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

A novel water-soluble near-infrared fluorescent probe for monitoring viscosity fluctuations in plants and zebrafish under abiotic stresses.

Smart molecules : open access·2026
Same author

Chemodivergent aminocarbonylation enabled by oxygen vacancy-engineered Pd-doped In<sub>2</sub>O<sub>3</sub> nanocatalysts.

Science advances·2026
Same author

Study on tensile mechanical response and microstructure of polypropylene fiber reinforced loess under freezing.

PloS one·2026
Same author

Global plastics treaty must be built on a foundation of monitoring.

Nature·2026
Same author

Pituitary adenomas associated with hydrocephalus: clinical characteristics, risk stratification, and clinical management.

Journal of neuro-oncology·2026
Same author

Covalently Hydrophobic Nanocarbon Supported Ni Single-Atom Catalysts for Highly Selective CO<sub>2</sub> Electroreduction.

Angewandte Chemie (International ed. in English)·2026

Related Experiment Video

Updated: May 26, 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

2.6K

A novel lightweight multi-scale feature fusion segmentation algorithm for real-time cervical lesion screening.

Jiahui Yang1, Ying Zhang2, Wenlong Fan1

  • 1College of Quality and Technical Supervision, Hebei University, Baoding, 071002, China.

Scientific Reports
|February 21, 2025
PubMed
Summary

A new lightweight AI algorithm, Light-MDDNet, enables rapid and accurate segmentation of cervical lesions in colposcopy images. This advancement holds significant potential for improving real-time cervical cancer screening efficiency.

Keywords:
Colposcopy imagesDeep learningImage segmentationLightweight

More Related Videos

Author Spotlight: Integrating Ultrasound Imaging with Biochemical Markers for Thyroid Disease Diagnosis
05:41

Author Spotlight: Integrating Ultrasound Imaging with Biochemical Markers for Thyroid Disease Diagnosis

Published on: February 9, 2024

518
Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging
06:08

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Published on: May 5, 2011

16.7K

Related Experiment Videos

Last Updated: May 26, 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

2.6K
Author Spotlight: Integrating Ultrasound Imaging with Biochemical Markers for Thyroid Disease Diagnosis
05:41

Author Spotlight: Integrating Ultrasound Imaging with Biochemical Markers for Thyroid Disease Diagnosis

Published on: February 9, 2024

518
Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging
06:08

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Published on: May 5, 2011

16.7K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Computer Vision

Background:

  • Cervical cancer screening relies on accurate lesion identification.
  • Current AI segmentation methods for cervical lesions are often too complex for real-time mass screening.

Purpose of the Study:

  • To develop a lightweight AI algorithm for efficient and accurate cervical lesion segmentation in colposcopy images.
  • To deploy and evaluate the algorithm on an edge device for real-time screening.

Main Methods:

  • Proposed a novel lightweight encoder-decoder segmentation framework, Light-MDDNet.
  • Utilized MobileNetV2 and DenseASPP modules for feature extraction and a multi-scale feature fusion (MFF) module.
  • Deployed the model on a JETSON ORIN NX edge device for real-time testing.

Main Results:

  • Achieved a pixel mean pixel accuracy (MPA) of 94.96% with an average speed of 19.60ms per image on a dataset of 971 images.
  • Demonstrated sustained accuracy and reduced interference speed (31.57ms/image) after mobile terminal deployment.
  • Outperformed existing state-of-the-art segmentation networks in accuracy and speed.

Conclusions:

  • Light-MDDNet offers an optimal balance between accuracy and speed for cervical lesion segmentation.
  • The proposed lightweight model shows great potential for practical deployment in mass cervical cancer screening systems.