Orbital computed tomography imaging characteristics of thyroid-associated ophthalmopathy

Abstract

This study aims to investigate the distribution characteristics and correlations of orbital soft tissue volume expansion in patients with thyroid-associated ophthalmopathy (TAO) by analyzing orbital computed tomography (CT) data and to determine the most appropriate parameters and corresponding cut-off values for imaging classification. Patients with TAO who met the inclusion and exclusion criteria, along with those with lacrimal duct obstruction requiring orbital CT examination, were included in the study. Raw CT data were imported into Mimics and RadiAnt software for analysis, and measurements of orbital volume, orbital fat volume, extraocular muscle volume, optic nerve volume, and exophthalmos were obtained. The results demonstrated that the combination of Mimics and other softwares for processing orbital CT scans allows for the quantitative analysis of various soft tissue volumes within the orbit. The MV/FV ratio was found to effectively reflect the relative changes in extraocular muscle volume, serving as a valuable indicator for identifying different subtypes of TAO and providing significant clinical reference value.

Keywords: Computerized tomography; Three-dimensional reconstruction; Thyroid associated ophthalmopathy.

Fig. 1

Computed tomography (ct) scanning measurement exophthalmos schematic diagram.

Fig. 2

Measurement of orbital volume. The inner wall of the orbit was filled and the orbit shape was semi-automatically outlined in three planes layer by layer according to each boundary auxiliary line. The three-dimensional orbit was obtained by calculating the mask to obtain the orbital volume data.

Fig. 3

Measurement of orbital fat volume. A mask was established according to the adult fat threshold (-200 to -20), and the mask was removed to obtain a three-dimensional model of orbital fat to obtain orbital fat parameters.

Fig. 4

Measurement of extraocular muscle volume. According to the adult muscle density, the oblique muscle and blood vessels of the eyeball were semi-automatically removed in the three-dimensional plane, and the boundaries of the four rectus muscles (inner, outer, superior and inferior) were carefully delineated. The mask was calculated to establish a three-dimensional model to obtain the volume parameters of the extraocular muscles.

Fig. 5

Measurement of optic nerve volume. The optic nerve mask was obtained by Boolean operation, and was checked layer by layer in the three-dimensional plane. The volume parameters of the optic nerve were obtained by calculating the mask to obtain the stereoscopic model of the optic nerve.

Fig. 6

Differences in orbital CT measurement parameters between the normal group and TAO groups according to the severity. N: control group; m-T: mild TAO group; ms-T: moderate to severe TAO group; st-T: very severe TAO group; nsp > 0.05, * p < 0.05, ** p < 0.01, *** p < 0.001.

Fig. 7

Differences in CT measurement parameters among the three groups. N: control group; a-TAO: active TAO group; ia-TAO: inactive TAO group; nsp > 0.05, * p < 0.05, ** p < 0.01, *** p < 0.001.

Fig. 8

Heat map of Pearson correlation analysis in the control group. The Age of the patients was the same. OV: orbital volume; FV: orbital fat volume; MV: extraocular muscle volume; Blue represents a positive correlation coefficient, red represents a negative correlation coefficient, the darker the color, the higher the correlation degree. * p < 0.05, ** p < 0.01, *** p < 0.001.

Fig. 9

Heat map of correlation between parameters in TAO group. Duration of illness; -Sheldon: The Severity of the disease is severe. -Sheldon: Activity; TRAb: thyroid stimulating antibody; TSH: thyroid stimulating hormone; OV: orbital volume; FV: orbital fat volume; NV: optic nerve volume; MV: extraocular muscle volume; EXO: exophthalmos; Blue represents a positive correlation coefficient, red represents a negative correlation coefficient, the darker the color, the higher the correlation degree. * p < 0.05, ** p < 0.01, *** p < 0.001.