HAM10000 Classification

Kalbe Digital Lab

Overview

The HAM10000 Skin Cancer Classification program serves the critical purpose of categorizing skin lesions into seven distinct classes: Actinic Keratosis, Basal Cell Carcinoma, Benign Keratosis, Dermatofibroma, Melanoma, Melanocytic Nevi, and Vascular Lesion. This program is trained using a ResNet18 model.

Dataset

The program has been meticulously trained on a robust and diverse dataset, specifically Skin Cancer : HAM10000 Dataset..
The program has been meticulously trained on a diverse and extensive dataset, known as the HAM10000 dataset. This dataset consists of 10,015 dermatoscopic images of skin lesions, meticulously collected and manually annotated. It is a valuable resource that has played a pivotal role in training our skin cancer classification model.

  • Objective: HAM 10000 Identification
  • Task: Classification
  • Modality: Colour Images

Model Architecture

The model architecture of ResNet18 to train images for classifying skin cancer part.

model-architecture

Demo

Please select or upload a skin cancer scan image to see the capabilities of skin part classification with this model