[Application of surface landmark positioning in assisting cone-beam CT scanning of temporomandibular joint]

Objective: To quantitatively measure the spatial relationship between the temporomandibular joint (TMJ) and surface landmarks such as the tragus and lateral canthus using cone-beam CT (CBCT), as to provide guidance for accurate positioning of CBCT scanning of the TMJ. Methods: DICOM format data from 112 patients (35 males and 77 females with 224 TMJs) were included in this study. The patients were between 12 and 66 years old, with a mean age of (25.6± 9.8) years and they underwent initial visits at the Department of Stomatology, General Hospital of Chinese PLA. CBCT images were imported into Mimics Medical 21.0 software for three-dimensional reconstruction. The distance between selected surface landmarks and corresponding projection points on the same side of the TMJ were measured in both the sagittal and coronal planes. In the frontal view, the distance from the lateral canthus to the perpendicular line passing through the center of the condyle (d^x). In the lateral view, the vertical distance from the horizontal line through the rear edge convex of tragus to the roof of glenoid fossa (d^y1); the vertical distance from the horizontal line through the rear edge convex of tragus to sigmoid notch (d^y2); the distance from the tragus to the perpendicular line through the center of the condyle (d^z). The results were compared between different genders and age groups (adolescent group aged 12 to 18 years, 33 cases; adult group aged>18 years, 79 cases). To verify the reliability and stability of the body surface landmarks obtained in this study, a comparison was made between the use of scout view positioning and the auxiliary positioning method that utilizes body surface landmarks to capture the mandibular condyle in a small field of view CBCT of the TMJ (scout view group, 25 cases; surface landmark-assisted positioning group, with 25 cases aged 12-18 years and 25 cases older than 18 years, totaling 50 cases), with regard to the deviation distance from the central point of the field of view to the central point of the mandibular condyle. Results: A total of 112 patients with 224 TMJs were included in this study. In the frontal view, d^x was (8.59±3.13) mm, with no significant difference between the left and right sides, between males and females (P>0.05). Notably, the d^x in the adolescent group [(7.43±3.02) mm] was significantly smaller than that in the adult group [(9.07±3.06) mm] (t=-3.68, P<0.001). In the lateral view, d^y1 was (14.80±3.90) mm, showing no significant difference between the left and right sides, males and females, or different age groups(P>0.05). Similarly, d^y2 was (6.82±3.95) mm, with no significant differences between the left and right sides or between males and females(P>0.05). However, the d^y2 in the adolescent group [(5.88±4.13) mm] was significantly smaller than that in the adult group [(7.22±3.83) mm] (t=-2.33, P=0.021). Regarding d^z, the measurement was (11.73±3.16) mm, with no significant difference between the left and right sides, between males and females (P>0.05). Meanwhile, the d^z in the adolescent group[(10.92± 2.74) mm] was also significantly smaller than that in the adult group [(12.07±3.27) mm] (t=-2.52, P=0.012). The verification results showed that both positioning methods could fully display the TMJ. The eccentricity in the scout view group [(8.84±3.79) mm ]was slightly greater than that in the body surface landmark-assisted positioning group[ (6.50±2.88) mm], and the difference between the two groups was statistically significant (t=4.20, P=0.032). Conclusions: The TMJ can be accurately positioned using stable surface landmarks such as the tragus and lateral canthus for reference.