Baltimore, MD
Joined May 2011
Ever wanted to do #tractography but didn't have #diffusion #MRI? Check out our preprint using #AI to propagate streamlines directly on T1w MRI, no diffusion needed! biorxiv.org/content/10.1101/… Big thanks to @grahamwjohnson @dc_moyer @kurtschilling5 @FrancoisRheault & @MASI_Lab team!
4
29
7
104
22,127
Convinced thrombectomy isn't beneficial in distal medium vessel occlusions (DMVOs)? New work from @LeonCai and @yedavaSNAPlab suggests an anterior DMVO-specific perfusion target mismatch profile may yet identify patients who would benefit! ahajrnls.org/49LcR36
1
6
9
1,092
Revisiting #tractography without #diffusion #MRI! Check out our #T1- and #AI-based streamline propagator applied to clinical cohorts! Let us know--what diseases would you want to see this technique studied with or validated on? My deepest thanks to @MASI_Lab @kurtschilling5 1/2
New paper in Imaging Neuroscience by Leon Y. Cai, Bennett A. Landman, et al: Tractography from T1-weighted MRI: Empirically exploring the clinical viability of streamline propagation without diffusion MRI doi.org/10.1162/imag_a_00259
2
7
2
29
4,933
Leon Y. Cai, MD, PhD retweeted
Poster is up at #MICCAI2023! We discuss learning strategies for integrating longitudinal Imaging + EHRs. I will be at the 9:30 AM poster session (T-3-73) and then presenting at 4:30 PM (oral 7). Looking forward to my first MICCAI with @MASI_Lab!
2
13
1,653
Tian presented his summer research project on using recurrent neural networks for brain tractography at the @VUEngineering poster session. Go @MASI_Lab !
1
6
585
Learn about SynBOLD-DisCo, a revolutionary deep learning technique developed by @UCSF_ci2's Dr. Joseph Vu, @UCSFImaging's Dr. Salvatore Torrisi & researchers from @VanderbiltU, @Harvard & @Sydney_Uni in our newest blog post. intelligentimaging.ucsf.edu/…
6
14
1,390
Stoked to share our multisite effort studying the brain in pediatric type 1 #diabetes with #MRI & #spectroscopy! Big thanks to @VUMCneurology @vumcpeds @VUMCradiology @sghetti1 @UCDavisHealth @CUAnschutz @vuiis @MASI_Lab collabs & more! @VandyBME @VanderbiltMSTP @EceVanderbilt
"Exploratory Multisite MR Spectroscopic Imaging Shows White Matter Neuroaxonal Loss Associated with Complications of Type 1 Diabetes in Children" #Spectroscopy #Diabetes #OpenAccess @theASFNR @The_ASPNR @LeonCai | bit.ly/3pSN1pX
8
1
30
6,196
Thank you everyone for coming to MIDL 2023. Here are the inside/outside group photos. @midl_conference, who can you tag? /1
3
4
1
16
2,088
Replying to @LeonCai
@LeonCai presenting his T1 tractography work at @midl_conference .
1
3
212
As far as I can tell, @luke_remedios won the most entertaining teaser slide @midl_conference.
1
4
1
15
1,607
Thrilled to have @LeonCai back from @VanderbiltMSTP (temporarily) for @midl_conference .
2
6
505
Our SynB0 distortion correction has been very well received for correcting spatial problems in diffusion weighted MRI when reverse phase encoded scans or field maps are not available. (This "should" not be a problems, but all too often is...). We now release a version for fMRI!/1
1
13
1
54
24,155
Hey everyone in diffusion MRI and tractography, if you have a cool project too complex for a 1-page abstract, consider submitting a paper to @CD_MRI workshop at MICCAI (deadline by the end of June)! 🇨🇦🧠👨‍🔬👩‍🔬💻🧮🔥
1
14
2
19
3,751
For those of you interested in the @CD_MRI workshop at MICCAI: if you want to submit your work on dMRI and Tractography, the submission process is open on CMT! You can search for "MICCAI 2023 Workshop on Computational Diffusion MRI".
1
7
1
10
2,275
1/3 Out of #GPU space in #medical #imaging #AI bc your data's too big? Those days may be over soon! @luke_remedios and I show we can use #AppleSilicon's shared CPU/GPU memory to train a #neural #network with gigapixel med images! That's >1 BILLON pixels! arxiv.org/abs/2304.12149
1
4
2
14
1,754
2/3 Med imaging AI is often limited by GPU capacity & big discrete GPUs cost $$$$. We were able to use >100GB of shared memory on a single @Apple #MacStudio w/ #M1Ultra for this #computational #pathology task for <$5k--potentially more cost-effective at larger memory capacities!
1
4
287
3/3 This hardware could transform how research labs ask medical AI questions! Big thanks to @sam_remedios & @MASI_Lab @YuankaiHuo @VUMC_Medicine colleagues! Also... @AppleEDU @tim_cook @mkbhd DM us if you want to collab :) @VanderbiltMSTP @EceVanderbilt @ViseVanderbilt @VandyBME
5
226