A multicompartment model for intratumor tissue‐specific analysis of DCE‐MRI using non‐negative matrix factorization

Abstract

Purpose: A pharmacokinetic analysis of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data is subject to inaccuracy and instability partly owing to the partial volume effect (PVE). We proposed a new multicompartment model for a tissue-specific pharmacokinetic analysis in DCE-MRI data to solve the PVE problem and to provide better kinetic parameter maps. Methods: We introduced an independent parameter named fractional volumes of tissue compartments in each DCE-MRI pixel to construct a new linear separable multicompartment model, which simultaneously estimates the pixel-wise time-concentration curves and fractional volumes without the need of the pure-pixel assumption. This simplified convex optimization model was solved using a special type of non-negative matrix factorization (NMF) algorithm called the minimum-volume constraint NMF (MVC-NMF). Results: To test the model, synthetic datasets were established based on the general pharmacokinetic parameters. On well-designed synthetic data, the proposed model reached lower bias and lower root mean square fitting error compared to the state-of-the-art algorithm in different noise levels. In addition, the real dataset from QIN-BREAST-DCE-MRI was analyzed, and we observed an improved pharmacokinetic parameter estimation to distinguish the treatment response to chemotherapy applied to breast cancer. Conclusion: Our model improved the accuracy and stability of the tissue-specific estimation of the fractional volumes and kinetic parameters in DCE-MRI data, and improved the robustness to noise, providing more accurate kinetics for more precise prognosis and therapeutic response evaluation using DCE-MRI.

A weakly supervised NMF method to decipher molecular subtype-related dynamic patterns in breast DCE-MR images

“…• KPC: the pixels within a tumor were grouped by their kinetics using k-mean clustering (Fan et al 2018). The number of clusters k was set to three, which has been suggested in previous studies (Chen et al 2011, Xie et al 2021. Then, the tumor was partitioned into three subregions, and the statistical features were extracted from these subregions.…”

“…The decomposition approach performed better than several partitioning approaches for the prediction of breast cancer subtypes on DCE-MRI data. Xie et al (2021) used a minimum-volume constraint nonnegative matrix factorization (NMF) method on pixel time-series curves to decompose a tumor into multiple compartments. The NMF method acquired a more accurate estimation of kinetic parameters than the CAM method on breast DCE-MRI data.…”

Multi‐parametric MRI for radiotherapy simulation

“…47 Moreover, it has been shown that sophisticated multi-compartmental models can be used to extract quantitative parameters based on the same temporal signal enhancement curves. 48 The most commonly derived parameters include K trans (the volume transfer constant of Gd from the intravascular space to the extracellular extravascular space [EES]), K ep (the volume transfer constant of Gd from the EES to the intravascular space), and V e (the fractional volume of the EES). Due to its ability to probe the physiology of the tumor microenvironment, DCE-MRI has shown great potential as a method to evaluate and predict the tumor response to RT and chemotherapy.…”

“…Based on the arterial input function (AIF), which refers to the concentration of tracer in blood‐plasma in an artery measured over time, it is possible to qualitatively assess tissue perfusion and to calculate several semi‐qualitative parameters as surrogates for tissue perfusion, such as the slope of the initial wash‐in curve and the area under the curve 47 . Moreover, it has been shown that sophisticated multi‐compartmental models can be used to extract quantitative parameters based on the same temporal signal enhancement curves 48 . The most commonly derived parameters include K trans (the volume transfer constant of Gd from the intravascular space to the extracellular extravascular space [EES]), K ep (the volume transfer constant of Gd from the EES to the intravascular space), and V e (the fractional volume of the EES).…”