skin-cancer-classification

Build a CNN based model which can accurately detect melanoma

A Multimodal Deep Learning Approach for Skin Cancer Classification using ViTs (Visual Transformers)

A comprehensive guide on how to implement and train a Compact Convolutional Transformer model that is specifically designed for classifying skin cancer.

Code and results for a paper on early melanoma skin cancer detection. Implements and compares WGAN, RaLSGAN, and CycleGAN to correct class imbalance in dermatological image datasets, with a VGG16-based classifier used to evaluate downstream detection performance on the balanced data.

A Convolutional neural network (CNN)model to train and detect skin cancer (benign and malignant) disease using DDI(Diverse Dermatology Images)Dataset.

This project uses TensorFlow to implement a Convolutional Neural Network (CNN) for image classification. The goal is to classify skin lesion images into different categories. The dataset used is HAM10000, which contains skin lesion images with associated metadata. The actual accuracy of the model is 90%. 🚀🚀

Skin Cancer Classification using HAM10000 Dataset – a deep learning project employing Convolutional Neural Networks to classify various types of pigmented skin lesions from dermatoscopic images, aiming to assist with early skin cancer detection.

Skin lesion classification using deep learning (HAM10000) with transfer learning and Grad-CAM explainability.

Deep learning-based skin lesion classification using CNNs for early melanoma detection. Trained on the HAM10000 dataset to improve diagnostic accuracy. 🚀 #AI #DeepLearning #SkinCancerDetection

Skin Cancer Lesion Detection

AI-powered skin lesion classifier: EfficientNet-B4 trained on HAM10000 (7 classes), MobileNetV3 skin detector, Grad-CAM visual explanations, GPT-4o-mini diagnosis text, Gradio web UI. Bachelor's thesis in Biomedical Engineering.

Official Fusion Model project for journal article titled "Explainable Depth-Wise and Channel-Wise Fusion Models for Multi-Class Skin Lesion Classification". Implements hybrid CNN-ViT Fusion model with XAI for skin cancer detection.

This repository contains the implementation and supplementary materials for the paper titled "Leveraging Spatial and Semantic Feature Extraction for Skin Cancer Diagnosis with Capsule Networks and Graph Neural Networks."

An automated skin lesion classification web application powered by PyTorch (EfficientNet-B4) and Django. Features explainable AI (Grad-CAM heatmaps), Celery/Redis for async inference, and a modern medical-grade dark UI.

An explainable AI Wrapper framework for automated skin cancer segmentation and classification while integrating custom CNN models across seven skin conditions.

Skin lesion classification using ResNet50 - 75% accuracy