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 17, 2025

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

Annotation-efficient deep learning for automatic medical image segmentation.

Shanshan Wang^1,2,3, Cheng Li⁴, Rongpin Wang⁵

¹Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong, China. ss.wang@siat.ac.cn.

Nature Communications

|October 9, 2021

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

TRPC4/TRPC5 are critical for neuronal modulation by transcranial focused ultrasound in retrosplenial cortex in male mice.

Nature communications·2026

Same author

MRI-based habitat radiomics for preoperative prediction of axillary pathological complete response in breast cancer after neoadjuvant therapy: a multicenter study.

Frontiers in oncology·2026

Same author

Tailoring the Extent of Lymphadenectomy for Esophageal Squamous Cell Carcinoma: Insights From a Comparative Study of Neoadjuvant Chemo-Immunotherapy and Surgery Cohort.

Thoracic cancer·2026

Same author

Discovering proteo-transcriptomic networks via biologically informed heterogeneous graph learning.

Nucleic acids research·2026

Same author

Deep Learning Radiomics of Multiparametric MRI for Individualized Prediction of Axillary Lymph Node Response After Neoadjuvant Chemotherapy in Breast Cancer.

Breast cancer (Dove Medical Press)·2026

Same author

Network model for alignment, stitching and slice-to-volume 3D reconstruction of large-scale spatially resolved slices.

Nature communications·2026

Same journal

Unlocking the capacity of Mn-based Prussian blue cathodes in capacitive deionization.

Nature communications·2026

Same journal

Scaling biodiversity-stability relationships from populations to meta-communities across trophic levels.

Nature communications·2026

Same journal

Thermodynamically programmed one-pot CRISPR platform for point-of-care SNP genotyping.

Nature communications·2026

Same journal

Engineering all-organic electrocatalysts with asymmetric dual-active sites for uncommon oxygen-evolving pathway.

Nature communications·2026

Same journal

Rapid GC content evolution in rice through GC-biased gene conversion and selection for translation efficiency.

Nature communications·2026

Same journal

Declines in organic matter persistence with increased soil carbon.

Nature communications·2026

See all related articles

Summary

This summary is machine-generated.

Annotation-efficient Deep Learning (AIDE) is a new framework for medical image segmentation that works well with limited or imperfect data. It significantly improves efficiency in using expert labels for better biomedical applications.

Area of Science:

Medical Imaging
Artificial Intelligence
Biomedical Engineering

Background:

Deep learning for medical image segmentation requires large, high-quality annotated datasets, which are often unavailable in clinical settings.
This limitation hinders the application of advanced AI models in real-world medical scenarios.

Purpose of the Study:

To introduce Annotation-efficient Deep Learning (AIDE), an open-source framework designed to address the challenge of imperfect training datasets in medical image segmentation.
To demonstrate AIDE's effectiveness in improving segmentation performance with scarce or noisy annotations.

Main Methods:

Developed and evaluated the Annotation-efficient Deep Learning (AIDE) framework.
Conducted methodological analyses and empirical evaluations on open datasets with limited annotations.

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Related Experiment Videos

Last Updated: Oct 17, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Performed a real-life case study on breast tumor segmentation using three datasets from multiple medical centers.

Main Results:

AIDE surpasses conventional fully-supervised models on datasets with scarce or noisy annotations.
In a breast tumor segmentation case study, AIDE achieved segmentation maps comparable to fully-supervised methods using only 10% of training annotations.
Demonstrated a 10-fold enhancement in the efficiency of utilizing expert labels.

Conclusions:

AIDE provides a robust solution for medical image segmentation with imperfect training data.
The framework's efficiency in label utilization has significant potential to advance various biomedical applications.
AIDE offers a practical approach to leveraging AI in clinical settings where data annotation is a bottleneck.