Joint system relaxometry (JSR) and Crámer-Rao lower bound optimization of sequence parameters: A framework for enhanced precision of DESPOT T1 and T2 estimation

Rui Pedro A G Teixeira*, Shaihan J. Malik, Joseph V. Hajnal

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

32 Citations (Scopus)
211 Downloads (Pure)

Abstract

Purpose: This study aims to increase the precision of single-compartment DESPOT relaxometry by two means: (i) a joint system relaxometry (JSR) approach that estimates parameters in a single step using all available data; and (ii) optimizing acquisition parameters by deploying a robust design tool based on the Crámer-Rao lower bound (CRLB). Methods: Following the development of the analysis and design capabilities, phantom and four in vivo subject experiments were performed to compare directly the precision achieved with DESPOT and JSR estimation using published protocols and protocols designed using a proposed CRLB framework. Results: Experimental data demonstrate JSR's ability to decrease relaxometry estimation variance. Phantom results show 72 to 77% improvement using the same data as conventional DESPOT. This is further improved to 81 to 87% using optimal parameters. Both experiments show systematic bias depending on the acquisition parameters used, which are shown to be highly reproducible and to vary with different magnetization transfer conditions. Conclusions: Compared with DESPOT, JSR produces reproducible relaxation maps with improved precision. Further improvement was achieved using CRLB as a protocol design tool. With this combined approach, it is possible to achieve submillimeter maps of ρ,T1,T2, and B0 in an 11-min examination, making the approach appealing for potential clinical use.

Keywords

  • CRLB
  • DESPOT
  • Relaxometry

Fingerprint

Dive into the research topics of 'Joint system relaxometry (JSR) and Crámer-Rao lower bound optimization of sequence parameters: A framework for enhanced precision of DESPOT T1 and T2 estimation'. Together they form a unique fingerprint.

Cite this