GastroVision: A Multi-class Endoscopy Image Dataset for Computer Aided Gastrointestinal Disease Detection (https://osf.io/84e7f/)

A multi-centre polyp detection and segmentation dataset for generalisability assessment https://www.nature.com/articles/s41597-023-01981-y

Noise Robust Learning with Hard Example Aware for Pathological Image classification

Official Implementation of our paper "Supervision meets Self-Supervision: A Deep Multitask Network for Colorectal Cancer Histopathological Analysis" [Best Paper Award at MISP 2022]

DL-model for multi-class tissue segmentation in colorectal cancer H&E slides, developed as part of the SemiCOL2023 Challenge.

Colorectal cancer risk mapping through Bayesian Networks

Decision model for colorrectal cancer screening. Based on bayesian networks and influence diagrams

Diagnosing colorectal cancer from histopathology images using deep learning: final project code.

UNSUPERVISED MACHINE LEARNING (CLUSTERING): TCGA data mining for studying the system of interactions between sub-branches of Wnt signalling pathway in colorectal cancer

Transfer learning & fine-tuning in Tensorflow for classification of textures in colorectal cancer histology

This repository contains all machine learning and statistical models used to analyze the landscape of colorectal cancer.

VDAC1 Gate-Opening Therapeutic Stack for MSS Colorectal Cancer — SAD v4.0 with complete bench protocol. DOI: 10.17605/OSF.IO/4KNQR

A production-grade computational pathology model (EfficientNetB1) for 9-class colorectal tissue classification, achieving 92.7% unbiased holdout accuracy.

Classifying images from the MNIST colorectal histology dataset

This project predicts the five-year survival rates of colorectal cancer patients using a Random Forest machine learning model.

A machine learning pipeline that predicts colorectal cancer from gut microbiome 16S rRNA sequencing data at the genus level.

It contains the requirements, code, and backlog of my bioinformatics master's thesis.

End-to-end AI-assisted CAD system for colorectal cancer detection from H&E histopathology images | EfficientNet-B0 | Grad-CAM | Streamlit | 99.93% accuracy