Artificial Intelligence in Medicine: AI is a Value-Added Improvement on all MRI Levels
Is there any part of the body that can’t benefit from magnetic resonance imaging (MRI)? So far, science hasn’t found one. In fact, not only can simple, straightforward MRI be applied to virtually all body parts, but MRI technology lends itself to endless development. For example, there are numerous variants in image acquisition that provide a range of information about bodily structures, allowing specific visual information to characterize tissues, fluids, bones, organs—you name it.
Attendees at the May, 2021 virtual annual meeting of the International Society for Magnetic Resonance in Medicine (ISMRM) were treated to an inspiring presentation by Dr. Shreyas Vasanawala, PhD (Stanford University) on the ways Artificial Intelligence (AI) can improve the benefits already offered by MRI. Broadly speaking, AI can inform the methods or processes by which the equipment is used and the images captured; and it is already proving its abilities to read the images much faster than human readers yet with competitive accuracy.
Some of the possibilities outline by Dr. Vasanawala are either still in development, not commercially available, or have not been widely integrated into clinical workflow because of slow implementation processes in hospitals or other institutions.
Still, Dr. Vasanawala’s vision includes exciting contributions from AI such as:
- Guiding how scans are prescribed and automating image acquisition
- Accelerating image acquisitions by enhancing accuracy and efficiency (less time in the scanner)
- Improving image reconstruction through deep-learning algorithms so it’s not only faster but also has greater accuracy through voxel-based quantitative analysis
- Detecting and even correcting image artifacts, such as the movement of a beating heart during a cardiac scan. “With help from AI, MR technologists could receive real-time feedback on image quality and make any adjustments before patients leave the scanner.”[i] Need for specific sequence corrections could be automatically implemented by repeating certain “slices”.
- “Automatically recognizing complex patterns in imaging data and providing quantitative assessments of radiographic characteristics. In radiation oncology, AI has been applied on different image modalities that are used at different stages of treatment”[ii] in clinical and research areas such as cancer, heart disease, neurology, etc.
Such contributions, and more, amplify the value of MRI for patients, doctors, academic centers, hospital systems, and researchers. Not only does MRI enable us to peer into the body to discern more and more fine details, it is safe to repeat as often as needed since the scan itself does not beam radiation into the body. (Although some types of MRI use radioisotopes injected into the body in order to highlight otherwise undetectable disease conditions, the patient is briefly exposed to very low doses of gamma rays and low energy electrons, not high energy beta rays. These isotopes are safe for humans.)
The time is fast approaching when the immense benefits of MRI are applied across all areas and branches of medicine. Meanwhile, Dr. Vasanawala and countless others recognize and are eagerly embracing the daily innovations by which AI is adding to MRI’s value.
NOTE: This content is solely for purposes of information and does not substitute for diagnostic or medical advice. Talk to your doctor if you are experiencing pelvic pain, or have any other health concerns or questions of a personal medical nature.
[i] Ridley, Erik L. “AI can add value in all parts of MR imaging chain.” AuntMinnie.com. May 15, 2021. https://www.auntminnie.com/index.aspx?sec=rca&sub=ismr_2021&pag=dis&ItemID=132398
[ii] Tang X. The role of artificial intelligence in medical imaging research. BJR Open. 2020;2(1): 20190031.
- Artificial Intelligence