MOOSE (Multi-organ objective segmentation) a data-centric AI solution that generates multilabel organ segmentations to facilitate systemic TB whole-person research.The pipeline is based on nn-UNet and has the capability to segment 120 unique tissue classes from a whole-body 18F-FDG PET/CT image.

Clinical evaluation of deep learning for tumour delineation on [18F]FDG-PET/CT of head and neck cancer

Cascade model ( Tumor appearance extraction + multi-channel nnUNet ) for AutoPET Challenge 2022

This script reads DICOM files in a source directory or in a list of source directories and searches for the patients in the given patients' list creates a DICOM DataBase in the destination directory, copies the files, and creates a DicomDataBase.csv file and a summary.txt file.

Deploying STIR on Azure via Terraform

This repository presents a radiodosiomics framework for personalized [¹⁷⁷Lu]Lu-PSMA-617 RLT in mCRPC. It includes feature selection and ML models using clinical, radiomic, and dosiomic features, plus nnU-Net and Swin UNETR DL models with SSL to predict Monte Carlo–based dose rate maps.

MOOSE (Multi-organ objective segmentation) a data-centric AI solution that generates multilabel organ segmentations to facilitate systemic TB whole-person research.The pipeline is based on nn-UNet and has the capability to segment 120 unique tissue classes from a whole-body 18F-FDG PET/CT image.

Machine learning for non small cell lung cancer nodal staging from PETCT molecular imaging

A project with the aim of predicting future PETCT scanning referrals based on time-series analysis of past referrals.

Code repository for the paper entitled "Segmentation-Free Outcome Prediction in Head and Neck Cancer: Deep Learning-based Feature Extraction from Multi-Angle Maximum Intensity Projections (MA-MIPs) of PET Images" published in "Cancers" journal.