Predictive medicine is a field of medicine that tries to predict the probability of a disease and institute preventive measures in order to prevent the disease. There are different prediction methodologies, including cytoomics, proteomics, and genomics. But the most fundamental way to predict future disease is based on genetics.
Proteomics and cytomics allow for early detection, but they detect biological markers that exist because a disease had already started. Comprehensive genetic testing, on the other hand, allows for the estimation of disease risk years to decades before it even exists.
Today, we will delve deep into the topic of predictive medicine, and try to explain it how it will develop in the future.
Predictive analytics is not something new. We are not reinventing the wheel. Predictive analytics basically applies what doctors have been doing for years on a larger scale. We now have a better and improved ability to measure, aggregate, and make sense of previously behavioral, psychosocial, and biometric data.
When you combine the new datasets with the existing science of epidemiology and clinical medicine, you can speed up the process of understanding the relationships between external factors and human biology. At the end, we get to enhanced reengineering of clinical pathways and truly personalized care.
Predictive analytics is the process of learning from historical data in order to make predictions about the future. That is the simple definition. In medicine, as in any other field, the goal is to reliably predict the unknown.
Some experts believe that the true benefits of artificial intelligence and machine learning will be when we move away from the current service model of healthcare and toward preventive medicine. The idea is that instead of waiting for people to get sick, we can predict illnesses and stop them from becoming a problem.
This might cost more up front, but it will save the healthcare industry a lot of money in the long run.
The big question regarding preventive medicine is where can we use it? Does it have real value at the moment? Are there are any examples? Well, we are at the beginning. But as we move deeper into the business of predicting health, the industry will redefine itself. There will be new opportunities for health maintenance and prevention.
The goal of predictive medicine is simple, flag risk factors so that physicians and patients can work together and reduce the chances of future problems. For example, if you have a greater risk of heart attack, you can get more regular EKG and cardiologist appointments. Predictive medicine can help both healthy people and those with existing diseases.
With that in mind, there is no shortage of great examples of predictive medicine. Here are some we already use:
Imagine a world where sensors do more than just monitor glucose levels. Combine that with real-time sensor data, full medical history, genetic information, lifestyle information, and more. The possibilities are endless.
But such technical revolution requires a lot of data from multiple platforms and healthcare systems. With digitalization of healthcare, we are already taking baby steps towards the end goal.
We have to stress that predictive medicine is not perfect, especially at its early stage. One of the risks, and biggest challenges is the risk of false positives. Even if we manage to get to 99.9% accuracy rate, that is still 1 in 1,000 people with a false diagnosis. He or she will suffer the unnecessary stress and strai of thinking he/she is at risk of something that will never develop. And we might spend millions treating something that will not happen.
Of course, there are also the ethical implications. Just imagine, what would happen if employers start mandating genetic testing for every employee? And they use that data to determine whether he or she is worthy of a job or promotion? What if health insurers begin to require genetic tests before they offer coverage?
Machine learning and artificial intelligence are still in their early evolutionary stage. But the end goal for predictive medicine is to get to personalized health through predictive analytics. We already see this in other areas.
For example, Netflix can offer recommendations and suggestions for content. No two people are likely to see the exact same customer menu. Facebook and Google do that for years. Their algorithms have evolved. Healthcare is no different, just a bit more complicated.
With modern technologies, we can work on solving the complicated challenges. Our ability to apply medical knowledge across vast population data can deliver new insights. They will have a great and direct impact on patient health outcomes.