> This makes it hard to predict when and how deep learning methods will fail (there are no theoretical guarantees that deep learning will work).<p>I actually think we know fairly well how deep learning methods work (and what the shortcomings are), we just have no way to interpret the models it produces. Wouldn't ML techniques to reduce scan times fail at the most critical moments, ie when patients had unusual or unexpected ailments? Using ML in on downsampled MRI images feels akin to having an artist with a lot of familiarity of human anatomy touch up a scan.
Things like MRIs are the last thing you want to be using ML to invent detail in.<p>This proposal basically says using ML we can quarter the number of frequencies we sample and still get good <i>looking</i> scans. But the full resolution is made by inventing details based on statistics from a biased input (most MRIs are taken due to something being wrong).<p>Again, as with super resolution, ML cannot add detail that isn’t there, anything it creates is simply based on the statistical model it formed from the training set.
> Though compressed sensing can improve the image quality relative to a vanilla inverse Fourier transform, it still suffers from artifacts.<p>Odd remark. FDA approves compressed sensing products (e.g., [1], [2], [3], [4]) precisely because it is possible (and provably so) to quantify and/or characterize such “artifacts” up to substantial equivalence.<p>[1] <a href="https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K162722" rel="nofollow">https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn...</a><p>[2] <a href="https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K163312" rel="nofollow">https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn...</a><p>[3] <a href="https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K173079" rel="nofollow">https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn...</a><p>[4] <a href="https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K173617" rel="nofollow">https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn...</a>
Can’t wait to do an MRI and hear the doc say “You’re all set, good to go!”, only to discover that I actually had a tumor but that really clever ML algorithm thought that it was noise and should’ve been smoothed out…<p>I don’t want to be part of it, thanks
Could this be taken one step further... Use the ML in-the-loop during an MRI scan, to look at the data collected so far, then decide which frequency should be measured next to most improve the quality of the result?<p>This can also all be simulated offline without an MRI machine to test on with just access to a few full scans... So could be a good weekend project for someone here on HN, and your technique might even be in use by the time you need an MRI scan and will mean your doctor can get results slightly quicker and you get better healthcare, together with hundreds of millions of other people!