Advanced product design and big data are changing our approach to healthcare. Take the human genome project for example. Using a genome sequencer, the whole human genome (about 3 billion base pairs) was successfully mapped in 2003. This breakthrough opened the door for personalized medicine where an individual’s genetic makeup influences the prescribed treatment plan. Since then, medical device innovations and big data have transformed how we approach health monitoring and analysis of health.
Remote health monitoring
Medical professionals traditionally evaluate health based on data collected from patients while at a medical clinic. Although these tests and evaluations are informative, office visits are infrequent with weeks or months between visits. In addition, medical tests can create burdens of time and means for individuals as well as clinics. As a result, the development and production of remote monitoring devices are increasing, allowing for data to be collected at a higher frequency and without the need to visit a clinic. For example, Reveal Biosensors has created a device worn on the arm which monitors tissue oxygen levels. The device has been used as a monitor for sleep apnea rather than requiring a patient to spend the night at a sleep lab. Oasis Diagnostics is another example. They are marketing a device that can be used anywhere to analyze saliva samples. The device is under development to analyze cortisol, testosterone, and hormone biomarkers with future tests under also underway to identify Alzheimer’s disease, Parkinson’s disease, and sleep disorders. Devices like these mentioned here have the potential to save time as patient monitoring can be done more frequently in the comfort of their own home and reduce the workload of medical facilities.
Predictive analytics
Predictive analytics in healthcare is the use of machine learning algorithms to build predictive models to identify states of health or recommend patient treatment. For example, one research group has developed a system to analyze the risk of a miscarriage for expecting mothers. The mothers wear sensors to collect data which is analyzed using machine learning algorithms, and the resulting information is sent to the patient and doctor by an app to indicate the mother’s state of health. This information can be used not only to identify risks, but to treat conditions before they become problematic.
Another predictive analytic system has been used and tested in clinics to identify cases of sepsis or heart failure. The study used algorithms to correctly identify the onset of sepsis shock up to 28 hours earlier than traditional identification methods allowing for quicker treatment and recovery. In addition, the algorithm was able to more accurately identify patients at risk for heart failure than traditional methods. As a result, patients were educated on their conditions and readmission rates dropped.
The development of medical devices in concert with big data will undoubtedly continue to improve and change the way we approach healthcare. So far, these advances are shifting to more personalized medicine and changing from a reactionary to a preemptive approach.