Revolutionizing MRI Efficiency with MR-Now

by Valeria Delgado

Coauthors: Acknowledgement: Thank you Dr. Sharief Taraman for helping me come up with this amazing abstract idea. Additionally, a big thank you to Dr. Timothy Flannery for providing me endless support and guidance on not only my journey in MI3, but also on my future path in life. Author: Valeria Delgado Chapman University ‘24│vdelgado432@gmail.com References https://pubmed.ncbi.nlm.nih.gov/30637838/ https://www.ajronline.org/doi/10.2214/AJR.17.19480

Medical Devices & Digital Health

Background:
Magnetic Resonance Imaging, most commonly known as the MRI, is a powerful tool that allows for diagnosis to be obtained in a noninvasive manner. Thus, MRI’s are highly requested by physicians in order to correctly treat the illness of the patient. However, buying and managing multiple MRI’s can be quite expensive. Hence, as a result of the combination of the high demand yet short supply, patients often have to wait for hours, if not days, to get an MRI. This process can be very upsetting for both patients and physicians; patients greatly yearn for this test to be conducted so that the illness which is negatively impacting their life can be treated. On the physician’s side, it can be frustrating to strongly desire to help the patient, yet be restricted from doing so simply due to the lack of test results. Overall, the lack of efficiency with MRI usage can be a stressful and even detrimental experience for many.

Method:
MR-Now is an algorithm that is designed to take MRI efficiency to a whole other level. This algorithm will learn by utilizing previous information about patients who had an MRI, their diagnosis, their demographics, and the time spent in the MRI machine. MR-Now will essentially learn to make timing predictions and adjustments based on the patient’s symptoms and past medical history. With this previous information, MR-Now will be able to predict which future patients require MRI’s and will automatically assign a time slot that allows for quick, reliable results. In other words, MR-Now will know if a patient needs an MRI even before the doctor or the patient knows of this requirement. Furthermore, MR-Now can be utilized to predict the urgency of the MRI test. In order to better utilize the limited inpatient MRI machines that are available, MR-Now will be able to predict which patients are stable enough to receive their MRI’s in an outpatient setting. Since MR-Now will have access to all MRI’s in the county, it will be able to route non-urgent patients to a local MRI in a swift manner. That way, the scheduling can be done in advance so that when the doctor actually goes in to put the order for an MRI, the test will be available within minutes at a nearby location. This will ultimately allow for the remaining time slots to be reserved for more urgent inpatient situations. Lastly, MR-Now also has the ability to classify and distinguish which MRI’s in the county will be reserved specifically for MRI’s that are related to emergencies and trauma. Thus, there will be an MR-Now for these two different categories of machines that will apply the algorithm based on the differing scenarios.

Conclusion:
With MR-Now it will be possible to know exactly what time ranges are needed for exams and thus how to utilize MRI’s efficiently. As time goes on, MR-Now will be able to expand to have the capacity of predicting and managing machine usage for CT scans, X-rays, and much more!