Introduction: Artificial intelligence (AI) will bring changes to the professional life of radiologists, as well as changing many other aspects of our lives. Since the invention of electricity, the internet and, more recently, artificial intelligence, the technologies of general use have made it possible for societies to progress and improve their quality of life.
Artificial intelligence and machine learning tools have the potential to analyse large data sets and extract meaningful information to improve patient outcomes, a skill that is also useful in radiology and pathology.
- Abundance of data
- Development of artificial neural networks
- Increased affordability of the hardware
The future of radiology augmented with Artificial Intelligence
Radiologists are not familiar with Artificial Intelligence, pioneering work in the perception of medical images in the 80s. We are experts in domains in medical imaging, medical physics and radiation safety. But in the last 6 to 12 years, there have been substantial innovations in obtaining images from deep learning methods of image classification. Today’s artificial neural networks have rates of accuracy that surpass those of human radiologists in narrow-based tasks, such as nodule detection The first step in formulating a strategy is to define our capabilities and identify the competitive forces that represent a threat. We are facing competition from other medical specialties who spend more time interacting with patients and who can choose to buy AI technologies. We also face competition from suppliers of equipment that make imaging devices, such as CT scanners.
General use cases, potential impact and implementation strategy
They can be divided into task-based categories:
Detection and prediction automation
Precision diagnostics and big data
Radiological decision support systems