Application of a self-made positioning device in laryngeal CT-MRI image fusion

Abstract

Objective To investigate the application of a self-made positioning device in CT-MRI image fusion in patients with laryngeal tumor, as well as the precision of image fusion and the changes in target volume delineation after fusion. Methods A total of 10 patients with laryngeal cancer were enrolled, and a self-made positioning device was used to collect CT and MRI images in a fixed position. These images were fused by mutual information combined with manual fusion. The precision of image fusion was assessed by the positional deviation of internal and external markers and degree of gross tumor volume (GTV) overlap (PCT-MRI) between CT and MRI images. GTV was contoured based on CT images (VCT), MRI images (VMRI), and fused images (VCT+ MRI). The overlapped volume of VCT and VMRI (VCT-MRI) was calculated, and the target volume was analyzed and compared. Results The positional deviations of three external markers in the three directions were 0.996±0.222 mm, 1.146±0.291 mm, and 1.368±0.298 mm (P=0.000), respectively, while those of the internal markers were 0.476±0.151 mm, 0.561±0.083 mm, and 0.724±0.125 mm (P=0.000), respectively. VCT, VMRI, VCT+ MRI, and VCT-MRI were 26.355±7.876 cm3, 33.556±7.407 cm3, 40.036±7.627 cm3, 19.875±8.588 cm3(P=0.000), respectively. PCT-MRI was 73.7%±9.8%. Conclusions The self-made positioning device can improve the consistency of position during the collection of CT and MRI images, and fused CT-MRI images can provide more information and improve the precision of target volume delineation. Key words: Tomography, X-ray computed; Magnetic resonance imaging; Image registration; Laryngeal neoplasms/radiotherapy; Self-made positioning device