Uncertainty-Aware, End-to-End Deep Learning for Functional Lung MRI Quantification Using ¹²⁹ Xe and ¹ H MRI

Joshua Astley; Helen Marshall; Laurie Smith; Alberto Biancardi; Martin Brook; Guilhem Collier; Paul Hughes; Laura Saunders; Jim M. Wild; Bilal Tahir

doi:10.1148/ryct.250371

June 1, 2026Radiology Cardiothoracic Imaging

Uncertainty-Aware, End-to-End Deep Learning for Functional Lung MRI Quantification Using ¹²⁹ Xe and ¹ H MRI

View at Publisher

Ask AI

Bookmark

Share

Key Points

The aim is to develop an end-to-end deep learning model for quantifying functional lung MRI data without the need for complex segmentation methods.
Utilized an end-to-end deep learning framework for functional lung MRI quantification.
Evaluated performance across various pathologies and hospitals using test-time augmentations.
Focused on uncertainty-aware techniques to enhance the reliability of quantification.
Achieved accurate quantification across diverse pathologies.
Demonstrated reliable performance without the need for multimodal segmentation.
Exhibited effectiveness in various hospital settings, indicating broad applicability.

Abstract

An end-to-end, uncertainty-aware deep learning approach enabled direct quantification of functional lung MRI without multimodal segmentation, performing accurately across a range of pathologies, hospitals, and test-time augmentations.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Astley et al. (2026) studied this question.

synapsesocial.com/papers/6a363172db0793dc1a5384c8 https://doi.org/10.1148/ryct.250371

View at Publisher

Ask AI

Bookmark

Share

Uncertainty-Aware, End-to-End Deep Learning for Functional Lung MRI Quantification Using ¹²⁹ Xe and ¹ H MRI

Key Points

Abstract

Discussion

Cite This Study

Uncertainty-Aware, End-to-End Deep Learning for Functional Lung MRI Quantification Using ¹²⁹ Xe and ¹ H MRI

Key Points

Abstract

Discussion

Cite This Study

Keep reading

Keep reading