Related Experiment Videos
An automatic approach for estimating bolus arrival time in dynamic contrast MRI using piecewise continuous regression
1School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798.
Physics in Medicine and Biology
|April 17, 2003
Summary
We developed two regression models to automatically estimate bolus arrival times (BATs) in dynamic contrast MRI. Our simulations show accurate BAT estimation even with significant noise.
Area of Science:
- Medical Imaging
- Radiology
- Biomedical Engineering
Background:
- Dynamic contrast-enhanced MRI is crucial for assessing tissue perfusion.
- Accurate estimation of bolus arrival time (BAT) is essential for quantitative analysis.
- Current methods for BAT estimation can be sensitive to noise and artifacts.
Purpose of the Study:
- To introduce two novel regression models for automated BAT estimation.
- To evaluate the performance of these models in dynamic contrast MRI datasets.
- To assess the robustness of the models against image noise.
Main Methods:
- Development of two distinct regression models for BAT estimation.
- Utilized dynamic contrast-enhanced MRI datasets for model training and validation.
- Employed Monte Carlo simulations to rigorously test model performance under varying noise levels.
Main Results:
- The proposed regression models successfully automate BAT estimation.
- Monte Carlo simulations demonstrated that estimated BATs are within the sampling interval.
- Model accuracy is maintained even in the presence of significant image noise.
Conclusions:
- The developed regression models offer a robust and automated solution for BAT estimation.
- These models show potential for improving the reliability of quantitative analysis in dynamic contrast MRI.
- The findings suggest improved diagnostic accuracy in perfusion imaging.