Sign in

Greg Holste

@giholste.bsky.social
84 followers 101 following 4 posts

PhD Student at UT Austin | @NSF Graduate Research Fellow Building deep learning methods to interpret medical images www.gholste.me

PostsRepliesMedia
Reposted by Greg Holste
Evangelos K. Oikonomou @ekoikonomou.bsky.social · 25/06/2025
Check out #PanEcho, now in @jama.com ✅ open weights ✅ multiview multitask reporting ✅ international validation ✅ works with POCUS Amazing effort @cardslab.bsky.social @giholste.bsky.social @rohankhera.bsky.social + @atlaswang.bsky.social, Márton Tokodi & Attila Kovács #CardioSky #EchoSky
184
Reposted by Greg Holste
JAMA @jama.com · 23/06/2025
An #AI system that automatically interprets echocardiograms maintained high accuracy across geography and time from complete and limited studies. ja.ma/4jYWBgI
Figure 4. Task-Specific ViewRelevance
194
Greg Holste @giholste.bsky.social · 24/06/2025
Thank you @erictopol.bsky.social for highlighting our work! Huge honor This is the product of deep interdisciplinary collaboration btw engineers & clinicians. Grateful to teammates & mentors @ekoikonomou.bsky.social @rohankhera.bsky.social + Model is publicly available! github.com/CarDS-Yale/P...
github.com
GitHub - CarDS-Yale/PanEcho: "PanEcho: Complete AI-enabled echocardiography interpretation with multi-task deep learning" by Gregory Holste, Evangelos K. Oikonomou, Zhangyang Wang, & Rohan Khera
"PanEcho: Complete AI-enabled echocardiography interpretation with multi-task deep learning" by Gregory Holste, Evangelos K. Oikonomou, Zhangyang Wang, & Rohan Khera - CarDS-Yale/PanEcho
033
Greg Holste @giholste.bsky.social · 22/05/2025
Vibe coded a new personal website: www.gholste.me Do I know how Next.js works? Not in the slightest. Am I using it in my website? Apparently! Have to say I'm very impressed with @v0.dev, saved me hours and hours
gholste.me
Gregory Holste
Ph.D. Candidate, The University of Texas at Austin
000
Greg Holste @giholste.bsky.social · 25/11/2024
Working on #AI for #echofirst? Use PanEcho! Why PanEcho? The model is trained on >1M echos from all views on 39 diverse tasks, meaning... - PanEcho is a powerful transfer learner - It takes *one line* of Python code See github.com/CarDS-Yale/P... for details + reach out with Qs
080
Reposted by Greg Holste
Rohan Khera @rohankhera.bsky.social · 22/11/2024
Presented at #AHA24 Late Breaking session, we are pleased to announce #PanEcho - Complete AI-enabled echocardiography interpretation with multi-task deep learning. Led by @giholste.bsky.social & @ekoikonomou.bsky.social @cardslab.bsky.social Preprint, code, and model: cards-lab.org/panecho
42610