ham10000

[MICCAI 2023] DermoSegDiff: A Boundary-aware Segmentation Diffusion Model for Skin Lesion Delineation

[MICCAI 2025] CENet: Context Enhancement Network for Medical Image Segmentation

FixCaps: An Improved Capsules Network for Diagnosis of Skin Cancer，DOI: 10.1109/ACCESS.2022.3181225

[ICML 2023] Official repository of paper: Dividing and Conquering a BlackBox to a Mixture of Interpretable Models: Route, Interpret, Repeat

The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions.

This repo includes classifier trained to distinct 7 type of skin lesions

PyTorch code to reproduce the key experiments and results presented in the paper: ELMAGIC: Energy-Efficient Lean Model for Reliable Medical Image Generation and Classification Using Forward Forward Algorithm.

Multiclass skin cancer detection using explainable AI for checking the models' robustness

Data quality analysis of DermaMNIST (MedMNIST), HAM10000, and Fitzpatrick17k datasets

HAM10000 image dataset classification using Pytorch and Scikit Learn

Skin cancer detection using CNNs with data analysis.

Cross-platform smartphone app capable of detecting skin cancer lesions using Computer Vision.

Deep Learning based Skin Cancer Detection using multiple CNN architectures (VGG, ResNet, DenseNet, EfficientNet, Inception) with image preprocessing using ESRGAN and performance comparison for clinical AI research.

Terminal application to perform skin lesion segmentation & classification

Discover DermaScan: A full-stack web app with MobileNetV2-based skin lesion classifier using Harvard's Ham10000 Dataset for precise dermatological diagnosis.

Convolutional neural network capable of identifying skin lesions (based on the skin lesion image data set HAM10000).

EffiDerm: An Efficient Deep Learning Model for Skin Cancer Prediction

This is a project that I worked on with my colleagues in the 6th Semester of my B.tech. In this project, we present a fully automatic method for skin lesion segmentation by leveraging UNet and FCN that is trained end to-end. For Skin lesion disease classification, we use a customized convolutional neural net. Designing a novel loss function base…

CNN ensemble for skin lesion classification — BACC 0.846 ± 0.009 · Clinical threshold calibration · ResNet-50 + DenseNet-121 + EfficientNet-B3 · ISIC 2018 Task 3

3-class dermoscopic lesion ensemble (EfficientNet-B0 + ResNet-50 + DenseNet-201 + CBAM). Lesion-level split, malignant operating points. Research only — not for diagnosis. Weights not released.