The crossover between artificial intelligence and healthcare has arrived.
If the thought of machine learning in healthcare makes you imagine an eerie, Brave New World-esque universe, don’t worry. This technology will help bridge the communication gap between doctors and patients. It could also save lives.
Samantha Kleinberg, a computer science professor and researcher at the Stevens Health & AI Lab, was recently awarded three grants from the National Science Foundation (NSF) and the National Institutes of Health (NIH) totaling $2.3 million.
Her goal? To develop artificial intelligence that personalizes the information patients receive in order to help them make better-informed health decisions.
Kleinberg said she hopes to “improve human health through computing.” Her aim mirrors the growing need for artificial intelligence and big data in healthcare. Though the method is technologically-based, the focus is on the patient. This adds unprecedented personalization to diagnosis, treatment, and prevention.
The three diabetes-focused investigations use artificial intelligence and machine learning in their calculation methods to help patients.
One $917,879 grant from the NSF is titled “Uniting Causal and Mental Models for Shared Decision-Making in Diabetes.”
Kleinberg and her team will create computational methods to personalize information for patients with diabetes so that they can pick optimal treatment plans. Since the technology combines machine and human input, it uses a collaborative decision-making approach so that the patient and the clinician can work together to develop the best treatment plan.
Kleinberg recognizes that each patient, doctor, and caregiver has different beliefs about preferred treatment. With this in mind, she will create training modules to educate clinicians about how patient beliefs influence both trust and decision-making.
The next grant comes from NIH. This $864,220 award is put towards “Harnessing Patient Generated Data to Identify Causes and Effects of Nutrition during Pregnancy.”
About 9% of women develop gestational diabetes during pregnancy. According to March of Dimes, a non-profit focused on mother and baby health, this type of diabetes can result in premature birth or stillbirth.
For this investigation, Kleinberg will collaborate with Andrea Deierlein, a researcher at New York University. They will collect patient-generated data through wearable activity monitors.
With this technology, patients will log meals in real time using photos that can sense food types and calories. Additionally, the monitors can record symptoms to observe what happens before the illness develops. Kleinberg aims to pinpoint which factors cause this disease, as this technology could help guide important decisions during pregnancy.
The last award, “Moving Beyond Knowledge to Action: Evaluating and Improving the Utility of Causal Inference,” worth $499,454, comes from the NSF.
Kleinberg wants to create artificial intelligence that directly benefits patients.
“My group’s work aims to potentially prevent people from developing chronic diseases in the first place, and to otherwise help them better manage their illnesses,” she says. “It will hopefully help many people lead healthier lives.”
The research will focus on understanding how the output of an algorithm can be useful for human decision-makers. For conclusive results, algorithms should be evaluated based on their ability to be helpful for decision-making rather than their effectiveness (how many causes they find or how fast they accomplish finding them).
This investigation will introduce new methods for making such models more personalized, allowing patients to improve their everyday decisions around diet, exercise, and overall treatment.
Kleinberg’s research is innovative. However, if you think that overlap between health and AI is an invasion of privacy, you aren’t alone.
Public outrage was recently sparked by the revelation that Google quietly partnered with Ascension to collect and analyze patient data. This raised questions about the security of the initiative and if safeguards have been enacted for consumers.
Some critics are worried about the protection of their information. Others think that such technology will eliminate the need for humans in the healthcare community.
While Google waits to update the public on the implementation of this service, critics should consider both sides of the matter. Though risky, artificial intelligence and machine involvement in healthcare could be the newest trend in fostering an innovative and personalized healthcare system.
Navigating the world of health can be tricky, but Kleinberg’s work could spark a major shift in personalized healthcare, one that doctors and patients alike appreciate. After all, Kleinberg says, her work is “to aid humans – not replace them.”
Be First to Comment