Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Machine Learning for MRI Image Reconstruction

rhotter.github.io

Post reply on HN
Print view View original

31–40 of 49 posts

Re: Machine Learning for MRI Image Reconstruction

#31

Post by 01acheru »

notarandomer wrote: post #29
01acheru wrote: post #26

Earlier quoted context omitted.

Well, I really hope you succeed on your experiments, but right now it seems to me that ML introduces an error factor we still cannot precisely account for. It looks like you are proposing some kind of mixed approach, not a simple “less data, faster scan, ML to the rescue”. I understand how MRI works but you are surely and obviously much more knowledgeable than me, so I simply wish you luck! My problem with the articl…

Thanks! I agree taht generative models on their own are definitely more risky than methods that combine ML for regularization along with data consistency terms that force the reconstructions to be consistent with the acquired data.

My opinion is pretty much worthless but I think this is a much more sensible approach, using the strengths of ML but putting constraints on the outputs.

Just a simple question: to achieve a guarantee that those results are at least equal or better than the ones we have now on our battle tested setups, shouldn’t we use the same sampling we use on a “default” MRI? I mean… using those reconstruction algorithms to try to achieve a better result, without downsampling so that a standard reconstruction can still be performed to be checked against?

Re: Machine Learning for MRI Image Reconstruction

#32

Post by selimthegrim »

01acheru wrote: post #16
ryan93 wrote: post #13

Earlier quoted context omitted.

Why would the ML algorithm necessarily change the scan. The radiologist could still look at the unadulterated MRI.

There is no unadulterated result, you are doing less sampling and relying on ML to fill the gaps. So you either have the ML reconstructed result or a subsampled MRI. Healthcare is an area where we need good and clean data as much as possible, let’s use ML reconstruction somewhere else.

There are definitely ways to work well with subsampled data, see Lester Mackey's recent work.

Re: Machine Learning for MRI Image Reconstruction

#33

Post by vitorsr »

> Though compressed sensing can improve the image quality relative to a vanilla inverse Fourier transform, it still suffers from artifacts.

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.

[1] https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn...

[2] https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn...

[3] https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn...

[4] https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn...

Re: Machine Learning for MRI Image Reconstruction

#34

Post by 01acheru »

selimthegrim wrote: post #32
01acheru wrote: post #16

Earlier quoted context omitted.

There is no unadulterated result, you are doing less sampling and relying on ML to fill the gaps. So you either have the ML reconstructed result or a subsampled MRI. Healthcare is an area where we need good and clean data as much as possible, let’s use ML reconstruction somewhere else.

There are definitely ways to work well with subsampled data, see Lester Mackey's recent work.

Could please share a link to the work you are referring to? I would really appreciate it (not ironically, it would be truly appreciated).

I know we can work around subsampling, actually we have always been very good at it since our sampling data back in the day was way smaller than what we refer as subsampled today.

Re: Machine Learning for MRI Image Reconstruction

#35

Post by selimthegrim »

01acheru wrote: post #34
selimthegrim wrote: post #32

Earlier quoted context omitted.

There are definitely ways to work well with subsampled data, see Lester Mackey's recent work.

Could please share a link to the work you are referring to? I would really appreciate it (not ironically, it would be truly appreciated). I know we can work around subsampling, actually we have always been very good at it since our sampling data back in the day was way smaller than what we refer as subsampled today.

http://stanford.edu/~lmackey/papers/thinning-slides.pdf

Re: Machine Learning for MRI Image Reconstruction

#37

Post by 01acheru »

selimthegrim wrote: post #35
01acheru wrote: post #34

Earlier quoted context omitted.

Could please share a link to the work you are referring to? I would really appreciate it (not ironically, it would be truly appreciated). I know we can work around subsampling, actually we have always been very good at it since our sampling data back in the day was way smaller than what we refer as subsampled today.

http://stanford.edu/~lmackey/papers/thinning-slides.pdf

Thanks!

Re: Machine Learning for MRI Image Reconstruction

#38

Post by notarandomer »

01acheru wrote: post #31
notarandomer wrote: post #29

Earlier quoted context omitted.

Thanks! I agree taht generative models on their own are definitely more risky than methods that combine ML for regularization along with data consistency terms that force the reconstructions to be consistent with the acquired data.

My opinion is pretty much worthless but I think this is a much more sensible approach, using the strengths of ML but putting constraints on the outputs. Just a simple question: to achieve a guarantee that those results are at least equal or better than the ones we have now on our battle tested setups, shouldn’t we use the same sampling we use on a “default” MRI? I mean… using those reconstruction algorithms to try to…

"Default MRI" (i.e. fully sampled) should definitely be acquired when possible when testing this out to compare to gold standard. But the benefit of using ML methods in the fully sampled case would be minimal (maybe some denoising), whereas they have a much larger effect when acquiring highly undersampled data that traditional reconstruction methods fail at. It's also not always possible to get fully sampled reference data. For example in functional MRI you might not be able to get matched fully sampled data because the benefit of undersampling is in improving the temporal resolution. These cases are definitely more researchy and less clinical though, and in my work we add a 2 minute highly undersampled scan to current standard protocols and compare what we can reconstruct from our 2 minute scan compared with the fully sampled (but often lower resolution) standard scans.

Re: Machine Learning for MRI Image Reconstruction

#39

Post by tangoed »

tqi wrote: post #3

> This makes it hard to predict when and how deep learning methods will fail (there are no theoretical guarantees that deep learning will work). 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 unexpe…

I'd argue that knowing fairly well and theoretical guarantees are significantly different.

As an example, you can run a million simulations on a satellite with different initial conditions to test your new control algorithm. However, you have infinitely many possible initial conditions, and you can't simulate all of them. If you however show that the closed loop system in stable sine sense, it's a more rigorous guarantee.

Re: Machine Learning for MRI Image Reconstruction

#40

Post by tangoed »

01acheru wrote: post #12

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… I don’t want to be part of it, thanks

You'll be surprised how many well-known doctors miss a ton of non-obvious anomalies in scan results. There's no way a doctor would have seen all prior records of confirmed diagnosis and their corresponding scans.

In an ideal world, a deep learning algorithm should provide an independent report of potential features of interest to a doctor to let him know if something that he could have missed. However, I hope it stays in the intended role and doesn't make the doctor less careful.

Post reply on HN