Cloud-Connected Medical Devices for Personalized Medicine: an ECG Ring Sensor and a Home Air Pollution Sensor
Open AccessThis dissertation describes two cloud-connected medical devices, an ECG ring sensor which measures multi-lead ECG on-demand to promote point-of-care (POC) diagnosis of acute cardiac disease and a portable ambient gas sensor measuring personalized indoor pollutants exposure patterns for pediatric asthma patients. The supplementary cloud-based informatics system enables auto collection and real-time visualization of personal health-related data and promotes timely medical intervention, remote monitoring, and personalized disease management. We have conducted human subject studies to validate the ring sensor performance and study personalized NO2 exposure in pediatric asthma patients’ homes.The first sensor is a wearable Electrocardiogram (ECG) sensor in a ring format. It can obtain multi-lead ECGs with quality comparable to an FDA-approved clinical ECG. Using a reconstruction algorithm, the multi-lead Ring ECGs can reconstruct conventional 12-lead ECGs, which are widely used in cardiovascular disease diagnosis. The performance of this reconstruction algorithm has been demonstrated with healthy human subjects and is on par with a clinical 12-lead machine. The combination of the accurate ring sensor and the 12-lead ECG reconstruction capability enables point-of-care (POC) diagnosis of acute cardiac conditions such as myocardial infarctions to be achieved with telemedicine or artificial intelligence. The real-time POC diagnostic capability can reduce prehospital delays by shortening decision time. As a result, it can significantly contribute to timely medical intervention. The application of the ring sensor is not limited to ECG measurement. The ring also provides a miniaturized wearable platform that can adopt more sensing modalities by integrating other sensors without major modification of the rest of the system. The second is a portable ambient gas sensor device for pediatric asthma research. Asthma is the leading chronic disease in children. Current research has identified many asthma triggers, including NO2. However, the underlying mechanism between the triggers and asthma exacerbations regarding susceptibility and timing is still unclear. This is due to a lack of effective tools to quantify personal trigger exposure. We developed a cloud-connected indoor air quality sensor to address this issue, which can measure residential NO2, ozone, humidity, and temperature at one-minute resolution and automatically uploads the data to a cloud-based informatics system for real-time storage, analysis, and visualization. The lower limit of detections (LLOD) of NO2 (10 ppb) and ozone (15 ppb) are well below the current EPA standards. By deploying these gas sensors into pediatric asthma patients’ homes in the inner-city population in the DC metropolitan area, we found a significant correlation between frequent-short duration NO2 exposures and gas appliance usage. Meanwhile, we also observed a potential positive relationship between these NO2 exposure and hospital visits, and a further study can be explored. Finally, a cloud-based informatics system connecting the ECG ring and portable gas sensor has been developed to facilitate data collection, storage, analysis, and visualization. It can also be a key for large-scale studies, telemedicine, and timely medical interventions. The informatics system is built using the Amazon Cloud Service (AWS). The system scaling is easy and can even be set to automatic when more data are uploaded to it and more sensors are connected to it. These two sensor devices are connected together by the cloud informatics system and formed a comprehensive system which closes the loop from the data measurement, collection, analysis to disease diagnosis/intervention and research/feedback. As a platform, our system can be applied to a broader scope beyond cardiovascular disease and asthma with more sensing modalities integrated into it. Based on the demonstrated capabilities and exciting human subjects research findings, we anticipate that our devices can have great potentials in personalized medicine.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.