Artificial Intelligence-Powered Remote Monitoring of Patients with Chronic Obstructive Pulmonary Disease
Chronic obstructive pulmonary disease (COPD) has emerged as the third leading cause of death globally, following heart disease and cancer. Acute exacerbations of COPD (AECOPD) significantly contribute to the morbidity and mortality associated with this condition. Early detection of deterioration and timely intervention can prevent AECOPD, reduce the severity of exacerbations, and minimize hospitalizations. Remote patient monitoring (RPM) has become an essential tool in both inpatient and outpatient care settings for COPD. The integration of artificial intelligence (AI) technologies into RPM systems has the potential to anticipate exacerbations and enable early therapeutic interventions, thereby improving patient outcomes.
AI-Enabled Technologies for Remote Monitoring of COPD Patients
Healthcare policies worldwide are increasingly advocating for telehealth-supported self-management of long-term conditions like COPD. Advances in sensor technology, miniaturized processors, and wireless data transmission have enabled the continuous assessment of environmental, physical, and physiological data without restricting patient activity. AI enhances the capabilities of these sensors, allowing for a more comprehensive understanding of patient conditions. This section explores the various AI-enabled technologies used in RPM for COPD patients, including body-worn and contactless sensors.
Physical Activity Monitoring
Physical activity monitoring is a critical component of RPM for COPD patients. Intelligent sensing technologies can continuously track body motions and detect various activities of daily living (ADL), such as sitting, walking, and sleeping. Monitoring physical activity levels helps identify the disease stage at which physical activity becomes limited, understand the relationship between physical activity and clinical characteristics, and drive innovative clinical procedures to predict patient risk and provide timely interventions.
Accelerometers are widely used sensors for monitoring physical activity. These small-scale micro-electro-mechanical system devices are integrated into consumer wearable devices like wristbands, smartwatches, and smartphones. One study demonstrated the feasibility of inferring physical activities by combining sensors available on a smartphone, such as accelerometers, gyroscopes, and gravity sensors, using machine-learning algorithms. This approach involves collecting a large dataset for training the classifier, extracting features from the data, and using a machine-learning algorithm to classify activities into categories like walking, running, sitting, standing, climbing stairs, and going downstairs. These experiments achieved an average accuracy of above 86%.
Camera-based sensors with action recognition techniques offer an unobtrusive alternative for remote tracking of COPD patients’ ADL. These sensors send live ADL video streams to processors, where algorithms like convolutional neural networks classify the activities. With proper training and model tuning, these models can achieve an accuracy of approximately 80%. Continuous detection of ADL can also aid in monitoring other diseases that affect ADL, such as Alzheimer’s disease and Parkinson’s disease.
Audio Symptoms Monitoring
Audio symptoms monitoring is another vital aspect of RPM for COPD patients. Recent studies have developed efficient solutions for monitoring symptoms like coughing and wheezing. One study presented a real-time low-power wireless respiratory monitoring system with cough detection. This system uses a microphone and an AI-based audio analysis algorithm to perform automated cough detection, achieving good results. Another study introduced four audio-only cough monitors using an artificial neural network to detect cough after signal processing.
Wheeze sound analysis has also been explored, with studies demonstrating that wheeze signals contain sufficient information for categorizing patients according to the severity of COPD. Machine learning algorithms like support vector machines have been implemented to build classifiers, achieving an average sensitivity of above 90% and an average accuracy of above 85%. AI-powered audio-based cough detection systems are increasingly applied in RPM and are becoming essential in studying other audio symptoms caused by COPD.
Environmental Sensors
Environmental sensors play a crucial role in RPM by monitoring factors like air quality, temperature, and humidity, which can significantly impact COPD patients. Various air quality sensors are available, most of which operate on the principle of an infrared light source with a photodetector measuring the light scattered by dust or haze particles in the air. These sensors can measure dust larger than 1 mm and provide particulate matter density readings.
The DHT22 sensor is widely used to monitor temperature and humidity. This low-cost sensor uses a capacitive humidity sensor and a thermistor to measure surrounding air and generate a digital signal on the data pin. It can achieve humidity readings from 0% to 100% with 2% to 5% precision and a temperature range from –40 to 80°C with ±0.5°C resolution.
Pulse Oximetry
Pulse oximetry is a non-invasive method for monitoring oxygen saturation in the blood, which is crucial for COPD patients. A wearable finger pulse oximeter is a thin clip-like device placed on a thin body part, such as the finger, ear lobe, or foot. The device operates based on the change in the amount of light absorbed during an arterial pulse. Two light sources in the visible red-light spectrum (660 nm) and infrared spectrum (940 nm) alternately illuminate the area under test. The microprocessor embedded in the system calculates the ratio of these two spectra absorbed and compares the results with a saturation value table to obtain the pulse oximetry.
Respiratory Rate
Respiratory rate monitoring is essential for assessing the respiratory health of COPD patients. Various sensors are available for this purpose, including electrocardiogram (ECG) sensors. Studies have shown that respiratory rate and even respiratory wave morphology can be approximated by ECG-derived respiration. This method detects the frequency by measuring the size of the R-wave in QRS signals, achieving a precision of over 97%.
Another appealing approach is contactless sensing using Doppler radar, which transmits radio waves and senses the signals reflected from the chest. The chest wall movement induced by the respiratory system is remotely monitored and captured as a waveform signal. A time-domain autocorrelation model is then applied to process the radar signals for rapid and stable respiratory rate estimation.
Predicting COPD Exacerbation with RPM
Most COPD exacerbation symptoms can be detected remotely using the sensing technologies described above. Respiratory sensors, pulse oximetry, and wearable/non-wearable devices for monitoring physical activity provide a data foundation for monitoring the flare-ups of physiological measures and symptoms. A previous study described a COPD exacerbation prediction method using respiratory signals. This method employs a machine-learning technique called decision tree forest to predict early AECOPD, achieving detection accuracies of 78.0% for detected episodes and 75.8% for reported exacerbations.
Prospect
AI technology has advanced rapidly in recent years, offering several promising applications in COPD management. AI-assisted computed tomography (CT) and other imaging diagnoses are helpful for accurately diagnosing COPD. CT scans at different levels can measure airway diameter and tube wall thickness, aiding in identifying the disease stage and clarifying its progression. The various wearable and contactless devices discussed will continue to improve, enabling more effective RPM in practice.
AI is expected to have a significant impact on the overall COPD remote monitoring market. In the future, COPD patients will be able to enjoy more hospital-grade medical care services in the comfort of their homes or outpatient facilities. Patients with complex and urgent hospitalization needs will receive more effective care in smart hospitals equipped with intelligent monitoring devices. AI-empowered tools will enable caregivers to better plan and coordinate care efforts according to patients’ individual demands, leading to better outcomes at lower costs.
Conclusion
RPM is crucial for managing patients with COPD. The development of sensors and AI technology has made it possible to effectively reduce the economic and medical burdens associated with COPD. AI-powered remote monitoring systems integrate data from various medical devices and sensors, analyze the data to understand patient conditions, determine trends, and generate early warning signals. This enables the medical team to monitor and treat patients more effectively, ultimately improving patient outcomes.
doi.org/10.1097/CM9.0000000000001529
Was this helpful?
0 / 0