Wearable Technology for Predicting Postoperative Pulmonary Complications

Postoperative Pulmonary Complications (PPCs) encompass a variety of lung-related issues that can arise after surgeries, especially after major operations like cardiac valve surgery. These complications range from pneumonia and respiratory failure to atelectasis (the partial or complete collapse of the lung), bronchospasm (airway constriction), pleural effusion (fluid accumulation around the lungs), and pulmonary embolism (a lung blood clot).

In the rapidly advancing field of healthcare technology, wearable devices are emerging as key tools for predicting and mitigating postoperative complications. Highlighting this trend, a recent exploratory study by Wang et al. explores the potential of wearables in predicting PPCs following cardiac valve surgery. Their study, "A wearable six-minute walk-based system to predict postoperative pulmonary complications after cardiac valve surgery: an exploratory study," examines how wearable technology can be integrated into clinical settings.

The development of wearable devices has surged in recent years. These devices, armed with sensors and sophisticated monitoring capabilities, play a crucial role in tracking a wide range of health metrics. The study underscores the significance of combining wearable technology with clinical practices, marking a critical evolution in the application of these devices.

A key goal of the study was to develop a predictive model for PPCs by monitoring respiratory physiological parameters during the preoperative 6-Minute Walk Test (6MWT). The SensEcho was the chosen device for this study, capable of collecting real-time respiratory data during physical activities.

The SensEcho device features a signal quality assessment algorithm to evaluate the quality of ECG and respiratory signals. While the accuracy of signal quality classification is notable, challenges like signal quality misjudgment persist, mainly because most Signal Quality Assessments (SQAs) are not designed for the daily usage of wearables. The most effective methods for ensuring accurate readings involve supervised machine learning models. The SensEcho, designed as a vest, provides a single-lead ECG signal, chest and abdominal respiratory signals through respiratory inductive plethysmography (RIP) technology, and triaxial acceleration signals. It also communicates with other wearable devices, such as oximeters and blood pressure monitors, and its battery supports at least 24 hours of continuous monitoring.

Interference in the form of Electromyography (EMG) noise can degrade the quality of an ECG recording. While power line interference can be relatively easily filtered out, unpredictable EMG interference, like that caused by muscle contractions or vibrations from speaking, poses a greater challenge.

The significance of Electrocardiogram (ECG) monitoring in diagnosing, preventing, and rehabilitating cardiovascular disease (CVD) cannot be overstated. With advancements in the Internet of Things (IoT), big data, cloud computing, and artificial intelligence (AI), wearable ECG technology is gaining increasing importance. This is particularly relevant as our population ages, necessitating a shift in CVD diagnostic methods. The use of AI to assist in the clinical analysis of long-term ECGs is emerging as a crucial area of focus, improving early detection and prediction capabilities for CVD. This integration of intelligent wearable ECG monitoring, leveraging both edge and cloud computing, underscores the need for precise implementation in medical contexts. Recent studies have highlighted the progress of AI in ECG analysis, illustrating the potential of AI-assisted wearable ECG in clinical settings and pointing to future opportunities and challenges in ensuring the reliability and effectiveness of this technology.

The study by Wang et al included 53 patients with cardiac valve diseases from West China Hospital's Department of Cardiovascular Surgery, Sichuan University. These patients were grouped based on whether they developed PPCs postoperatively. The 6MWT, a standard exercise capacity metric, served as the basis for continuous monitoring with the SensEcho.

Analysis of ongoing respiratory physiological parameters showed significant differences between patients who experienced PPCs and those who did not. Key indicators included respiratory rate and oxygen saturation. These findings indicate a promising direction in using wearables to predict PPCs in cardiac valve surgery patients, offering a novel model for incorporating wearables into clinical practice.

Integrating wearable devices into preoperative assessments heralds new possibilities for proactive healthcare. Continuous monitoring of physiological parameters allows clinicians to detect potential complications earlier, facilitating timely interventions. The findings of this study add to the growing evidence for incorporating wearables into clinical care, highlighting the potential for personalized, data-driven medicine.

As technology continues to progress, the seamless incorporation of wearables into clinical procedures promises to enhance patient outcomes, especially in the critical postoperative phase.

REFERENCES

Wang Y, Wang J, Zhang J, Luo Z, Guo Y, Zhang Z, Yu P. [A wearable six-minute walk-based system to predict postoperative pulmonary complications after cardiac valve surgery: an exploratory study]. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2023 Dec 25;40(6):1117-1125. Chinese. doi: 10.7507/1001-5515.202305007. PMID: 38151934; PMCID: PMC10753314.

Xu H, Li P, Yang Z, Liu X, WaWang X, Li Q, Ma C, Zhang S, Lin Y, Li J, Liu C. [Artificial intelligence in wearable electrocardiogram monitoring]. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2023 Dec 25;40(6):1084-1092. Chinese. doi: 10.7507/1001-5515.202301032. PMID: 38151930; PMCID: PMC10753313.

Xu H, Li P, Yang Z, et al. Construction and Application of a Medical-Grade Wireless Monitoring System for Physiological Signals at General Wards. Journal of Medical Systems. 2020 Sep;44(10):182. DOI: 10.1007/s10916-020-01653-z. PMID: 32885290; PMCID: PMC7471584.