TL;DR
This paper introduces a novel biofeedback respiratory rehabilitation training system that leverages virtual reality (VR) to make the process more engaging and effective. By visualizing respiratory data in real-time within interactive VR environments, the system aims to improve lung function and patient adherence compared to traditional, often monotonous, rehabilitation methods.
What Problem Does This Paper Address?
Traditional respiratory rehabilitation training, while essential for patients with conditions like Chronic Obstructive Pulmonary Disease (COPD), suffers from several drawbacks. The process is often described as "simple and boring," leading to low patient engagement and adherence. Furthermore, it lacks clear visualization and quantification of respiratory data, making it difficult for patients to understand their progress and for clinicians to objectively assess improvements in lung function. This can hinder the effectiveness of rehabilitation and impact patients' quality of life.
Methods
The core of this research is a biofeedback respiratory rehabilitation training system that integrates virtual reality technology with respiratory sensors. The system architecture involves:
- Respiratory Data Collection: A vital capacity sensor (HKF-20C) is used to measure various respiratory parameters such as exhalation time, vital capacity (FVC), peak flow rate (FEF), and different flow rates (MEF25, MEF50, MEF75), as well as forced expiratory volume in one, two, and three seconds (FEV1, FEV2, FEV3).
- VR Environment: The system utilizes HTC Vive Pro2 VR equipment, including a head-mounted display and controllers, to create immersive virtual scenes. The Unity 3D engine is used to develop these scenes.
- Biofeedback Integration: Respiratory data collected by the sensor is transmitted to a PC and then used to drive changes within the VR environment. For example, in a "blowing candles" scenario, the intensity of the user's breath determines if the virtual candle is extinguished.
- Data Analysis and Visualization: Respiratory data is analyzed to extract characteristic points and parameters. This data is then visualized in a digital format within the VR environment, providing patients with immediate feedback on their performance.
- Interactive Training Scenarios: Two specific virtual scenes were designed: "blowing candles" and "blowing dandelions." These scenarios are intended to make the training more active and interesting, encouraging deeper breathing and improved lung function.
Datasets
Not discussed in the paper. The study relies on real-time data collection from participants using the developed system.
Key Insights and Ideas
- Bridging the Gap Between Data and Experience: The system's primary innovation is its ability to translate raw respiratory data into tangible, visual feedback within an engaging VR environment. This makes abstract physiological metrics understandable and actionable for patients.
- Gamification for Adherence: By designing interactive VR scenarios like "blowing candles" or "dandelions," the system transforms a potentially tedious rehabilitation task into a more enjoyable experience, which is crucial for long-term adherence.
- Quantifiable Progress: The system provides objective, quantified respiratory data before and after training, allowing for a more precise evaluation of rehabilitation effectiveness than traditional methods.
- Reduced Physical Burden: Unlike some VR rehabilitation systems that require heavy physiological sensors, this system aims for a more intuitive and portable setup, potentially reducing physical burdens on patients.
Key Results
The study involved 10 participants, divided into a control group (traditional training) and an observation group (VR-based training).
- Improved Lung Function: Both groups showed improvements in respiratory data after training. However, the observation group using the VR system demonstrated significantly better training effects. Specifically, the first-second vital capacity (FEV1) of the observation group increased by 5% compared to a 2% increase in the control group.
- Enhanced User Experience: The observation group reported higher scores for immersion, participation, and fun. While initially experiencing some boredom, their scores for boredom decreased over time as they became more accustomed to the system, unlike the control group whose boredom scores increased.
- Visualization Effectiveness: The paper highlights that visualizing respiratory data in a digital and visual form helps trainees understand their shortcomings and improve their breathing training.
Limitations and Future Work
- Data Transmission Interference: The authors acknowledge that respiratory equipment can generate Gaussian white noise and random noise during data transmission, affecting accuracy.
- Sensor Noise and Heat Loss: During data collection, the equipment can produce noise and heat loss, potentially impacting the accuracy of vital capacity measurements.
- Future Work: The authors suggest focusing on further improving and perfecting the respiratory data collection process and enhancing the accuracy of respiratory data transmission. This includes optimizing the system for better accuracy and potentially exploring more advanced VR scenarios.
Summary updated