Time-efficient estimation of the magnetic resonance dispersion model parameters for quantitative assessment of angiogenesis. (April 2016)

• Record Type: Journal Article
• Authors: Turco, Simona; Janssen, Augustus J.E.M.; Lavini, Cristina; de la Rosette, Jean J.; Wijkstra, Hessel; Mischi, Massimo
• Is Part Of: Biomedical signal processing and control. Volume 26(2016)
• Page Start: 23
• Page End: 33
• Publication Date: 2016-04
• Subjects: Dynamic contrast enhanced magnetic resonance imaging -- Physiological modeling -- Parameter estimation -- Cancer angiogenesis -- Computational efficiency; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2015.11.006
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract : Highlights: Magnetic resonance dispersion imaging (MRDI) permits detection of cancer angiogenesis. Analytical solutions of the dispersion model in frequency and time domain are derived. Three methods are proposed to improve the computational efficiency of MRDI. A reduction of computation time of about 50 Abstract: The limitations of the available imaging modalities for prostate cancer (PCa) localization result in suboptimal protocols for management of the disease. In response, several dynamic contrast-enhanced imaging modalities have been developed, which aim at cancer detection through the assessment of the changes occurring in the tumor microenvironment due to angiogenesis. In this context, novel magnetic resonance dispersion imaging (MRDI) enables the estimation of parameters related to the microvascular architecture and leakage, by describing the contrast agent kinetics with a dispersion model. Although a preliminary validation of MRDI on PCa has shown promising results, parameter estimation can become burdensome due the convolution integral present in the dispersion model. To overcome this limitation, in this work we provide analytical solutions of the dispersion model in the time and frequency domains, and we implement three numerical methods to increase the time-efficiency of parameter estimation. The proposed solutions are tested for PCa localization. A reduction by about 50% of computation time could be obtained, without significant changes in the estimation performance and in the clinical results. With the continuous development of new technological solutions to boost the spatiotemporal resolution of DCE-MRI, solutions to improve the computational efficiency of parameter estimation are highly required.