Keras unet and then chromogen analysis for prostate histopathology images.

Script Release, Training and Evaluation Pipeline

Histopathologic Cancer Detection

Histopathology Atlas contains whole slide images of various diseases

Benchmarks for the segmentation of H&E stained histopathology images

This project leverages the DeepVit model, a variant of the Vision Transformer (ViT), for the classification of histopathology images from the BreakHis dataset.

Histopathology of Glycogen Storage Disease in Liver Biopsy

Histopathology of Fat Necrosis

tubular adenoma

Data challenge by Owkin (2023).

Source code for paper: https://ieeexplore.ieee.org/abstract/document/9816283

Flat Tubular Adenoma

Automatic stain unmixing and augmentation for histopathology images in PyTorch

CLI tool to run specimen-level inference on whole slide images

Breast Pathology

Skin cancer misdiagnosis by dermatologists is not uncommon. 25-33% of the skin cancers are incorrectly diagnosed as eczema or another less serious disease. The overall goal is to design and implement an application that can effectively classify whether a patient developed skin cancer based on photographs of skin lesions on dermatologic slides.

Automatic segmentation tool extracted from the BD-PATH website.

Implementation of CHOWDER method for automatic detection of ROI on gigapixel tumor images