Magnetic Resonance Imaging (MRI) has been a crucial imaging method for healthcare for more than 40 years and has made it possible to detect abnormalities in a way that was not possible previously. Now, with the help of AI, Ezra, under Emi Gal’s leadership, is making it possible to have a full-body MRI scan to detect cancer faster and cheaper than ever before.
When a person gets an MRI, they are put inside an electromagnet that magnetizes all the protons in the water retained in the body. While magnetized, interference is introduced into the field (in the form of radio waves) and then removed. Once removed, the protons relax back; depending on how quickly that process happens, you can create internal images of the body.
“The problem is that MRIs are a very “noisy” environment, so you have to do the scan multiple times,” said Gal in a recent interview conducted by Cleo Abram at the recent Converge event. By doing the scan multiple times, the signal will be the same, but the noise will be different, so you can average out the noise over numerous images to get a readable image.
How is Ezra different?
Using artificial intelligence gathered from existing scans of cancer, Ezra’s algorithm allows a patient to be scanned just once rather than numerous times, which is normally required to get a readable image. The AI helps decipher what is noise vs. what is relevant to the image. Having just one scan takes a sometimes hours-long process and shortens it to just 30 minutes.
Saving time in the MRI machine equates to saving money, making the technology an approachable way to cancer prevention and, in the best-known way, early detection.
Globant’s annual flagship event, Converge 2023, highlighted Emi Gal and Ezra in an interview with Cleo Abram, among many other speakers such as Walter Isaacson, Jane Lauder, and Marc Benioff. To view the recorded sessions, visit converge.globant.com.