Unequivocal resolution of multiplets in MR spectra for prostate cancer diagnostics achieved by the fast Padé transform

Abstract

Listen

Translate article

We perform mathematical modeling with the fast Pade transform (FPT) according to magnetic resonance (MR)-time signals as encoded in vitro from normal glandular and stromal prostate tissue and from prostate cancer. This is one of the most demanding signal processing problems in MR spectroscopy due to the abundance of diagnostically important multiplets (notably doublet and triplet resonances). The FPT provided exact reconstruction at short acquisition times (i.e. using only a fraction of the full signal length) of all the input spectral parameters for the data corresponding to prostate cancer and to normal glandular as well as stromal prostate tissue. This was achieved without any fitting or numerical integration of peak areas. The converged parametric results remained stable at longer partial signal lengths, including the case using the full signal length. The Pade absorption component spectra yielded unequivocal resolution of all the extracted physical resonances, including multiplet resonances and closely overlapping peaks of different metabolites. The capacity of the FPT to resolve and precisely quantify the physical resonances as encountered in normal tissue from two distinct regions of the prostate, as well as in prostate cancer is demonstrated. The spectra from prostate tissues are dense, which suggests that there is a rich array of metabolic information to be gleaned. The FPT is hereby shown to be optimally suited to retrieve that information. The FPT reliably yields the metabolite concentrations that could be of critical importance for distinguishing non-malignant from cancerous prostate tissue. Pade-optimized MRS could clearly aid prostate cancer diagnostics. This line of investigation will continue with experimentally encoded data from normal, hypertrophic and cancerous prostate tissue, in vitro and in vivo. We anticipate that Pade-optimized MRS will improve the specificity as well as sensitivity of MR-based modalities with respect to prostate cancer. This could have an important impact upon timely and accurate diagnosis of this malignancy, as well as aiding decision-making for therapeutic dilemmas.