Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

From Slice to Sequence: Autoregressive Tracking Transformer for Consistent 3D Lymph Node Detection in CT Scans.

IEEE transactions on medical imaging·2026
Same author

Clinical Knowledge-Guided PET/CT Lesion Segmentation With Interpretable Fusion of Metabolic and Structural Cues.

IEEE transactions on medical imaging·2026
Same author

Preoperative Prediction of Esophageal Cancer Survival in CT via Tumor and Lymph Node Context and Geometry Modeling.

IEEE transactions on medical imaging·2026
Same author

Clinical Validation of a Deep Learning-Based 2D Ultrasound Steatosis Algorithm: Cutoff Transferability, Scanner Generalizability, and Comparison with FibroScan.

Diagnostics (Basel, Switzerland)·2026
Same author

Pretreatment CT Identification of Extranodal Extension in Laryngeal and Hypopharyngeal Cancers Using Deep Learning.

Radiology·2026
Same author

DistAL: A Domain-Shift Active Learning Framework With Transferable Feature Learning for Lesion Detection.

IEEE transactions on medical imaging·2025

Related Experiment Video

Updated: Nov 24, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

3.1K

Learning From Multiple Datasets With Heterogeneous and Partial Labels for Universal Lesion Detection in CT.

Ke Yan, Jinzheng Cai, Youjing Zheng

    IEEE Transactions on Medical Imaging
    |December 28, 2020
    PubMed
    Summary

    This study introduces a novel deep learning framework, Lesion ENSemble (LENS), to improve lesion detection in medical imaging by effectively handling incomplete and varied datasets. LENS significantly enhances detection accuracy, achieving a 49% improvement in average sensitivity.

    More Related Videos

    Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
    10:25

    Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

    Published on: September 25, 2019

    48.9K
    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
    07:15

    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

    Published on: August 16, 2020

    7.2K

    Related Experiment Videos

    Last Updated: Nov 24, 2025

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

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

    Published on: November 30, 2022

    3.1K
    Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
    10:25

    Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

    Published on: September 25, 2019

    48.9K
    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
    07:15

    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

    Published on: August 16, 2020

    7.2K

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Deep Learning

    Background:

    • Accurate deep learning models require large, high-quality labeled datasets, which are scarce in medical imaging due to annotation costs.
    • Existing medical imaging datasets often suffer from partial labeling (missing annotations) or heterogeneous label scopes (different lesion types across datasets).
    • These data limitations, exemplified by datasets like DeepLesion, LUNA, and LiTS, hinder the development of universal lesion detection algorithms.

    Purpose of the Study:

    • To develop a universal deep learning algorithm for detecting a variety of lesions in medical images.
    • To address the challenges posed by heterogeneous label scopes and partial labeling in existing datasets.
    • To improve the performance of lesion detection models by leveraging multiple datasets and mining missing annotations.

    Main Methods:

    • Developed Lesion ENSemble (LENS), a framework for multi-task learning from heterogeneous lesion datasets using proposal fusion.
    • Implemented strategies to mine missing annotations from partially-labeled datasets by incorporating clinical prior knowledge and cross-dataset knowledge transfer.
    • Trained the LENS framework on four public lesion datasets and evaluated its performance on manually-labeled DeepLesion sub-volumes.

    Main Results:

    • The LENS framework demonstrated significant improvements in lesion detection.
    • Achieved a 49% relative improvement in average sensitivity compared to the current state-of-the-art approach.
    • Publicly released manual 3D annotations for DeepLesion to facilitate further research.

    Conclusions:

    • The proposed LENS framework effectively addresses the challenges of heterogeneous and partial labels in medical imaging datasets.
    • This approach enables more accurate and universal lesion detection, advancing the capabilities of deep learning in radiology.
    • The study contributes valuable annotated data and a robust methodology for future research in medical image analysis.