Sheng Wang
@wangshengpkucn
Followers
916
Following
583
Media
2
Statuses
261
Assistant Professor CSE@University of Washington. AI for Medicine.
Seattle, WA
Joined February 2012
Our paper is out at Nature today! A whole-slide foundation model for pathology images.
Just out @Nature .The 1st whole-slide digital pathology #AI foundation model pre-trained on large-scale real-world data, from over 1.3 billion images, 30,000 patients.@hoifungpoon @HanwenXu6 @Microsoft @UW @naotous @MSFTResearch
5
18
138
Beyond pushing the state of the art in drug combo prediction, Pisces also introduces new benchmarks and datasets for multi-modal learning with lots of modalities (8 modality in our case) — an underexplored but increasingly important space.
1
0
0
Pisces is versatile foundation model that demonstrates superior performance on cell line-based drug combinations, xenograft-based drug combinations, and drug-drug interaction.
1
0
0
While the number of modalities increases, they also introduce more missing data/modalities. Pisces decomposes each multi-modal data point into 8 uni-modal data points, enabling us to train a uni-modal model that effectively approximates a full multi-modal model.
1
0
0
Excited to share our latest work: Pisces — a new multi-modal model for drug combination synergy prediction!.Pisces is a multi-modal framework that integrates 8 different drug modalities, including SMILES, 3D structure, molecular graphs, and textual descriptions.
1
1
7
RT @JAMACardio: EchoNet-TR is a model for screening tricuspid regurgitation from single-view TTE videos with the potential to enhance TR sc….
0
6
0
Thank you for the nice review of our Scorpius paper!
Thank you @AMIAinformatics for an having me for another fun Translational Bioinformatics Year in Review! #YIR25 #IS25. Here are the slides of the top papers of the year:.
1
0
5
Wei is a rising star in AI for Medicine. She is on the faculty market this year!.
📢I am on the academic job market this year! .My research interest involves utilizing AI and explainable AI to explore the mechanisms of aging and age-related diseases. I'm looking for faculty positions in AI for Biomedicine. Check out my website:.
1
2
8
RT @uwnews: .@uwcse researchers & @natashajaques created a method for training AI systems — both for large language models like ChatGPT and….
0
2
0
RT @hoifungpoon: @arjunmanrai @zakkohane @sheng_zh @naotous @HanwenXu6 @TristanNaumann @wangshengpkucn Multimodal GenAI is the next frontie….
0
3
0
Thank you @hoifungpoon for being the best leader! So excited to see that BiomedCLIP has been published. It is already a milestone in multi-modal biomedical AI since it is first released two years ago. Great work led by @sheng_zh @naotous Yanbo Xu, @HanwenXu6!.
Happy to conclude 2024 by sharing that BiomedCLIP has been published in NEJM AI (right before Christmas no less :-):
0
1
17
RT @martinjzhang: Using #AI to increase #RareDisease #GWAS discoveries by up to 100%. One of the largest integrations of #GWAS and #Functio….
0
4
0
Amazing work! Congrats!.
0
0
2
Thank you for sharing! I appreicate Allen School's great support to do interdisciplinary research!.
#UWAllen - Q&A: A new medical AI model can help spot systemic disease by looking at a range of image types
0
0
2
RT @peteratmsr: So nice to see @naturemethods publish this review of our BiomedParse @MSFTResearch paper on medical imaging. I like that th….
0
9
0
RT @hoifungpoon: Excited to see this review of BiomedParse in Nature Methods News: We're psyched that the authors….
0
2
0
Our paper is out at Nature Methods! A biomedical foundation model that handles 9 different imaging modalities trained from 6.8 million medical images. Amazing team work led by @IceBubble217 . I have the privilage to work with @hoifungpoon, the most visionary and heroic leader!.
New from Wei, Poon, and Wang--BiomedParse is a biomedical foundation model that can jointly conduct segmentation, detection and recognition across nine imaging modalities.
1
2
34
RT @yingheng_wang: 📢 We're excited to introduce LC-PLM, a Long-Context Protein Language Model that redefines the potential of protein seque….
0
23
0