MR Spectroscopy in Prostate Cancer: New Algorithms to Optimize Metabolite Quantification

Area of Science:

• Medical Imaging
• Biophysics
• Computational Biology

Background:

• Prostate cancer (PCa) is a leading cause of cancer death, necessitating early, non-invasive diagnostic methods.
• Proton magnetic resonance spectroscopy (1H-MRS) and imaging (1H-MRSI) detect prostate metabolites, but quantification is challenging due to poor signal and citrate J-coupling.
• Existing magnetic resonance spectroscopy (MRS) software is established for brain and liver but not optimized for prostate metabolite analysis.

Purpose of the Study:

• To develop a software prototype for the automatic quantification of key metabolites in prostate tissue.
• To address the challenges in prostate metabolite quantification, including poor signal-to-noise ratio and citrate J-coupling.
• To improve early-stage prostate cancer diagnosis through advanced magnetic resonance techniques.

Main Methods:

• Developed an original fitting routine utilizing a fixed step gradient descent minimization algorithm (FSGD).
• Employed MRS simulations with GAMMA libraries in C++ for accurate prediction of J-modulation and spin systems.
• Validated software accuracy using homemade phantoms and in vivo acquisitions on a Philips Ingenia 3T scanner.

Main Results:

• The software prototype enables automatic quantification of citrate, choline, and creatine in prostate tissue.
• MRS simulations accurately predicted J-modulation under various NMR sequences and coupling parameters.
• Testing on phantoms and healthy volunteers demonstrated the software's performance and accuracy for in vivo applications.

Conclusions:

• The developed software offers a promising tool for automatic prostate metabolite quantification.
• This advancement can aid in the non-invasive, early-stage diagnosis of prostate cancer.
• The study highlights the potential of advanced MRS techniques and computational methods in oncology.