IADR Abstract Archives
Artificial Intelligence in Oral Diseases: Clinical-Histopathological Correlation ( Every day Practice )
Objectives:
Methods:
10 cases for each disease were enrolled in the study. The study sample included clinical images, patient symptoms, radiographs & histopathology texts for the oral diseases of interest in the current study (premalignant lesions, oral cancer, salivary gland diseases, immune-mediated oral mucosal lesions)
Results: The diagnostic performance of CAS was comparable to experienced oral pathologist and significantly superior to inexperienced clinicians. CAS provided faster differential diagnostic list and reliable recommendations for further investigatory tests in challenging cases.
Conclusions:
Developing a software with all needed input data downloaded in the computer device to act as machine learning program to diagnose variety of oral diseases with accuracy and proposal of treatment plan especially for life threatening conditions. So our research question was: Can we develop a Computer-Aided Software (CAS) for accurate diagnosis for oral diseases based on clinical and histopathological data inputs?.
Methods:
10 cases for each disease were enrolled in the study. The study sample included clinical images, patient symptoms, radiographs & histopathology texts for the oral diseases of interest in the current study (premalignant lesions, oral cancer, salivary gland diseases, immune-mediated oral mucosal lesions)
Results: The diagnostic performance of CAS was comparable to experienced oral pathologist and significantly superior to inexperienced clinicians. CAS provided faster differential diagnostic list and reliable recommendations for further investigatory tests in challenging cases.
Conclusions:
CAS has a potential to be used as diagnostic guidance tool for clinicians as well as could be used for accurate confirmation of diagnosis, test and treatment plan.