Three-dimensional reconstruction based on computed tomography/magnetic resonance imaging multimodal image fusion for parotid gland tumor diagnosis and treatment

Abstract

This research evaluated a novel computed tomography (CT)/magnetic resonance imaging (MRI) fusion-based three-dimensional (3D) reconstruction method for parotid gland tumor diagnosis and treatment. In 11 patients with primary parotid tumors adjacent to the facial nerve, contrast-enhanced CT and 3D multi-echo in steady acquisition MRI images were automatically fused, segmented, and reconstructed to visualize tumors and critical anatomical structures. Preoperative tumor localization and tumor–nerve relationship assessments matched intraoperative findings in 90.9 % of cases. The method outperformed facial nerve line identification in deep lobe tumor assessment. Mean deviation distances for total, nerve, vascular, and tumor markers were 3.14 ± 1.80, 3.51 ± 1.85, 2.40 ± 1.49, and 2.47 ± 2.04 mm, respectively. Of five cases with facial nerve invasion on 3D reconstruction, three received nerve sacrifice and anastomosis, whereas two with nerve continuity preservation received postoperative radiotherapy. This technique enables precise spatial mapping of parotid tumors, aiding surgical planning and facial nerve preservation.