating physiological indices such as arterial compliance and sympathetic activity may improve the robustness and accuracy of BP estimation models. While machine learning shows promise for enhanced feature extraction, calibration tolerance and real-world reliability remain critical challenges. Importantly, the evaluation and development of cuffless BPM technologies should align with validation standards appropriate to the intended application. We recommend that future early-stage studies apply the CDA and PTE framework as supportive accuracy metrics to better assess methodological performance and inform device development and validation.</p><p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Advancements and challenges in blood pressure monitoring using pulse wave propagation: a comprehensive review and ISO 81060-2 based statistical analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月7日

查看全文
  • Yang Yu,
  • Andrew Lowe

摘要

Cardiovascular diseases, particularly hypertension, remain a major global health burden, highlighting the need for accurate and accessible blood pressure (BP) monitoring. Cuffless BP measurement (BPM) based on pulse wave propagation methods (PWPM), including pulse arrival time (PAT), pulse transit time (PTT), and pulse wave velocity (PWV), has attracted increasing research interest. This review comprises two components. First, a narrative review of studies published up to June 2025 examines sensing technologies, mathematical models, and validation protocols used in PWPM-based BPM. Second, a statistical re-evaluation of 22 studies published between 2015 and 2025 was conducted using the Credence of Device Acceptability (CDA) and the Probability of Tolerable Error (PTE), grounded in the statistical principles of ISO 81060-2. Accuracy varied widely across physiological conditions, sensing technologies, and study designs, with no single approach demonstrating consistent superiority. The re-evaluation provided a more stringent assessment of performance: only five studies achieved CDA values exceeding 0.95 for both systolic and diastolic BP. Overall, diastolic BP estimation demonstrated superior accuracy compared with systolic BP. Incorporating physiological indices such as arterial compliance and sympathetic activity may improve the robustness and accuracy of BP estimation models. While machine learning shows promise for enhanced feature extraction, calibration tolerance and real-world reliability remain critical challenges. Importantly, the evaluation and development of cuffless BPM technologies should align with validation standards appropriate to the intended application. We recommend that future early-stage studies apply the CDA and PTE framework as supportive accuracy metrics to better assess methodological performance and inform device development and validation.