From pixels to precision: a narrative review of AI-driven 3D morphological analysis and digital twinning in alveolar cleft management

Alveolar cleft is a congenital craniofacial anomaly of common occurrence and is frequently seen in cleft lip and palate patients. This condition affects the patient's chewing, speech, and psychological and social life. This review aims to offer a broad overview of the role of artificial intelligence (AI) throughout the entire management of an alveolar cleft from diagnosis to treatment and to life after surgery, in terms of quality of life.

This is a narrative review, and the information was obtained from the literature and clinical guidelines. From the previous publications, we reviewed the advancements in the use of AI for the diagnosis, treatment, and management of alveolar clefts and other fields.

In the field of alveolar ridge defects, the applications of AI have been mainly used in combination with cone-beam computed tomography (CBCT). It accurately identifies the extent of the bone defect and calculates the volume of the bone defect from imaging. Moreover, AI can support personalized surgical planning and predict bone resorption and maxillofacial growth patterns. AI and CBCT are shifting the subjective, "experience-driven" qualitative assessment of alveolar clefts to an objective, "data-driven" evaluation that is multidimensional.

AI holds significant potential for applications in 3D image analysis, quantitative evaluation, and surgical planning of alveolar clefts, but it is currently hindered by several challenges, such as limited generalizability of models, lack of interpretability, and privacy concerns. Solutions to these problems include advancing towards multicenter standardized databases, federated learning, and model efficiency. The future of alveolar cleft management is expected to be improved with the use of generative AI and digital twins throughout their lifespan, in a personalized way.