International Journal of Chemical and Biochemical Sciences (ISSN 2226-9614)[/vc_column_text][/vc_column][/vc_row]
VOLUME 28(22) (2025)
Artificial Intelligence in Periodontal Diagnosis: Current Evidence and Future Clinical Integration (Narrative Review)
Nariman Shaker1✻, Ola M. Ezzatt 2, Waleed Abbas2, Mahetab M. Abdalwahab 2
1✻Department of Oral Medicine, Periodontology and Diagnosis, Faculty of Oral and Dental Medicine, Future University, Cairo, Egypt
2Department of Oral Medicine, Oral Diagnosis and Periodontology, Faculty of Dentistry, Ain Shams University, Cairo, Egypt
Abstract
Periodontitis, a chronic inflammation that affects the supporting structures of teeth, can result in the loss of alveolar bone and potentially tooth loss. Traditional diagnostic techniques of periodontitis rely heavily on clinical and radiographic evaluations, which are subject to human variability and limited predictive capability. Advances in artificial intelligence (AI) have given rise to potential tools for achieving better and more efficient diagnosis in periodontology. The diagnosis of periodontitis is addressed in this mini-review using machine learning (ML) and deep learning algorithms implemented in artificial intelligence. Convolutional neural networks (CNNs) are the primary focus of research due to their demonstrated capacity for achieving high diagnostic sensitivity and specificity in detecting periodontal bone loss from radiographic inputs. Moreover, this review notes that “Hybrid models” are being developed to better reflect the predictive power of these models by integrating clinical risk factors such as smoking and diabetes with radiographic features. The integration of the 2017 classification system with AI models enables automated staging and diagnosis to meet modern clinical criteria resulting in accurate diagnostic outputs. Recent work shows that AI, particularly convolutional neural networks, can accurately interpret radiographs for periodontal diagnosis. However, challenges remain, including inconsistent datasets, absence of unified standards, and limited use in daily clinical practice. Future progress will depend on developing large multi-center databases, more transparent AI models, and systems linked to electronic dental records for real-time decision support.
Keywords: Artificial Intelligence, periodontal disease, machine learning, deep learning, diagnosis.
Full length article *Corresponding Author, e-mail: narimanhesham@dent.asu.edu.eg, Doi # https://doi.org/10.62877/2-IJCBS-25-28-22-2, Submitted: 14-11-2025; Accepted: 20-12-2025; Published: 22-12-2025