Early Detection of Diabetic Kidney Disease Using Contrast-Enhanced Ultrasound Perfusion Parameters

Overview

This repository contains research related to the early detection of Diabetic Kidney Disease (DKD) using Contrast Enhanced Ultrasound Perfusion Parameters. The project explores the use of different perfusion models and their application to diabetic and control examples.

Research Presentation

The research is presented in a document titled "Towards Early Detection of Diabetic Kidney Disease Using Contrast Enhanced Ultrasound Perfusion Parameters." The document provides insights into the importance of early DKD detection, the limitations of current markers, and a comparison of CT, MRI, and Ultrasound for renal perfusion measurement.

Contents

'Towards Early Detection of Diabetic Kidney Disease Using Contrast-Enhanced Ultrasound Perfusion Parameters.pdf': a PowerPoint presentation of the research

Perfusion Models

The research evaluates three perfusion models:

Lagged Normal

Log-Normal

Gamma Variate

Parameters studied include Peak Enhancement (PE), Time to Peak (TP), Mean Transit Time (MTT), Rise Time (RT), Fall Time (FT), Wash-in Rate (WiR), Wash-out Rate (WoR), WiAUC (Wash-in Area Under the Curve), and WoAUC (Wash-out Area Under the Curve).

Discussion Highlights

Clear differences observed between Lagged and Log-Normal models.

Gamma Variate and Log-Normal show more similarity.

Example-specific observations for diabetic and control cases.

Future Work

Apply models to a larger sample size for more robust conclusions.

Consideration: Should the focus be on minimizing RMSE or fitting data to the parameters?