Validation of 2D lateral cephalometric analysis using artificial intelligence-processed low-dose cone beam computed tomography
Eun-Ji Chung,
Byoung-Eun Yang,
Sam-Hee Kang,
Young-Hee Kim,
Ji-Yeon Na,
Sang-Yoon Park,
Sung-Woon On,
Soo-Hwan Byun
Affiliations
Eun-Ji Chung
Department of Conservative Dentistry, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea
Byoung-Eun Yang
Department of Oral and Maxillofacial Surgery, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea; Graduate School of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Institute of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Dental AI-Robotics Center, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea
Sam-Hee Kang
Department of Conservative Dentistry, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea
Young-Hee Kim
Institute of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Department of Oral and Maxillofacial Radiology, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea
Ji-Yeon Na
Institute of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Dental AI-Robotics Center, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea; Department of Oral and Maxillofacial Radiology, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea
Sang-Yoon Park
Department of Oral and Maxillofacial Surgery, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea; Graduate School of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Institute of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Dental AI-Robotics Center, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea
Sung-Woon On
Graduate School of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Institute of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Department of Oral and Maxillofacial Surgery, Hallym University Dongtan Sacred Heart Hospital, Hwaseong, 18450, Republic of Korea
Soo-Hwan Byun
Department of Oral and Maxillofacial Surgery, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea; Graduate School of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Institute of Clinical Dentistry, Hallym University, Chuncheon, 24252, Republic of Korea; Dental AI-Robotics Center, Hallym University Sacred Heart Hospital, Anyang, 14066, Republic of Korea; Corresponding author. Dentistry, Hallym University Sacred Heart Hospital, Gwanpyung-ro 170, Anyang, 14066, Republic of Korea.
Objectives: Traditional cephalometric radiographs depict a three-dimensional structure in a two-dimensional plane; therefore, errors may occur during a quantitative assessment. Cone beam computed tomography, on the other hand, minimizes image distortion, allowing essential areas to be observed without overlap. Artificial intelligence can be used to enhance low-dose cone beam computed tomography images. This study aimed to clinically validate the use of artificial intelligence-processed low-dose cone beam computed tomography for generating two-dimensional lateral cephalometric radiographs by comparing these artificial intelligence-enhanced radiographs with traditional two-dimensional lateral cephalograms and those derived from standard cone beam computed tomography. Methods: Sixteen participants who had previously undergone both cone beam computed tomography and plain radiography were selected. Group I included standard lateral cephalometric radiographs. Group II included cone beam computed tomography-produced lateral cephalometric radiographs, and Group III included artificial intelligence-processed low-dose cone beam computed tomography-produced lateral cephalometric radiographs. Lateral cephalometric radiographs of the three groups were analyzed using an artificial intelligence-based cephalometric analysis platform. Results: A total of six angles and five lengths were measured for dentofacial diagnosis. There were no significant differences in measurements except for nasion-menton among the three groups. Conclusions: Low-dose cone beam computed tomography could be an efficient method for cephalometric analyses in dentofacial treatment. Artificial intelligence-processed low-dose cone beam computed tomography imaging procedures have the potential in a wide range of dental applications. Further research is required to develop artificial intelligence technologies capable of producing acceptable and effective outcomes in various clinical situations. Clinical significance: Replacing standard cephalograms with cone beam computed tomography (CBCT) to evaluate the craniofacial relationship has the potential to significantly enhance the diagnosis and treatment of selected patients. The effectiveness of low-dose (LD)-CBCT was assessed in this study. The results indicated that lateral cephalograms reconstructed using LD-CBCT were comparable to standard lateral cephalograms.