The plurality of radiology leaders (27 percent) intend to adopt machine learning capabilities within one to two years. However, these healthcare stakeholders vary in how — and why — they think artificial intelligence will change clinical practice, according to a Reaction Data survey.
For the survey, Reaction Data asked 133 directors of radiology, imaging directors and radiologists, among other stakeholders, about their attitudes toward machine learning, a type of artificial intelligence in which a computer learns over time, rather than having to be programmed like typical software.
Practicing radiologists were most "skeptical" of machine learning's applicability and usability, while department leadership were more likely to perceive the technology as being "very important" to the medical imaging field, according to Reaction Data.
The three types of medical imaging participants were most likely to plan to use machine learning capabilities on were breast imaging (68 percent), lung imaging (61 percent) and chest X-rays (58 percent).
However, a minority of respondents (16 percent) reported they don't ever think their department will use machine learning. Their reasons for hesitancy tended to focus on human clinicians being "better" than AI (46 percent), lack of "forward-thinking" leadership (39 percent) and uncertainty surrounding the usefulness of machine learning (15 percent).
To access Reaction Data's survey, click here.