A unique anisotropic R₂ of collagen degeneration (ARCADE) mapping as an efficient alternative to composite relaxation metric (R₂‐R_1ρ) in human knee cartilage study

Abstract

Purpose Anisotropic transverse R2 (1/T2) relaxation of water proton is sensitive to cartilage degenerative changes. The purpose is to develop an efficient method to extract this relaxation metric in clinical studies. Methods Anisotropic R2 can be measured inefficiently by standard R2 mapping after removing an isotropic contribution obtained from R1ρ mapping. In the proposed method, named as a unique anisotropic R2 of collagen degeneration (ARCADE) mapping, an assumed uniform isotropic R2 was estimated at magic angle locations in the deep cartilage, and an anisotropic R2 was thus isolated in a single T2W sagittal image. Five human knees from 4 volunteers were studied with standard R2 and R1ρ mappings at 3T, and anisotropic R2 derived from ARCADE on the T2W (TE = 48.8 ms) image from R2 mapping was compared with the composite relaxation (R2 – R1ρ) using statistical analysis including Student's t‐test and Pearson’s correlation coefficient. Results Anisotropic R2 (1/s) from ARCADE was highly positively correlated with but not significantly different from standard R2 – R1ρ (1/s) in the segmented deep (r = 0.83 ± 0.06; 8.3 ± 2.9 vs. 7.3 ± 1.9, P = .50) and the superficial (r = 0.82 ± 0.05; 3.5 ± 2.4 vs. 4.5 ± 1.6, P = .39) zones. However, after eliminating systematic errors by the normalization in terms of zonal contrast, anisotropic R2 was significantly higher (60.2 ± 18.5% vs. 38.4 ± 16.6%, P < .01) than R2 – R1ρ as predicted. Conclusion The proposed anisotropic R2 mapping could be an efficient alternative to the conventional approach, holding great promise in providing both high‐resolution morphological and more sensitive transverse relaxation imaging from a single T2W scan in a clinical setting.

An efficient R1ρ dispersion imaging method for human knee cartilage using constant magnetization prepared turbo‐FLASH

“…Before quantifying R 1ρ dispersion, R 1ρ -weighted 3D images, including an intra-and an inter-series acquired from each subject's unilateral knee, were co-registered using the free software elastix 38 following an established protocol. 23 Then, the femoral, tibial and patellar cartilage was manually outlined using the free software ITK-SNAP, 39 followed by an angular-radial segmentation as previously demonstrated. 23 Note that, for the tibial and patellar cartilage, the angular segmentations were evenly partitioned into five regions of interest (ROIs) horizontally and vertically.…”

“…23 Then, the femoral, tibial and patellar cartilage was manually outlined using the free software ITK-SNAP, 39 followed by an angular-radial segmentation as previously demonstrated. 23 Note that, for the tibial and patellar cartilage, the angular segmentations were evenly partitioned into five regions of interest (ROIs) horizontally and vertically.…”

“…41 In highly ordered biological tissues, S characterizes an intrinsic property of bound water molecular reorientation anisotropy. 24 An REF was determined as described before 23 for each of eight R 1ρ -weighted 3D image datasets, and included in the curve fitting to improve the accuracy of the fits due to an enhanced dynamic range of the measured data. Specifically, these REF signals were considered as those measured using the tailored T SL values (ranging from 13 to 24 ms) and ω 1 /2π = 10 kHz (rather than infinity).…”

“…7,8,19,20 These observations are in accordance with two previous studies from the early 2000s 21,22 as well as with some recent investigations relevant to the underlying R 1ρ relaxation mechanisms in cartilage. 19,20,23 Recently, a theoretical framework of R 1ρ dispersion in cartilage has been outlined, 7 suggesting that an orientation-independent MR metric, named the order parameter, S, 24 can be derived for detecting early collagen degeneration in joint osteoarthritis. Traditionally, the R 1ρ dispersion profile was obtained by collecting a series of R 1ρ mappings by varying ω 1 , with each R 1ρ mapping in turn being created from another series of R 1ρweighted images with varying SL durations (T SL ).…”

“…This extra information, ie S 0 exp −R i 2 T SL , has been referred to as an internal reference (REF) in the literature. 23 To our knowledge, only two quantitative R 1ρ dispersion investigations on the human knee cartilage in vivo at 3 T have been reported in the past. 7,25 The so-called inflection point (ω ip ) on the R 1ρ dispersion curve is determined by setting the second derivative of Equation 2 to zero, lead-…”