Normally your GP may see you for a few minutes a few times a year and take a handful of test results. What if a computer system was constantly monitoring you 24/7 and could compare your results with readings from thousands of other patients worldwide and use the knowledge from clinical guidelines and studies to treat you?
In 2016 this idea is closer to becoming a reality. A machine in California has been designed by Sentrian, an early-stage machine learning and biosensor analytics company for remote patient management to be able to do this and is currently being trialed on patients. “We are trying to build a system that will enable us to listen to the lives and bodies of patients all the time, so we can make better, earlier and more personalised decisions.”
Currently one or two wireless ‘biosensors’ can collect simple data such as body temperature and heart rate as well as more complex information like oxygen saturation and potassium levels. If a patient could constantly wear several sensors at a time then the amount of data produced would be enormous.
Collecting all of this data and using learning algorithms to detect subtle patterns on chronic conditions such as heart disease and COPD can allow the machine to be able to detect early warning signs of an exacerbation or an attack and notify the doctor and patient when needed to warn them and allow more accurate treatment.
Many patients with COPD and similar chronic conditions will have symptoms a few days before an attack occurs but not feel ill until nearer the time. The computer could detect early signs of an upcoming attack and warn the doctor and patient so that medications could be altered, oxygen levels adjusted or warn the patient to eat or sleep more etc. It is currently being tested in the US and UK and early evidence has shown that the computer can spot early signs of a congestive heart failure exacerbations up to 10 days in advance. Early research is showing that even small things like heart rate change, sleep duration and body temperature may be indicators of an impending crisis and the system can even predict when someone may suffer a fall well in advance.
However the question being discussed is whether we should hand over all decision making to a computer when it comes to health-care? The profession itself is non-trusting of the results and of a computer being in control but Sentrian have suggested having joint control with humans. That doctors are notified of results and warning indicators and can use their judgment accordingly. The computer system would also be continually learning to know exactly which rules and interventions work best for which patients, and if a false alarm occurs the doctor can report this and the more feedback the system gets, the more it can learn and adapt.
Normally the human brain can remember the last 30 patients but the computer would be able to remember more than 30,000. Machine learning is also a lot faster than human learning. With the medical advances and research of the last 100 years medical knowledge has grown beyond belief and is becoming more difficult for the human brain to be able to intake and even during the course of a GP’s life medicine is constantly progressing and it is difficult for a GP to keep up-to-date. On average it can now take 17 years for new evidence-based findings to find their way to the GP and be clinically used.
What if a doctor was able to bring up every single case study and worldwide guidelines on a particular disorder to the forefront of their minds and instantaneously have all the possible data they need? The computer system can do this and therefore will you not be treating the patient with the most up-to-date knowledge possible? It could read all textbooks, medical databases of journals and literature, thousands of patients records and be subject to 1000’s of hours of teaching sessions from clinicians. Things that can take scientists 30 years to uncover may take the computer only a month, especially when it comes to research.
It could assist your GP by analyzing a patient’s medical record in conjunction with all of it’s acquired knowledge and suggest potential treatment recommendations. It could also match patients to clinical trials which usually takes hours of a human’s time to trawl through patient’s notes to see if they match the criteria for a trial and eliminates human error.
In theory by wearing a few sensors a computer system can monitor and record your readings, recognize subtle patterns and differences in your results, use data collected from around the world and if there are indicators for different treatments, a warning sign of an attack or deterioration of your condition then it can alert your doctor and you. It may just require a slight alteration in your medication dosage or oxygen saturation levels or ask that you eat, exercise or sleep more; things that the patient can do themselves or it may require intervention from the GP or hospital.
“Sentrian’s system is currently undergoing several randomised controlled trials, which means that the data of thousands of patients will be added to the platform. While we wait on clinical evidence, it is just a matter of time before this sort of artificial intelligence becomes a regular occurrence in the doctor’s office.”