About this document
Neural Classifiers for Melanoma Detection by Oana-Liliana Zamfir is a document available to read on EtoBox.
This document summarizes a study that used neural networks to classify skin tumors as benign, dysplastic, or malignant melanoma based on color and texture features extracted from photographs. A neural network was trained on 180 images and tested on 60 images. The network achieved 76.0% accuracy for detecting melanoma cases. Pruning techniques were applied to optimize the network architecture and reduce overfitting, resulting in improved generalization performance compared to the fully connected network. Key
- Author
- Oana-Liliana Zamfir
- Language
- EN