Skin Cancer Classification using Inception Network and Transfer Learning

TL;DR

This paper presents a transfer learning approach using Inception Network to classify imbalanced skin lesion images from HAM10000 with high precision and efficiency.

Contribution

It introduces a novel application of a pretrained convolutional neural network for skin cancer classification on imbalanced dermatoscopic data.

Findings

Achieved high classification accuracy on HAM10000 dataset

Demonstrated low resource requirements for the model

Outlined potential extensions for improved performance

Abstract

Medical data classification is typically a challenging task due to imbalance between classes. In this paper, we propose an approach to classify dermatoscopic images from HAM10000 (Human Against Machine with 10000 training images) dataset, consisting of seven imbalanced types of skin lesions, with good precision and low resources requirements. Classification is done by using a pretrained convolutional neural network. We evaluate the accuracy and performance of the proposal and illustrate possible extensions.