We interact with “intelligent machines” every day. They help us find the fastest route home, recommend movies based on our recent viewing patterns, protect us against credit card fraud, and even answer our customer service questions over the phone.
This is artificial intelligence at its most helpful. Still, AI’s growing ubiquity raises a larger concern: Will it eventually replace us in the workforce?
Health care professionals, especially radiologists, face this question today as talk swirls of AI and digital analytics being the future of medicine. Will intelligent machines invade the radiology reading room and push trained human doctors aside? It’s a battle between AI and the human eye, say the headlines, and AI will win in the end.
But this is just not true.
Radiologists already use AI – likely in a hospital near you – to help make patient care more effective, efficient and faster. It is no zero-sum game; it is a partnership between clinicians and code that augments, not replaces, human caregivers in ways previously unimagined.
Take the complex process of generating a Magnetic Resonance Image (MRI) of a beating heart. Capturing the heart using medical imaging is not just technically difficult, but also time-consuming. It involves terabytes of data requiring hours of expert interpretation, as well as patient discomfort and worry not just while being scanned, but also while waiting for results. This is a process ripe for improvement.
A Deep-dive Into Specialty Pharma
A specialty drug is a class of prescription medications used to treat complex, chronic or rare medical conditions. Although this classification was originally intended to define the treatment of rare, also termed “orphan” diseases, affecting fewer than 200,000 people in the US, more recently, specialty drugs have emerged as the cornerstone of treatment for chronic and complex diseases such as cancer, autoimmune conditions, diabetes, hepatitis C, and HIV/AIDS.
Digital health company Arterys recently joined doctors from Stanford University Medical Center and the University of California-San Diego to develop cloud-based software that can analyze seven clinically-vital dimensions of heart blood flow data simultaneously: three in space, one in time, and three in velocity direction. With this first-of-its-kind, FDA-approved deep learning application, radiologists can now take cardiac MRI analysis from about one-hour to only minutes, or even seconds.
Another example is with healthcare’s oldest form of medical imaging: X-ray. X-rays account for three-fifths of all medical imaging, but X-ray “reject rates,” the number of images that cannot be used due to poor image quality or patient positioning, can approach 25 percent. Reducing these reject rates could save time and resources, and improve the patient experience.
The University of Washington has pioneered X-ray analytics that help clinicians automate their data collection to help understand the root causes of rejected images. This work will eventually help radiologists and technologists get the best image on the first try – improving department productivity and freeing up more time for clinical interpretation and patient interaction.
Finally, the challenge of tackling America’s number one disabler and number three killer: stroke. Roughly 800,000 Americans experience a stroke each year, affecting patients and their families and causing $30+ billion in medical and indirect costs. Healthcare facilities in many parts of the world lack the ability to optimally triage, detect and treat acute stroke due to lack of access to imaging technology, gaps in expertise and time delays.
Partners Healthcare and its Center for Clinical Data Science have begun work on a holistic, AI-for-stroke “Solution Suite” that aims to simplify and organize MRI and computed tomography (CT) images for faster evaluation and triage of stroke patients. In situations where early diagnosis and treatment have been shown to reduce brain injury, “time = functioning brain tissue” – and AI-augmented radiologists could be a life-saver.
There is no question the practice of radiology is changing and will evolve even more in the future. Still, the process of turning traditionally point-and-shoot medical imaging machines into intelligent scanners is not an easy or quick task.
In the same breath, remember that these incredible AI-for-radiology advances are being developed by and in partnership with radiologists themselves. Why would radiologists embrace AI like this? Because it can improve their work and workflows, and ultimately improve patient care with real-time assessment, point-of-care interventions, faster exam times, more confident diagnoses, and even predictive analytics.
So, ignore the dire warnings and sensational headlines claiming radiology’s future is now minds versus machines. It’s minds with machines. AI is not a doctor or a cure; it is a tool. And it’s empowering and augmenting radiologists, improving the health care system and helping patients.
Image: Getty Images
