Artificial Intelligence Makes MRI Scans Safer for Patients

Artificial intelligence (AI) may soon make magnetic resonance imaging (MRIs) safer for patients by reducing the dose of gadolinium-based contrast agent, which is administered to patients to enhance image quality of the scan.

Highlights:

Artificial intelligence (AI) may help synthesize MRI images similar in quality to full dose contrast-enhanced pictures although the patient is given a significantly lower dose of gadolinium-based contrast agent
Gadolinium-based contrast material is administered to persons undergoing MRI scanning to give clearer and better quality images. It enhances areas of tumor, inflammation, and blood vessels
Traces of gadolinium are left behind in the patient’s body including brain tissue and may pose a health hazard although the exact effects on the patient remain unclear

Artificial intelligence (AI) can be used to create clear and good quality images by giving very low dose gadolinium-based contrast material to the patient, that look similar to images obtained after giving high dose of contrast. Gadolinium is a heavy metal and traces are left behind in the patient’s body with potential health risks including possible kidney damage.

The study was led by Enhao Gong, Ph.D., scholar at Stanford University in Stanford, California. The findings of the research will be presented at the annual meeting of the Radiological Society of North America (RSNA).

Artificial Intelligence in Healthcare - An Overview
Artificial intelligence (AI) in healthcare is the discipline using data intensive computer based solutions to improve patient care and outcome.

‘Artificial intelligence (AI) has the potential to synthesize MRI images with comparable quality to full dose contrast-enhanced MRI pictures without actually using any contrast agent.’
Read More..

Using Artificial Intelligence to Make MRIs Safer for Patients

The members of the research team hope to make MRIs safer for patients and reduce any potential health risks due to gadolinium-based contrast agents. In looking for a solution, they turned to artificial intelligence.

Gong said, "There is concrete evidence that gadolinium deposits in the brain and body. While the implications of this are unclear, mitigating potential patient risks while maximizing the clinical value of the MRI exams is imperative."

For the study, the team used MRI scans of 200 patients who had undergone contrast-enhanced MRI testing for various conditions
For each patient, there were three sets of images based on amount of contrast used while imaging. These included zero dose scans or precontrast scans taken initially before giving the contrast agent. The low dose scans were taken after giving 10 percent of the total gadolinium dose and the full dose scans were imaged after giving the full dose of gadolinium
The AI machine was shown the three sets of images and gradually learnt to artificially synthesize full dose contrast type of images from the low dose (10% dose) contrast images
The synthesized images were analyzed by neuroradiologists for image quality and contrast enhancement
The quality of the AI created images (using low dose contrast images) was found to be comparable to actual full-dose, contrast-enhanced MR images

The findings of the study suggest that the quality of images in MRI scans can be preserved while drastically reducing the amounts of gadolinium given to patients by using artificial intelligence. In fact the team members are hopeful that in the future, AI might have the potential to totally overcome the need for contrast agent if it can create full contrast-enhanced images from the zero dose contrast images.

Gong says, "Low-dose gadolinium images yield significant untapped clinically useful information that is accessible now by using deep learning and AI."

Artificial Intelligence System to Identify Often Missed Cancer Tumors Developed
Artificial intelligence (AI) may offer a new promising way to identify often missed cancer tumors.

What is Artificial Intelligence?

Artificial intelligence (AI) is also known as machine intelligence. In AI, the computer is trained over a period of time to perform tasks similar to humans. This is achieved by a process called deep learning technology where, for example the computer is fed with images and taught over a period using established parameters to recognize patterns and shapes (that may not be evident to the human eye). Once it learns, the computer is then able to process the information/images fed into it and interpret it as well as and much faster than humans.

Artificial Intelligence Could Better Predict Psychosis Depression Risk
Machine learning, also known as artificial intelligence could predict social functioning in early psychosis and depression.

Future Plans

The team plans to take their research to the clinical setting
Also, plans to test the AI algorithm in a wider range of MRI scanning machines and other contrast agents as well

Training Artificial Intelligence to Diagnose Rare Diseases Using Artificial X-rays
Artificial intelligence can be used to diagnose rare diseases fast and accurately and researchers are trying to 'teach' the algorithms what to look for by generating artificial X rays in AI training sets.

Conclusion

In conclusion, the findings of the study may be a potential game changer in MR imaging and could make imaging safer for patients.

Concluding with the remarks of Dr Gong, "We're not trying to replace existing imaging technology. We're trying to improve it and generate more value from the existing information while looking out for the safety of our patients."

Reference:

Artificial Intelligence May Help Reduce Gadolinium Dose in MRI - (https://press.rsna.org/timssnet/media/pressreleases/14_pr_target.cfm?ID=2049)

Source-Medindia