A few years ago, machines suggesting doctors on how to treat their patients was considered fantasy, but today, with the advent of AI, this is slowly becoming a reality. Recently, IDx LLC, an AI diagnostics company that develops clinically-aligned autonomous algorithms to detect disease in medical images, developed an FDA-approved medical device to detect diabetic retinopathy—a disease where high blood sugar levels damage the blood vessels inside the retina of eyes, leading to loss of vision. The device analyzes the captured images of the eye and provides the doctor with accurate information regarding the intensity of the medical condition.
AI’s potential application in healthcare can be broken into two separate categories: algorithmic solutions and visual tools. Most commonly used AI applications in healthcare are evidence-based approaches programmed by researchers and clinicians. Evidence-based approaches integrate clinical experience and patient values with the best available research information to aid in making decisions about the care of patients. Using consensus algorithms—a process in computer science that is used to achieve agreement on a single data value among distributed processes or system—along with EMR data recorded by the physicians, an AI-driven system can review hundreds of established treatment alternatives and recommend the most appropriate diagnosis for a patient. Meanwhile, visual tools enable clinicians to see what the human eye fails. These tools are driven by visual pattern recognition software, which can store and compare tens of thousands of images in a more accurate manner. As machines advance to be more powerful and deep-learning approaches gain greater traction, AI’s capability will continue to expand into diagnostic fields such as radiology, pathology, dermatology, and ophthalmology.
Without a doubt, machine learning and AI have the potential to take the field of medicine far beyond what it is capable of today.