Medical devices generate huge amounts of essential, timely data that can have significant clinical impact. New technologies make it possible to effectively capture this data; however, discovering new actionable insights has not been common, with data captured from medical devices remaining vastly underutilised and wasted. The rise in predictive analytics is changing this technological lag as insights gained from big data analysis are set to augment the growth of the medical device industry.
Medical device industry growth
Predictive analytics using big data is a technology that learns from experience (data) to predict future behaviour in order to drive better decisions. Predictive models for patient risk and resource use can improve the quality of care and patient outcomes, as well as helping physicians gain useful insights. Numerous companies have been utilising this technology by gathering key patient vital signs along with other observations from devices in order to make decisions about the overall health of patients.
Medical device company Medtronic has recently worked with IBM to create a mobile personal assistant application, which provides real-time glucose insights for individuals with diabetes. The diabetes management system allows Medtronic to anticipate millions of data points; understand the potential links between glucose readings, drug administration, and lifestyle choices; and enable patients to make more informed decisions about their medication.
Digital therapeutics company Propeller Health has developed a digitally guided therapy platform for chronic respiratory disease, which integrates information from multiple sources, including connected medications, then uses machine intelligence to help individuals manage their medication. The data is then sent via Bluetooth to a smartphone app that uses machine learning algorithms to gather valuable insights.
Predictive analytics goes hand in hand with the diversity and multi-functionality nature of medical devices. In the future, this technology can be developed further, such that a clinical data analytics system may flag a patient at risk for a certain condition based on a pattern of medical device readings. The ability to correlate synchronous data from multiple anatomic locations may provide insights that cannot be achieved through single-device data analysis alone.