The term artificial intelligence might have once seemed like a part of science fiction: a dream that we would never see in the real world. We now have driverless cars and limited AI technology in things like our wrist watches. Medical researchers are now taking these AI systems into the world of healthcare.
These applications are all about shrinking the time between symptoms and diagnosis. We have always looked for ways that computer science can improve human health. Now medical device companies are working on ways to use machine intelligence to make their products more effective for patients everywhere.
Making Rare Diseases Easier to Find
ThinkGenetic is a digital health startup company. It is finding a way for AI to help patients discover if they might have a genetic disease. David Jacob, ThinkGenetic’s CEO, recently moderated a panel called: How Artificial Intelligence is Moving the Needle in Medtech. “The panel will be looking for ways to use artificial intelligence to make their devices more knowledgeable,” says Jacob, “one of the underlying things will be how do we shorten the diagnosis odyssey – whether that’s with machine learning or with a wearable device.”
ThinkGenetic believes that it can locate people with diagnosed and undiagnosed genetic diseases by using AI systems. Jacob likes to call it, “GPS for genetics.” One of the most significant challenges for drug companies is finding patients. They create drug treatments for genetic conditions then they struggle to find the patients suffering from these illnesses in the real world.
How the AI Works
“What ThinkGenetic does is finds these patients on the internet when they’re out searching for answers and walks them through the process of learning more about themselves,” says Jacob. The application will take the patient step-by-step through a process. This application will help them figure out if they might have a genetic issue and then what it might be. This computer power will step in for human intelligence. First, it will take in all the signs and symptoms, and then it will connect them to all the possible diagnoses.
- Genetic diseases are among the rarest and difficult conditions to diagnose.
There is AI research filtering the symptoms and asking questions to narrow down the prospects. The artificial neural networks will then talk about what it could be before telling the patient what they need to talk about with their doctor. ThinkGenetic has genetic counselors on their staff to guide people to the next steps.
Other Uses for Artificial Intelligence
Advanced AI has become a component of many new medical leaps forward in the field of diagnosis. San Francisco based, Freenome, has created clinical studies to make AI-Genomic blood tests for colorectal cancer more normal. These test trials would be able to learn from its own mistakes over time, making cancer screening more accurate.
AI technology is also now being used alongside programs that will monitor a woman’s health. Ava, a startup firm, is currently working on a bracelet that will be able to track and monitor a woman’s cycle using AI. This application could aid in both pregnancy prevention and those women trying to become pregnant.
One of the most ambitious reported uses for AI is the company Beta Bionic. They are trying to create a bionic pancreas that can perform two crucial services for diabetes patients.
1. Monitor the blood sugar of a patient easily.
2. Regulate a patient’s blood sugar all on its own.
Beta Bionic has recently received FDA approval to start recruitment for in-home test studies of an insulin-only version of the device. Most of these new technologies are not meant to replace or simulate human doctors. These advances will hopefully only advance human medical care.
David Jacob predicts that the amount of uses for AI is only going to get bigger over the next few years. “In healthcare, I see AI making us more proactive rather than reactive, We’re going to see things coming before they actually come… All these devices are going to be sending data that can be useful to the healthcare system. If the algorithms are written correctly then we can basically see problems [ahead of time].”