거동 불편한 환자의 전신 CT 검사 시 DLIR 기법의 유용성 평가

Abstract

When whole-body computed tomography (WBCT) is performed after a single injection of contrast media, the arm's position is altered during each examination to minimize beam hardening artifacts caused by the arms. In this study, the image quality of neck CT images using the deep learning iterative reconstruction (DLIR) algorithm for patients whose arms were elevated during WBCT scans was compared with images reconstructed by FBP and IR algorithms in patients who were examined with their arms lowered. The Signal-to-Noise Ratio (SNR) and Contrast-to-Noise Ratio (CNR) of the reconstructed images were quantitatively compared, while the image quality and artifacts were qualitatively evaluated. As a result, TF-H was evaluated higher in both SNR and CNR in all organs compared to FBP. Compared to the ASIR-V at 40%, the CNR was high, and the SNR was similar. In the qualitative evaluation, it was confirmed that the quality of the image was higher when compared to the image examined with the arm lowered, surpassing that of the existing reconstruction method. Therefore, it is believed that enhancing the quality of the neck CT image through reconstruction using deep learning after the examination with the arm raised, without altering the posture, during the WBCT examination of patients with mobility impairments.