Artificial Intelligence and analytics have become deeply integrated into our lives. Most of us encounter these algorithms multiple times per day: a customer service voice recognition program, an online shopping recommendation, or a well-targeted Facebook ad.
As our ability to harness machine learning becomes increasingly sophisticated, we see that healthcare is where these applications have the potential to deliver the most profound impact on our lives.
Tools that can churn through enormous sets of data and images faster than any human could – and then provide quality analysis that will aid physicians in improving patient care – are within our grasp.
Researchers already are experimenting with algorithms created to detect skin cancer, spot tumors in mammograms, and discover retinal damage in patients with diabetes. The FDA recently approved the sale of software using an algorithm that can calculate – in just 30 seconds – the volume of blood a patient’s heart can hold and pump.
After several days in Chicago attending the Radiological Society of North America (RSNA) annual conference this week, I am particularly excited about radiology as the next frontier for machine learning.
Radiology always has been a future-forward leader in healthcare – more than a century ago, it gave us the ability to literally see inside the human body in ways that transformed the practice of medicine.
Today, imaging remains one of the most effective, and often-used diagnostic tools in clinical practice, accounting for nearly 10 percent of medical costs in the U.S.
Enter radiology machine learning and AI: which can be used to create tools that can be taught to identify (or rule out) pneumonia, lumbar fractures, pulmonary embolisms, and many other health issues, as well as automatically move patient-critical findings to the top of a radiologist’s worklist.
“The objective of harnessing the power of deep learning for medical image analysis, and embedding it in an effective program of clinical care, is one of the most important challenges in artificial intelligence,” Tom Davenport wrote this month on Forbes.com.
Yet, machine learning should not be one more burden on radiologists. It is vital that we deliver useful, actionable data through AI-powered, speech recognition, and image sharing tools that already are in use.
This week, we unveiled the Nuance AI Marketplace for Diagnostic Imaging, the world’s first open AI marketplace for diagnostic imaging. Similar in concept to globally available “app stores” for businesses and the public, the Nuance AI Marketplace empowers radiologists and AI developers to build, test, and share AI algorithms for improved detection, diagnosis, and treatment.
Our hope is that these algorithms will amplify radiologist expertise – giving them the most comprehensive, timely information to best care for and treat patients.
Learn more by connecting with us on LinkedIn and Twitter or download a white paper about the AI Marketplace here.