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 Experiment Videos

Melanoma segmentation based on deep learning.

Xiaoqing Zhang1

  • 1a College of Information Science and Technology , Donghua University , Shanghai , China.

Computer Assisted Surgery (Abingdon, England)
|October 19, 2017
PubMed
Summary
This summary is machine-generated.

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

Self-heating mitochondrion-induced free radical blast for immunogenic cell death stimulation and HCC immunotherapy.

Journal of controlled release : official journal of the Controlled Release Society·2024
Same author

Autologous Tumor-Infiltrating Lymphocyte Mono-Therapy Can Rapidly Shrank Tumor in Asian Patient with Stage III/IV Cervical Cancer: Two Cases Report.

International journal of women's health·2024
Same author

Superhydrophilic and oleophobic sponges prepared based on Mussel-Inspired chemistry for efficient oil-water separation.

Chemistry, an Asian journal·2024
Same author

Biodegradable, Bifunctional Electro-acoustic Transducers Based on Cellular Polylactic Acid Ferroelectrets for Sustainable Flexible Electronics.

ACS applied materials & interfaces·2024
Same author

LCRNet: local cross-channel recalibration network for liver cancer classification based on CT images.

Health information science and systems·2023
Same author

Multi-style spatial attention module for cortical cataract classification in AS-OCT image with supervised contrastive learning.

Computer methods and programs in biomedicine·2023
Same journal

Toward real-time alignment of 3D CT and 2D X-ray with multi-stage CNNs.

Computer assisted surgery (Abingdon, England)·2026
Same journal

DMRNet: a dynamic multi-scale residual network for Shamrock view and lumbar plexus segmentation.

Computer assisted surgery (Abingdon, England)·2026
Same journal

Association between preoperative serum albumin levels and the incidence of postoperative nausea and vomiting in patients undergoing laparoscopic surgery: a single-center retrospective cohort study.

Computer assisted surgery (Abingdon, England)·2026
Same journal

An automatic detection model for spread through air spaces in postoperative pathological sections based on deep learning in NSCLC.

Computer assisted surgery (Abingdon, England)·2026
Same journal

Leg positioner improves efficiency on selected steps in robotic-assisted total knee arthroplasty: an analysis of surgical workflow efficiency and team experience.

Computer assisted surgery (Abingdon, England)·2026
Same journal

Development of an in silico hybrid model for scoliosis correction using Ansys Motion.

Computer assisted surgery (Abingdon, England)·2026
See all related articles

This study introduces an improved neural network for melanoma image segmentation, enhancing diagnostic accuracy for skin cancer detection. The new framework effectively identifies subtle lesions, reducing errors and improving early diagnosis of malignant melanoma.

Area of Science:

  • Dermatology
  • Artificial Intelligence
  • Medical Imaging

Background:

  • Malignant melanoma is a deadly and rapidly growing skin cancer.
  • Early diagnosis is critical for effective treatment and improved patient outcomes.
  • Current diagnostic methods face challenges in identifying subtle or unknown lesions, leading to potential screening errors.

Purpose of the Study:

  • To develop and validate an improved neural network framework for accurate melanoma image segmentation.
  • To enhance the efficacy of identifying indistinguishable lesions compared to traditional methods.
  • To reduce screening errors and improve overall diagnostic accuracy in skin cancer detection.

Main Methods:

  • Utilized an improved neural network architecture featuring convolution layers, dropout, softmax, filters, and activation functions.
Keywords:
Melanomaconvolutional neural networksdeep learningimage segmentation

Related Experiment Videos

  • Incorporated non-linear activation functions (ReLU, ELU) to mitigate gradient disappearance and RMSprop/Adam optimizers.
  • Added batch normalization layers to stabilize training and prevent gradient issues, with data augmentation via rotation.

    • Main Results:

    • The proposed neural network achieved satisfactory segmentation of melanoma images.
    • Demonstrated improved efficacy in distinguishing between normal and abnormal skin lesions.
    • Experimental results showed higher accuracy in melanoma image segmentation compared to existing approaches.

    Conclusions:

    • The improved neural network framework provides a robust basis for accurate melanoma diagnosis.
    • This AI-driven approach has the potential to significantly reduce diagnostic errors in skin cancer screening.
    • The study highlights the effectiveness of advanced neural networks in medical image analysis for critical diagnoses.