Some computational problems are so inherently difficult that conventional computers, including large supercomputers, can take years to solve them. One such problem has been encountered by researchers at Case Western Reserve University, who have developed a way to improve and speed up the detection of tumors on MRI scans and improve MRIs in many other ways. Their technique, known as magnetic resonance fingerprinting, which was featured in Nature a few years ago, can significantly increase the amount of useful information that MRI scans provide. But, to take full advantage of this technology, each scan has to be tuned to the individual patient and the target that’s being searched for.
This requires enormous computing power, but now Microsoft is partnering with Case Western to apply its quantum computing technology to make magnetic resonance fingerprinting clinically practical. Though actual quantum computers will not be used, algorithms inspired by them will play a major role.
If this partnership is successful, the consequences would be enormously consequential for radiology and all the fields of medicine relying on it, let alone the types of research that would be unleashed.
Here’s some details about what Microsoft will be offering, according to the company:
Microsoft will help optimize the pulse sequences by mapping the problem to a form suitable for quantum computers and then use a quantum-inspired algorithm, which runs on the classical computers we have today. By generating an optimized pulse sequence, researchers at CWRU will be able to create a solution that improves the diagnostic capability of MRI. This work will lead to an improved patient experience, requiring less time in the MRI machine and providing more accurate, rapid results. Once the scan is complete, Microsoft HoloLens will be used for a 3D, holographic model of the results.
Find out more about the project at its info page…