As artificial intelligence (AI) continues to advance, it stands to become a significant disruptor in healthcare, including radiology. A new study published in Nature suggests that some AI models could interpret mammogram results better than human radiologists.
What New Research Shows
Researchers trained and deployed an AI model created by Google's London-based AI subsidiary DeepMind. The model scanned de-identified mammogram data from more than 76,000 U.K. women and more than 15,000 U.S. women.
According to the results, the AI model reduced the number of false positives and false negatives in breast cancer detection compared to human radiologists. You can read about the full story in Time online.
Does AI Threaten Radiologists?
Of course, these findings have raised questions about whether radiology will stay relevant as the use of AI expands in the healthcare industry. The use of AI and machine learning in radiology is still in the early stages, with researchers continuing to examine its value.
There is also more to radiology than just looking at medical images. Radiologists are highly involved in other vital medical tasks, including:
● Consulting with other physicians
● Performing image-guided interventions
● Comparing imaging findings with other medical records and test results
● Defining parameters for imaging examinations
● Discussing imaging results and procedures with patients
There are also patients themselves to consider. Newer technologies like AI take time to become accepted among the general population. Many patients may continue to prefer that a human radiologist review and interpret their imaging results!
Grow and Improve Your Radiology Practice
You may not be using AI in your radiology practice yet, but there are other solutions to help you communicate more efficiently, improving engagement and patient outcomes.
With Novarad’s suite of innovative solutions, medical imaging can become a better experience for radiology practices and their patients! Schedule a demo here and let our experts help you set up a custom radiology workflow.