Assessing Molecular Pathways Driving Conjunctival Melanoma

Abstract

The purpose of this project is 2-fold: perform a comprehensive molecular analysis of conjunctival melanomas and to develop a computer vision model to analyze histopathology images of conjunctival melanomas. The scope is to develop complementary diagnostic aids: a molecular classifier and a convolutional neural network driven computer vision model. We have made rapid progress on the computational aim: we have developed an artificial intelligence model to identify melanocytes in histopathological slides, reported in a published manuscript1 presented at the International Conference on Medical Image Computing and Computer Assisted Intervention(MICCAI 2022). These results are significant in being the first step towards training the computer vision model to distinguish benign from malignant melanocytic neoplasms.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 2022
Accession Number
AD1193670

Entities

People

  • Maria Wei

Organizations

  • Northern California Institute for Research and Education
  • University of California

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Artificial Intelligence Computing
  • Artificial Intelligence Software
  • Automata Theory
  • Cancer
  • Computational Science
  • Computer Vision
  • Computers
  • Convolutional Neural Networks
  • Deep Learning
  • Dermatology
  • Health Services
  • Machine Learning
  • Medical Personnel
  • Neoplasms
  • Neural Networks
  • Physicians
  • Skin Cancer
  • Skin Diseases
  • Training

Readers

  • Molecular and Cellular Biology
  • Neural Network Machine Learning.
  • Oncology and Biomarker-Based Cancer Detection.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks