Quantitative MRI: Defining repeatability, reproducibility and accuracy for prostate cancer imaging biomarker...

Area of Science:

• Medical Imaging
• Oncology
• Radiotherapy

Background:

• Quantitative MRI (qMRI) parameters are vital for monitoring prostate cancer treatment response post-radiotherapy.
• High repeatability and reproducibility of qMRI parameters are essential for accurate predictive models.

Purpose of the Study:

• To measure uncertainties in qMRI parameters using commercial and in-house phantoms.
• To guide the development of robust predictive models for prostate cancer treatment monitoring.

Main Methods:

• Acquired ADC, T1, and R2* values on three 3T scanners using prostate-specific qMRI protocol and phantoms.
• Assessed within-session and between-session repeatability, and accuracy of qMRI parameters.
• Utilized B1 field mapping for T1 correction and evaluated parameter uncertainties.

Main Results:

• Repeatability CV for ADC was <2.40%, reproducibility <3.98%, with accuracy between -8.0% to 2.7%.
• B1 correction improved T1 repeatability/reproducibility (p<0.05) but increased accuracy error (p<0.001).
• R2* repeatability was <4.5% and reproducibility <7.3% in the system phantom.

Conclusions:

• This study quantified repeatability, reproducibility, and accuracy of qMRI parameters for prostate cancer.
• Findings will identify robust qMRI parameters for predictive models in clinical trials.
• Results support longitudinal monitoring of prostate cancer treatment response.