Establishment of magnetic resonance imaging 3D reconstruction technology of orbital soft tissue and its preliminary application in patients with thyroid‐associated ophthalmopathy. (6th March 2018)

• Record Type: Journal Article
• Authors: Shen, Jie; Jiang, Wei; Luo, Yaosheng; Cai, Qiuyue; Li, Zhangfang; Chen, Zhiyi; Hu, Shidi; Tang, Lei
• Is Part Of: Clinical endocrinology. Volume 88:Number 5(2018)
• Page Start: 637
• Page End: 644
• Publication Date: 2018-03-06
• Subjects: 3D reconstruction -- orbital soft tissue volume -- thyroid associated ophthalmopathy; Endocrinology -- Periodicals; 616.4005
• DOI: 10.1111/cen.13564
• Languages: English
• ISSNs: 0300-0664

Abstract

Summary: Objective: Effective management of thyroid‐associated ophthalmopathy (TAO) requires precise identification of the disease activity period as it is responsive to immunosuppressive treatment. Quantitative evaluations of orbital soft tissue are useful for analysing disease stages. We aimed to establish a method for orbital soft tissue volume calculation based on magnetic resonance imaging (MRI) data using 3D reconstruction technology. Furthermore, we validated the accuracy and precision of this method and investigated volume differences between patients with TAO and healthy individuals. Materials and Methods: Using Mimics software for 3D reconstruction based on orbital MRI data, we quantitatively measured orbital fat volume (FV) and extraocular muscle volume (MV) using a manual phantom, and in patients with TAO and healthy volunteers (n = 10 each). The phantom was made using a combination of butter and chicken muscle and 2 observers measured its volume. Volume calculations were compared to a previously established standard volume. One observer measured a typical TAO case 10 times to calculate intra‐observer variability while 3 observers independently measured 10 patients with TAO each to calculate interobserver variability. Orbital soft tissue volumes between 10 patients with TAO and 10 healthy individuals were compared. Results: The precision of calculations for the phantom between the 2 observers varied from −4.60% to −2.78% for FV and between −4.13% to 0.71% for MV. Mean differences among repetitive calculations were lower than 4%, except during measurement of MV, which was 5.84%. The intraclass correlation coefficient varied from 0.976 to 0.996. FV was 15.53 ± 3.06 mL in patients with TAO and 11.32 ± 1.68 mL( P = .001)in healthy individuals, while MV was 3.19 ± 0.82 mL in patients with TAO and 2.45 ± 0.57 mL( P = .030)in healthy individuals. Conclusions: This method of calculating orbital soft tissue volumes based on MRI data and 3D reconstruction is both reliable and accurate as it yielded significant differences in tissue volume between patients with TAO and healthy individuals.