A computational measurement of cartilaginous endplate structure using ultrashort time-to-echo MRI scanning. (May 2017)

• Record Type: Journal Article
• Authors: Jin, Ri-Chu; Huang, Yong-Can; Luk, Keith D.K.; Hu, Yong
• Is Part Of: Computer methods and programs in biomedicine. Volume 143(2017)
• Page Start: 49
• Page End: 58
• Publication Date: 2017-05
• Subjects: Ultrashort time-to-echo -- MRI -- Intervertebral disc -- Cartilaginous endplate -- Image segmentation; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2017.02.024
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: A new method based on the UTE data was developed to detect cartilaginous endplate (CEP) of intervertebral disc. Global properties of CEP, e.g., morphology, thickness distribution and signal intensity distribution, were acquired. It is helpful for accurately qualitative and quantitative assessment of the CEP structure after UTE MRI scanning. Abstract: Background and Objective: Ultrashort time-to-echo (UTE) MRI scanning has been applied to observe the cartilaginous endplate (CEP) in intervertebral disc. CEP plays a critical role in IVD health and disease. Nevertheless, current measurements of CEP based on UTE MRI technique are still by manual segmentation, and observation of signal abnormality was usually time-consuming and often disturbed by subjective bias. This study hence proposed an efficient way to harvest the global parameters of CEP after UTE MRI scanning. Methods: E x-vivo UTE-MRI scanning was performed using 12 goat lumbar spine specimens. After the UTE-MRI data were collected, the computational method for CEP segmentation and assessment was developed. Global view of CEP, e.g., surface morphology as well as distributions of thickness and signal intensity, were measured. Histological staining of the CEP as well as manual CEP segmentation was then conducted to validate the accuracy. Results: Segmentation of CEP by the proposed method presented a good agreement with manual measurement, with mean Jaccard index of 0.7296 and mean Cohen's Kappa coefficient of 0.8286. The processing time for CEP segmentation and property measurements was 59.2 s which was much shorter than the manual measurement. Conclusions: This newly-developed technique is able to qualitatively and quantitatively assess the CEP structure, which is very valuable for the clinicians and researchers to accurately evaluate the endplate health after UTE MRI scanning.