Clinical and Applied Machine Learning Group @MIT_CSAIL. Led by Prof John Guttag, we focus on clinically inspired machine learning with real-world relevance.
Cambridge, MA
- Tweets 111
- Following 26
- Followers 617
- Likes 61
MIT Clinical and Applied Machine Learning retweeted
X-rays are everywhere in medicine, but extracting general, quantitative anatomical information from them remains remarkably difficult.
Announcing FleXray: an open-source, open-weight model for zero-shot anatomical segmentation across the human body—from different regions, viewing angles, and acquisition settings.
FleXray learns from large-scale synthetic supervision generated from 3D anatomy, then transfers directly to real clinical X-rays.
Try it now on your own X-rays in our in-browser demo!
🧵 1/N — project page, demo, weights, code & paper below ↓
18
63
424
69,868
Check out this awesome new work from our lab! The first general purpose X-ray segmentation tool 🤯
X-rays are everywhere in medicine, but extracting general, quantitative anatomical information from them remains remarkably difficult.
Announcing FleXray: an open-source, open-weight model for zero-shot anatomical segmentation across the human body—from different regions, viewing angles, and acquisition settings.
FleXray learns from large-scale synthetic supervision generated from 3D anatomy, then transfers directly to real clinical X-rays.
Try it now on your own X-rays in our in-browser demo!
🧵 1/N — project page, demo, weights, code & paper below ↓
5
297
MIT Clinical and Applied Machine Learning retweeted
Very excited to present ScribblePrompt at #ECCV2024 this afternoon (Tue Oct 1, 16:30-18:30 CEST)! Stop by poster # 70 if you are around!
Presenting ScribblePrompt: a lightweight interactive segmentation tool that enables users to perform new biomedical image segmentation tasks using a few bounding boxes, clicks and scribbles, to appear at #ECCV2024 🎉
Work w/ @MarianneRakic, John Guttag and @AdrianDalca
1/
7
33
3,396
MIT Clinical and Applied Machine Learning retweeted
Presenting ScribblePrompt: a lightweight interactive segmentation tool that enables users to perform new biomedical image segmentation tasks using a few bounding boxes, clicks and scribbles, to appear at #ECCV2024 🎉
Work w/ @MarianneRakic, John Guttag and @AdrianDalca
1/
3
9
40
6,761
Checkout @HalleeWong 's awesome work on building a general-purpose interactive segmenter, appearing now at #ECCV2024!
Presenting ScribblePrompt: a lightweight interactive segmentation tool that enables users to perform new biomedical image segmentation tasks using a few bounding boxes, clicks and scribbles, to appear at #ECCV2024 🎉
Work w/ @MarianneRakic, John Guttag and @AdrianDalca
1/
2
10
697
MIT Clinical and Applied Machine Learning retweeted
.@MIT_CSAIL PhD student Marianne Rakic's most recent project, Tyche, is a medical image segmentation model that aims at generalizing new tasks & capturing uncertainty in the medical image.
Learn more about Marianne and her recent projects: bit.ly/4d6X39t
4
15
767
MIT Clinical and Applied Machine Learning retweeted
.@MIT_CSAIL PhD student Marianne Rakic's most recent project, Tyche, is a medical image segmentation model that aims at generalizing new tasks & capturing uncertainty in the medical image.
Learn more about Marianne and her recent projects: bit.ly/4d6X39t
3
7
709
MIT Clinical and Applied Machine Learning retweeted
To the untrained eye, a medical image like an MRI or X-ray appears to be a murky collection of black-and-white blobs. 🩻
When trained to understand the boundaries of biological structures, AI systems can delineate regions of interest for biomedical workers. MIT CSAIL, MGH, and Harvard Medical School researchers’ interactive “ScribblePrompt” can help rapidly segment any medical image, even types it hasn’t seen before: bit.ly/3XzR3S6
A thread on the upcoming paper at @eccvconf 🧵
3
27
103
32,607
MIT Clinical and Applied Machine Learning retweeted
🏆Our Bench-to-Bedside Award goes to:
"ScribblePrompt: Fast and Flexible Interactive Segmentation for Any Biomedical Image"
by Hallee E. Wong, Marianne Rakic, John Guttag, Adrian V Dalca
Congratulations! 🎉
6
20
971
New in-context learning work from our lab just dropped!
Checkout out Tyche, to appear at #CVPR2024 as a ✨highlight✨, led by @MarianneRakic.
So excited that Tyche is a highlight at @CVPR 🥳
Tyche is a stochastic strategy for in-context medical image segmentation, to both generalize to new tasks and capture uncertainty.
Work with @AdrianDalca @HalleeWong @jjgort John Guttag and @CiminiLab
1
6
714
MIT Clinical and Applied Machine Learning retweeted
Presenting UniverSeg: an in-context learning model for medical image segmentation to appear at #ICCV2023 🎉! (w/ @jjgort, @mertrory, @AdrianDalca, and others)
@MIT_CSAIL, @MIT, @MGHMartinos
🧵1/N (project-page, demo, and paper links 🔗at the end)
3
24
84
16,293
Check out UniverSeg! Recent #ICCV2023 work from our lab led by co-first authors @VictorButoi and @jjgort.
Presenting UniverSeg: an in-context learning model for medical image segmentation to appear at #ICCV2023 🎉! (w/ @jjgort, @mertrory, @AdrianDalca, and others)
@MIT_CSAIL, @MIT, @MGHMartinos
🧵1/N (project-page, demo, and paper links 🔗at the end)
1
7
1,139
MIT Clinical and Applied Machine Learning retweeted
In New York today to present a paper with @rajivmovva on the impact of coarse race variables on the study of disparities in clinical risk score performance!
1
3
38
4,324
MIT Clinical and Applied Machine Learning retweeted
@SCzolbe is presenting at #CVPR2023 today (TUE-PM-200 in person)
Neuralizer: General Neuroimage Analysis without Re-Training
code: github.com/SteffenCzolbe/neu…
video: youtu.be/_wgCESDwjZI
@FreeSurferMRI @MGHMartinos @mit_caml @MIT_CSAIL @MITEECS @CVPR
2
4
24
1,806
MIT Clinical and Applied Machine Learning retweeted
In Vancouver for #CVPR2023 ! Excited to chat about multimodal generative models, especially in the context of creativity or health - please reach out if these are of interest to you.
3
15
2,650
MIT Clinical and Applied Machine Learning retweeted
New working paper! We study the impact of coarse race categories - think "Asian", "Black", "White" - on the study of predictive disparities in health.
1/ Patient race in health datasets is often reported coarsely: for example, both Indian and Chinese patients are categorized as “Asian”. Does the coarse coding of race hide disparities in clinical machine learning performance?
In our new working paper (!), we find that it does.
2
8
39
6,889
MIT Clinical and Applied Machine Learning retweeted
1/ Patient race in health datasets is often reported coarsely: for example, both Indian and Chinese patients are categorized as “Asian”. Does the coarse coding of race hide disparities in clinical machine learning performance?
In our new working paper (!), we find that it does.
7
29
153
49,034
MIT Clinical and Applied Machine Learning retweeted
I'm presenting Kaleidoscope, a system for user-driven, context-specific, and semantically-meaningful ML model evaluation at #CHI2023! dl.acm.org/doi/10.1145/35445…
1
6
44
9,473
MIT Clinical and Applied Machine Learning retweeted
"UniverSeg: Universal Medical Image Segmentation"
What if we could train a single neural net to highlight important structures in any medical image given just a few examples? [1/13]
4
59
267
63,423
