Real-time and early detection of minute changes in the functioning of the cardiovascular system is crucial for managing critically ill patients, such as newborns and older adults, and can significantly affect their outcomes. Heart rate variability (HRV) is the minute, yet normal, fluctuations between consecutive heartbeats, usually measured through the electrocardiogram (ECG). HRV is a well-established, quantitative, and noninvasive measure for assessing autonomic nervous system activity.
However, despite its high value for patient monitoring in clinical practice, accurate real-time HRV monitoring often faces two fundamental challenges. The first is the high inter-individual variability of HRV indices that depend on the patients' age and sex. Conventional monitoring systems usually rely on fixed, population-based thresholds for alerts, resulting in more false positives or negatives and misrepresentation of the results. The second challenge is the frequent data contamination from procedural artifacts that are inherent in clinical settings, such as patient movement, intense emotions, or nursing care. These artifacts cause non-physiological fluctuations in the readings, thereby compromising the reliability of the analysis. In addition, conventional frameworks lack the ability to simultaneously visualize long-term trends and short-term fluctuations across multiple HRV indices, limiting clinicians' understanding of subtle and evolving changes in a patient's state. While existing HRV tools are valuable in research and offline analysis, they are not designed to perform efficiently in real-world clinical settings.
There is a need for a computational framework that can provide robust, personalized, and artifact-resilient bedside HRV monitoring in real-time. Consequently, researchers from Fujita Health University (FHU) developed and validated a novel computational framework for reliable, real-time, and personalized HRV monitoring that is suitable for clinical applications. Their findings were made available online on January 29, 2026, and will be published in Volume 50, Issue 1 of the Journal of Medical Systems on December 1, 2026.
Dr. Takashi Nakano, Associate Professor at FHU's Department of Computational Biology, along with Dr. Masayuki Fujino, Dr. Masafumi Miyata, and Dr. Tetsushi Yoshikawa, created the framework by integrating an adaptive, personalized algorithm with a practical, workflow-integrated artifact management mechanism and implemented it in a cross-platform software tool called 'CODO Monitor.' "CODO Monitor is clinically oriented for real-time, long-duration ECG/HRV analysis that visualizes both short-term fluctuations and long-term trends simultaneously. It also enables patient-specific, adaptive alerts and improves robustness through artifact exclusion and event annotation," explains Prof. Nakano.
The framework is built on four key components. The first is an adaptive alerting algorithm that dynamically calculates personalized thresholds for HRV indices based on each patient's unique data, which reduces false alarms and improves clinical decision-making. The second is an artifact management mechanism that allows clinicians to manually flag and exclude artifact-prone periods for accurate HRV analysis. Next is the simultaneous analysis and visualization of both time- and frequency-domain indices, providing a comprehensive view of a patient's autonomic function. Finally, it utilizes a multi-scale visualization approach that provides short-term fluctuation readings and long-term trend analysis of HRV indices, facilitating a more unified clinical interpretation.
The framework was validated using open-access ECG databases for both pediatric and adult recordings and synthetic noise-contaminated signals to confirm its accuracy and the robustness of R-wave detection. Furthermore, it was operationally validated at the bedside using ECG data from 24 newborn patients. Additionally, the framework boasts cross-platform compatibility with both Windows and macOS systems.
The groundbreaking CODO Monitor represents a substantial improvement over conventional fixed-threshold systems, offering the potential to increase the specificity of alerts, reduce alert fatigue among clinical staff, and ultimately make real-time monitoring more clinically meaningful.
Envisioning the possibilities that this innovative tool opens up, Prof. Nakano says, "Our system could support real-time HRV monitoring in bedside clinical settings, enabling early detection and intervention. Moreover, the personalized alerting feature can reduce 'one-size-fits-all' alarm issues and improve day-to-day monitoring workflows, allowing for more individualized management strategies. Over time, this could contribute to safer patient care and better outcomes in both neonatal and critical care settings. Finally, the high-quality HRV data gathered may also accelerate research that clarifies links between autonomic function, HRV metrics, and disease, leading to new predictive markers and monitoring standards."
