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

Temporal Trends in Vulvar Cancer Cases and Associated Factors of Adverse Clinical Features: A 6-year Retrospective Study From a Tertiary Hospital in Indonesia.

International journal of women's health·2026
Same author

Expression of miR-92a and miR-125b and Their Association with Chemoradiotherapy Response in Locally Advanced Cervical Cancer.

International journal of molecular sciences·2026
Same author

AI-assisted diagnosis of cervical dysplasia from cervicography images.

Scientific reports·2026
Same author

Exploring Knowledge, Attitudes, Practices, and Information-Seeking on HPV Infection and Cervical Cancer Prevention in Rural Indonesia.

Asian Pacific journal of cancer prevention : APJCP·2026
Same author

The 11th IUPS Workshop for Physiology Education: "Towards a new global horizon in the teaching of Physiology," Löwenstein, Germany, September 14-16, 2025.

Advances in physiology education·2026
Same author

Efficacy and Acceptability of Thermal Ablation in Patients with Positive Documented Visual Inspection with Acetic Acid (Do-VIA) Results.

International journal of women's health·2025

Related Experiment Video

Updated: Sep 3, 2025

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
08:05

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence

Published on: June 10, 2025

191

Automated Precancerous Lesion Screening Using an Instance Segmentation Technique for Improving Accuracy.

Patiyus Agustiansyah1,2, Siti Nurmaini3, Laila Nuranna4

  • 1Doctoral Program, Biology Science, Faculty of Medicine, Universitas Sriwijaya, Palembang 30139, Indonesia.

Sensors (Basel, Switzerland)
|July 28, 2022
PubMed
Summary
This summary is machine-generated.

Automated detection of cervical precancerous lesions using visual inspection with acetic acid (VIA) is crucial for low-resource settings. A Mask R-CNN model accurately segments and detects columnar areas and acetowhite lesions from VIA images, improving diagnostic accuracy.

Keywords:
acetowhite lesionscolumnar areainstance segmentationsquamocolumnar junctionvisual inspection of acetic acid

More Related Videos

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
05:33

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

215
Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.2K

Related Experiment Videos

Last Updated: Sep 3, 2025

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
08:05

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence

Published on: June 10, 2025

191
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
05:33

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

215
Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.2K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence in Healthcare
  • Gynecologic Oncology

Background:

  • The World Health Organization recommends visual inspection with acetic acid (VIA) for cervical precancer screening in low- and middle-income countries (LMICs).
  • VIA screening is often performed by nurses and midwives in LMICs, leading to challenges in accurately identifying human papillomavirus (HPV) infection pathophysiology due to interobserver variability and limited sensitivity/specificity.
  • Accurate automated detection of columnar areas (CA), the squamocolumnar junction (SCJ), and acetowhite (AW) lesions is essential to support diagnosis and overcome limitations of manual VIA interpretation.

Purpose of the Study:

  • To propose and evaluate a Mask R-CNN based deep learning architecture for simultaneous segmentation, classification, and detection of CA and AW lesions in VIA cervicograms.
  • To develop a computational tool that assists medical workers in low-resource settings for more accurate cervical precancer screening.
  • To improve the reliability and reduce interobserver variance in VIA-based cervical cancer screening.

Main Methods:

  • A Mask R-CNN architecture was employed for instance segmentation, classification, and detection tasks.
  • The model was trained and tested using a dataset comprising 262 VIA-positive and 222 VIA-negative cervicograms.
  • Performance was evaluated using metrics such as Intersection over Union (IoU), Dice Similarity Coefficient (DSC), mean Average Precision (mAP), sensitivity, and specificity.

Main Results:

  • The proposed Mask R-CNN model achieved satisfactory performance with an IoU of approximately 63.60% for CA and 73.98% for AW lesions.
  • Dice Similarity Coefficients were reported as 75.67% for CA and 80.49% for AW lesions.
  • The model demonstrated high accuracy in detecting cervical cancer precursor lesions, with mAP of 86.90% for CA and 100% for AW lesions, alongside 100% sensitivity and 92% specificity.

Conclusions:

  • The developed Mask R-CNN model effectively segments, detects, and classifies cervical cancer precursor lesions from VIA cervicograms using an instance segmentation approach.
  • This automated detection system shows significant potential to support accurate diagnosis and improve cervical cancer screening in resource-limited settings.
  • The findings suggest that AI-powered analysis of VIA images can enhance diagnostic performance and overcome limitations associated with manual interpretation by non-specialist healthcare providers.