Deep learning synthesis of cone-beam computed tomography from zero echo time magnetic resonance imaging

Abstract

Cone-beam computed tomography (CBCT) produces high-resolution of hard tissue even in small voxel size, but the process is associated with radiation exposure and poor soft tissue imaging. Thus, we synthesized a CBCT image from the magnetic resonance imaging (MRI), using deep learning and to assess its clinical accuracy. We collected patients who underwent both CBCT and MRI simultaneously in our institution (Seoul). MRI data were registered with CBCT data, and both data were prepared into 512 slices of axial, sagittal, and coronal sections. A deep learning-based synthesis model was trained and the output data were evaluated by comparing the original and synthetic CBCT (syCBCT). According to expert evaluation, syCBCT images showed better performance in terms of artifacts and noise criteria but had poor resolution compared to the original CBCT images. In syCBCT, hard tissue showed better clarity with significantly different MAE and SSIM. This study result would be a basis for replacing CBCT with non-radiation imaging that would be helpful for patients planning to undergo both MRI and CBCT.

Multi-planar 2.5D U-Net for image quality enhancement of dental cone-beam CT

“…For the 18 original CBCT (oCBCT) and corresponding predicted (or corrected) CBCT (pCBCT) images, the evaluation was conducted using a PACS viewer (Zetta, Tae-young, South Korea) in random order. Two radiologists made an evaluation using a modified version of the clinical CBCT image evaluation chart provided by the Korean Academy of Oral and Maxillofacial Radiology ( Table 2 ) [ 13 ]. The evaluation procedure was conducted individually in blind condition.…”

“…Deep learning (DL) based approaches for CBCT image improvement have recently emerged as a possible viable solution in clinics. While prior studies have demonstrated the possibility of improving CBCT through the use of DL by reducing artifacts and noise and by standardizing pixel intensity [ 6 – 12 ], DL inference is still limited by obscured anatomic fine details and blurred edges in the image [ 7 , 13 ]. These factors may limit the diagnostic utility of CBCT for intricate features such as the teeth, alveolar bone pattern, sinuses, or the temporomandibular joint (TMJ) complex.…”

Clinical application of high-resolution spiral CT scanning in the diagnosis of auriculotemporal and ossicle

“…The study’s central tenet is to improve diagnostic accuracy and efficiency by combining High-Resolution Spiral Computed Tomography (HRSCT) [ 16 ] scanning with Deep Learning Techniques (DLT) [ 17 ]. Incorporating the CNN-UNet deep learning method, which has been fine-tuned to perform exceptionally well in catching the finest distinctions inside medical images, is central to this groundbreaking method.…”