Software and Analytics are Revolutionizing Medical Sensors
From Monitoring to Disease Fighting: Software and Analytics are Revolutionizing Medical Sensors
Electronic sensors have been in use in medicine for decades, but their status is currently being elevated from monitoring tools to one of the most powerful sources of data used in disease fighting. This revolution is not being spurred directly by innovations in sensor technology or other hardware, but rather by innovations in software and in data collection, aggregation, and analysis that can make even apparently mundane information useful in improving healthcare outcomes and patient care.
Medicine Gets Data-Hungry
Before this shift in data collection and analysis, sensors could provide a variety of patient metrics like pulse rate, blood pressure, and EEG, but were mostly limited to monitoring, particularly in clinical settings. With nowhere for vast amounts of data to be stored, no way to access it if it were stored, and no effective way to analyze it, sensors remained in the critical but limited role of reactive monitoring.
Today, data collected by those same types of sensors, now interconnected, can be stored in the cloud and made accessible to patients, doctors, and even researchers, and can be analyzed and monitored automatically. Sensors are becoming part of a proactive healthcare effort rather than being merely reactive.
One thing that’s changing is the emergence of a new market for wearable monitoring products. They leverage compact sensors and connectors, often including wireless devices and some processing capabilities as well. That requires compact and fairly inexpensive sensors and connectors.
What this means from a hardware perspective is that sensors and the data they collect have gotten a lot more valuable. This is true in nearly every industry, but healthcare is unique in its high stakes nature and complex regulatory environment. Makers of sensor components and supporting connectors and other hardware have a massive business and research opportunity, but meeting industry demand and regulatory requirements will be a challenge.
Decades of Experience
Fortunately, sensor and connector companies are not starting at square one. As in many industries that are seeing the business and research potential of data analytics, the underlying sensors in many healthcare settings have been in place for a long time. MEMS (microelectromechanical systems) components have been in use since the 1970s, using tiny structures micromachined in silicon to convert mechanical factors like a patient’s heartbeat to electrical signals and data.
Sensors for healthcare applications, including MEMS components, continue to shrink and improve, encouraged at least in part by the heightened role of data in health decisions and the patient experience.
Components like Analog Devices’ AD8232, a single-lead heart rate monitor for ECG and other biopotential measurement applications, are using less power and making monitoring less obtrusive than their predecessors. These newest sensors, including the AD8232, are being created with wearables in mind, enabling designers to include them quickly and easily in an expanding range of products.
With data becoming more useful and valuable, the demand is strong for hardware innovations enabling wearability and implantability of medical devices; but the demand for health-related sensors won’t come entirely from the healthcare industry itself.
Bundles of Sensors
Industries are finding new ways to collect more and more valuable sensor data every day by putting sensors in everything. Much of this is sold to consumers on the basis of convenience or personal interest, like the accelerometer-derived fitness data collected by FitBits and iPhones. Sensors like STMicroelectronics’ LIS331DLH 3-axis digital accelerometer are part of the latest generation of continually shrinking MEMS sensors being included in consumer products like smartphones.
The practical upshot of this widespread inclusion of connected sensors in consumer products is that their data can eventually be aggregated with and augment more formal medical data to become an important diagnostic and research tool. With data aggregation, nearly every sensor could potentially have medical applications.
Kaiser Permanente, one of the largest HMOs in the United States, is actively working to put these principles of data aggregation, analysis, and reporting into practical use. One of the goals of the company’s Garfield Center for Innovation is to provide a testing ground where new technologies can be tested and developed under the stresses of a real hospital environment.
One of the issues being studied at the Garfield Center is a direct result of the pervasiveness of connected sensors: alarms. In the monitoring role of healthcare sensors, machines are capable of alerting hospital staff to emergency situations. However, false alarms produced by these sensors and machines are a major source of fatigue and distraction; in these cases, more information could actually mean worse patient outcomes.
Major healthcare companies like GE and Philips are all developing advanced software systems for what they call “alarm management.” Algorithms and automated monitoring can reduce the number of false alarms referred to hospital staff, improving their work environment and enabling them to serve patients better.
Kaiser’s Garfield Center is helping bring these innovations to real-world use more quickly. In an industry with high stakes, confirming a technology’s effectiveness before it goes into use is critical.
“We look at the technology in a level of detail that’s not possible anywhere else,” said Trevor Hogberg, Executive Director, Facilities Strategic Solutions & Clinical Operations at the Garfield Center. “We ran somewhere in the neighborhood of 2,500 different test cases on how an alarm can be generated, processed, and reacted to by a human being, and we made sure that, one, the technology was sound, but two, that the workflow processes were optimal”.
Without software innovations that can automate the handling of data, and computing and data storage technologies that mean sensor data can be used to enhance patient outcomes and fuel research, the large scale expansion of the use of sensors would remain impractical. Because of these software, computing, and storage advances, the demand for sensors, connectors and their supporting hardware will continue to grow.
Neil Shurtz is a contributor to ConnectorSupplier.com based in San Francisco. As a freelancer and in his work in public relations for high-tech companies, he has written about technology in the oil and gas, aerospace, and manufacturing industries. Shurtz specializes in framing complex and niche technical topics in a broader social context.