By Dr. William Randazzo
Radiology is inseparable from the locomotive-like advancement of technology, forcing change over time. Those of us in the field can choose either to hold on to the past and get passed by, or to scrupulously utilize the latest advancements in tech for the benefit of our patients and our industry.
An early example of technology’s impact on radiology is the digitization of medical images, which allowed for widespread access to patients’ imaging files through hospital-wide PACS. Physicians no longer needed to go to the radiology department to assess a patient’s X-ray, which were previously under the control of the radiologist. This advancement allowed physicians to more quickly determine a patient’s condition and start treatment.
The subsequent advent of the internet – plus the use of digitized images – sparked the teleradiology revolution, turning the conventional model of radiologists’ interactions with patients and physicians upside down. This barreling train brought with it 24/7 radiology coverage, which is now moving more and more towards around-the-clock subspecialty radiology coverage.
Yet another disruptive wave has been born from the fast-moving technology sector, causing some distress among those who hold tightly to the existing tech paradigm.
Artificial intelligence has gained mainstream attraction and diagnostic radiology is a natural fit: a profession purely operating in the digital space. Of particular interest is the promise of autonomous medical image interpretation. This is not to be confused with computer-aided detection, which has been in the marketplace for years and is used rather sparingly in clinical practice. Due to the high false positive and false negative rates, radiologists largely rely on their own internal neural networks to interpret images instead of being slowed down by a traditional computer-aided detection system.
However, the latest and greatest advancements from the artificial intelligence sector have far superior computing power than previous generations of computer-aided detection, utilizing algorithms that can be trained via the same data sets used in the education of radiologists.
Apply Moore’s Law describing the exponential advancement of transistor density in a circuit board over time to almost every area of technology development and you will find little reason to doubt that the future holds realities beyond what we think is currently possible. We, the radiologists, have always been early adopters of technology and the medical community is paying attention to our reaction to these latest advancements.
Peruse radiology blogs and radiology forums and you will find the full spectrum of opinions about the future of radiology in a world where artificial intelligence is the new norm. Some hold that despite Moore’s Law, computer-aided detection is not be relied upon. Others suggest that we start applying for new jobs.
The truth will only be seen in time. However, one thing is certain: we as a society are standing before a new horizon of technological advancement that will be widely applied to the software we use everyday and have come to take for granted.
With radiology, technology and traditional medicine converge to give us the best shot we have to understand and heal the human body. To hold stubbornly to the status quo when introduced to these potentially revolutionary technologies would be not only unwise, but irresponsible.
Artificial intelligence is coming, and it will undoubtedly reshape the interaction of patients, radiologists and physicians. If this change is indeed for the betterment of patient care, it is here to stay.
