Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Oct 3, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Transfer learning for data-efficient abdominal muscle segmentation with convolutional neural networks.

Dónal M McSweeney^1,2, Edward G Henderson^1,2, Marcel van Herk^1,2

¹Division of Cancer Sciences, University of Manchester, Manchester, UK.

Medical Physics

|February 16, 2022

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Associations between body composition and radiotherapy-related side-effects and health-related quality of life in patients with prostate or lung cancer: sub-analysis of the REQUITE trial.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2026

Same author

Bayesian networks as prognostic models in oncology: a systematic review and recommendations for clinical practice.

BMJ oncology·2026

Same author

Causal re-assessment of the spatial association between normal tissue dose and overall-survival in lung radiotherapy.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2026

Same author

Limitations and future directions in identifying risk factors for radiotherapy related pelvic insufficiency fractures in gynecological cancers: A scoping review.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2026

Same author

Assessing body composition using auto-segmentations of muscle and subcutaneous adipose tissue in prostate cancer patients receiving magnetic resonance-guided radiotherapy.

Physics and imaging in radiation oncology·2026

Same author

Using machine learning to predict patient-reported symptom clusters in prostate cancer patients receiving radiotherapy.

Health and quality of life outcomes·2025

Same journal

Correction to "On the shape of the radiation survival curve in tumor spheroids: The role of oxygen heterogeneity".

Medical physics·2026

Same journal

Multi-view constrained semi-supervised vertebra detection for 3D ultrasound spine volume.

Medical physics·2026

Same journal

Accuracy of quantitative <sup>177</sup>Lu SPECT/CT imaging: A systematic review.

Medical physics·2026

Same journal

Physics-constrained dual-domain network for CBCT reconstruction from orthogonal X-rays in gynecologic radiotherapy.

Medical physics·2026

Same journal

Decomposition-based harmonization for quantitative PET imaging across scanners and radiotracers.

Medical physics·2026

Same journal

Development and evaluation of an in vivo dose-based monitoring system for electron FLASH radiation therapy.

Medical physics·2026

See all related articles

This summary is machine-generated.

Transfer learning significantly reduces data needs for skeletal muscle segmentation, achieving human-level performance with minimal annotation. This method enables robust muscle health assessment, even with limited data.

Area of Science:

Medical imaging analysis
Deep learning in radiology
Computational anatomy

Background:

Skeletal muscle segmentation is crucial for sarcopenia assessment, a key indicator of patient frailty.
Current deep learning models for auto-segmentation are hindered by extensive data annotation requirements.

Purpose of the Study:

To explore transfer learning methodologies for adapting segmentation models to new domains with reduced annotation.
To define efficient strategies for applying deep learning models across different anatomical regions or imaging modalities.

Main Methods:

Empirical evaluation of transfer learning source tasks: classification, segmentation, reconstruction, and jigsaw solving.
Training segmentation models on limited annotated CT data from 204 cancer patients.

Keywords:

deep learning sarcopenia segmentation

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Related Experiment Videos

Last Updated: Oct 3, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Cross-validation and inter-observer studies with 10 radiographers to establish human-level performance benchmarks.

Main Results:

Accurate skeletal muscle segmentation models can be trained using a fraction of previously required data.
Models pre-trained on segmentation tasks and fine-tuned on just 10 images achieve performance comparable to trained observers.
The approach yields reliable measures of muscle health and segmentation accuracy.

Conclusions:

Transfer learning enables the development of convolutional neural networks for abdominal muscle segmentation with human-level performance.
This method reduces data requirements by an order of magnitude, overcoming annotation bottlenecks.
Facilitates skeletal muscle assessment in data-scarce anatomical sites and addresses unmet clinical needs.