P2Echo uses text prompts for multi-view echocardiography segmentation
P2Echo, introduced in an arXiv paper submitted on November 10, 2026, uses structured, view-aware text prompts to control segmentation of cardiac structures across multiple standard echocardiographic views with a single model. Its Dynamic Image-Text Alignment module conditions decoder features on the requested view and anatomical structure. The authors report that P2Echo matched or exceeded view-specific specialists and state-of-the-art universal models on a large internal dataset and three external datasets, while using a fraction of their computational cost. The paper provides a GitHub code repository, and the official pretrained weights are available through the moein99/P2Echo model repository on Hugging Face.
The authors present P2Echo as an efficient and controllable approach to multi-plane cardiac analysis that unifies multiple echocardiographic views in one model. The model is intended for research in medical image analysis and has not been validated for clinical decision-making.
Computer Science > Computer Vision and Pattern Recognition
arXiv:2611.13425 (cs)
[Submitted on 10 Nov 2026]
Title:Prompt2Echo (P2Echo): Text-Prompted Multi-View Echocardiography Segmentation with One Model
Authors:Moein Heidari, Alvin Kimbowa, Mahdi Tabatabaei, Afshin Bozorgpour, Sina Ghorbani Kolahi, River Jiang, Wenjin Chen, David J. Foran, Jasmine Grewal, Dorit Merhof, Ilker Hacihaliloglu
View a PDF of the paper titled Prompt2Echo (P2Echo): Text-Prompted Multi-View Echocardiography Segmentation with One Model, by Moein Heidari and 10 other authors
#P2Echo#Dynamic Image-Text Alignment#Hugging Face#moein99/P2Echo