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

Long non-coding RNA DSCAM-AS1 promotes the progression of gastric cancer via targeting miR-101-3p.

Panminerva medica·2019
Same author

A simultaneous grafting/vinyl polymerization process generates a polycationic surface for enhanced antibacterial activity of bacterial cellulose.

International journal of biological macromolecules·2019
Same author

Coexistence Of A Huge Venous Thromboembolism And Bleeding Tendency In Cytokine Release Syndrome During CAR-T Therapy.

OncoTargets and therapy·2019
Same author

A rapid method for on-line solid-phase extraction and determination of dioscin in human plasma using a homemade monolithic sorbent combined with high-performance liquid chromatography.

Analytical and bioanalytical chemistry·2019
Same author

Cranberry extracts promote growth of Bacteroidaceae and decrease abundance of Enterobacteriaceae in a human gut simulator model.

PloS one·2019
Same author

SCN4A p.R675Q Mutation Leading to Normokalemic Periodic Paralysis: A Family Report and Literature Review.

Frontiers in neurology·2019

Related Experiment Video

Updated: Sep 1, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K

Object Detection in Medical Images Based on Hierarchical Transformer and Mask Mechanism.

Yuntao Shou1, Tao Meng1, Wei Ai1

  • 1School of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410082, Hunan, China.

Computational Intelligence and Neuroscience
|August 15, 2022
PubMed
Summary

This study introduces the MS Transformer, an advanced object detection model for medical imaging. It effectively addresses challenges like low resolution and small object sizes, improving diagnostic accuracy.

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

488
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

2.9K

Related Experiment Videos

Last Updated: Sep 1, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

488
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

2.9K

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Artificial Intelligence

Background:

  • Object detection in medical imaging is vital for computer-aided diagnosis and detection.
  • Existing methods struggle with low resolution, high noise, and small object sizes in medical images.

Purpose of the Study:

  • To propose a novel algorithmic model, the MS Transformer, for enhanced medical object detection.
  • To improve the accuracy and robustness of object detection in challenging medical image conditions.

Main Methods:

  • Utilized a self-supervised learning approach with random masking for feature reconstruction and noise reduction.
  • Introduced a hierarchical transformer model with a sliding window and local self-attention mechanism to focus on small objects.
  • Employed a single-stage object detection framework for bounding box and class prediction.

Main Results:

  • The MS Transformer demonstrated improved performance across multiple evaluation metrics on the DeepLesion and BCDD datasets.
  • The model effectively learned richer feature vectors and filtered out excessive noise.
  • Enhanced attention was given to small objects, a common challenge in medical imaging.

Conclusions:

  • The proposed MS Transformer model offers a significant advancement in medical object detection.
  • The approach effectively handles low resolution, noise, and small object sizes, crucial for clinical applications.
  • This work contributes to the development of more reliable computer-aided diagnosis and detection systems.