Thesis
Application of artificial intelligence in airway evaluation of cone beam computed tomography scans
University of Iowa
Master of Science (MS), University of Iowa
Spring 2025
DOI: 10.25820/etd.008029
Abstract
Introduction: Multiple studies have supported the idea that new AI-based technologies can provide airway analysis comparable to that of human practitioners. Tested products include ‘Mimics’ (Materialise N.V., Leuven, Belgium) and ‘Diagnocat’ (Diagnocat, USA). According to the most comprehensive systematic review performed of these studies, “in order to have strong evidence on how accurately CBCT airway analysis is performed with the use of artificial intelligence, more randomized prospective studies should be performed. Those studies should limit all kind of bias to provide high quality evidence” (Tsolakis et al., 2022). Current literature has focused on testing these new technologies in very rigid conditions with the largest of these studies excluding datasets with artifact completely (Orhan et al., 2022). This is important because other reviews of these technologies have found that “overfitting is a common issue in the whole field of AI. This means the model performs excessively well in the training datasets but shows unsatisfactory performance in the testing dataset. Factors like data insufficiency, low data heterogeneity and excessive variables could all lead to overfitting” (Liu et al., 2023).Materials and Methods: 100 anonymized cone beam computed tomography datasets obtained from the records of the University of Iowa were oriented and analyzed by two calibrated oral and maxillofacial radiology residents using InVivo software (InVivo6, Anatomage, San Jose, CA). Measurements made were total airway volume and minimum surface area of the airway. These measurements were then compared to those produced by uploading the same datasets to Diagnocat AI software (Diagnocat, San Francisco, CA).
Results: The results indicate that all comparisons showed strong to excellent reliability (ICC > 0.75), suggesting high agreement between raters. Specifically: the interhuman rater comparisons demonstrated excellent reliability for both total airway volume (ICC = 0.974, 95% CI: 0.96 – 0.983) and minimum surface area (ICC = 0.968, 95% CI: 0.952 – 0.979), indicating very consistent measurements between these two raters.
Conclusion: Diagnocat (Diagnocat, San Francisco, CA) artificial intelligence software is capable of analyzing patients’ oropharyngeal airway volume and minimum cross-sectional area in cone beam computed tomography datasets with minimal to moderate motion artifact at a level comparable to qualified human practitioners.
Details
- Title: Subtitle
- Application of artificial intelligence in airway evaluation of cone beam computed tomography scans
- Creators
- Steven Dorris
- Contributors
- Trishul Allareddy (Advisor)Sindhura Anamali (Committee Member)Juan P. Castro (Committee Member)Shareef Dabdoub (Committee Member)
- Resource Type
- Thesis
- Degree Awarded
- Master of Science (MS), University of Iowa
- Degree in
- Oral Science
- Date degree season
- Spring 2025
- DOI
- 10.25820/etd.008029
- Publisher
- University of Iowa
- Number of pages
- vii, 23 pages
- Copyright
- Copyright 2025 Steven Dorris
- Language
- English
- Date submitted
- 04/29/2025
- Description illustrations
- color illustrations
- Description bibliographic
- Includes bibliographical references (page 21-23).
- Public Abstract (ETD)
The use of artificial intelligence as a diagnostic tool in dentistry is increasing across multiple sectors. One of these applications has been in the analysis of cone beam computed tomography (CBCT) datasets which have been increasingly used in the field of airway analysis. While previous studies have shown promising results, the question of whether diagnostic tools using artificial intelligence to analyze airways in clinical conditions can produce results comparable to qualified human practitioners remains underexplored. Previous studies have examined these tools in the absence of radiographic artifact. One hundred CBCT datasets with minimal to moderate motion artifact were obtained and analyzed to determine each patient’s total oropharyngeal airway volume and minimum cross-sectional area of their airway. Airway analysis was performed by two residents in the oral and maxillofacial radiology program at the University of Iowa college of dentistry to determine interrater reliability. The same one hundred datasets were analyzed by Diagnocat (Diagnocat, San Francisco, CA) artificial intelligence software to produce the same measurements. The human and AI measurements were then compared. It was determined that the Diagnocat software produced comparable results to that of the human practitioners in the presence of minimal to moderate motion artifact.
- Academic Unit
- Oral Pathology, Radiology and Medicine
- Record Identifier
- 9984831229202771
Metrics
1 File views/ downloads
9 Record Views