Liver PDFF estimation using a multi-decoder water-fat separation neural network with a reduced number of echoes - Summary - MDSpire

Liver PDFF estimation using a multi-decoder water-fat separation neural network with a reduced number of echoes

  • By

  • Juan Pablo Meneses

  • Cristobal Arrieta

  • Gabriel della Maggiora

  • Cecilia Besa

  • Jesús Urbina

  • Marco Arrese

  • Juan Cristóbal Gana

  • Jose E. Galgani

  • Cristian Tejos

  • Sergio Uribe

  • April 4, 2023

  • 0 min

Share

Objective:

To achieve liver PDFF estimation with accuracy comparable to the reference 6-echo graph cut technique using 3-echo images, ensuring that the accuracy matches rather than just approaches.

Key Findings:
  • MDWF-Net allows for accurate PDFF quantification with reduced echo acquisitions, enhancing clinical workflow.
  • The 3-echo protocol significantly decreased scan time from 120 to 54 seconds, improving patient comfort.
  • MDWF-Net maintains accuracy comparable to the gold standard graph cut method, indicating its potential as a reliable alternative.
Interpretation:

The multi-decoder architecture effectively addresses the water-fat separation problem while reducing scan times, making it a promising tool for clinical applications in assessing hepatic fat content, particularly in NAFLD patients.

Limitations:
  • Validation of reduced echo acquisitions was not thoroughly assessed against shorter duration protocols, which may affect reliability.
  • Study conducted at a single center, limiting generalizability to broader populations and clinical settings.
Conclusion:

MDWF-Net demonstrates potential for efficient and accurate liver PDFF estimation, paving the way for improved non-invasive assessments of NAFLD.

Original Source(s)

Related Content