Artificial Intelligence Based Algorithm For Assessment of Skeletal Age and Detection of Cervical Vertebral Anomalies in Lateral Cephalogram in Children with Cleft Lip and Palate

In this study, 1000 lateral cephalograms (head x-rays) will be collected to develop a machine learned program which will assess the upper spine and detect the maturity of the bone utilizing the Cervical Vertebral Maturation (CVM) method. The benefit of choosing lateral cephalograms is the prevention of additional radiation exposure to the patients as it is a routine orthodontic diagnostic radiograph. additionally, the program can detect any abnormalities in the morphology of the upper spine deviating from the normal. 

The correction of disharmonies in the sagittal, transverse, and vertical planes, orthodontic therapy and interventions are often timed to occur before or during the pubertal growth spurt or peak of growth velocity. Therefore, the thorough understanding of the timing of the craniofacial growth and development and skeletal age is critical in the correct diagnosis, treatment planning, outcome and overall stability of orthodontic treatment, secondary alveolar bone grafting, and orthognathic surgery in CL/P children.

Assessment of skeletal age using the CVM method is usually performed by orthodontists and other dental specialists and is a tedious procedure with variations in reproducibility and is subjective, leading to inaccuracies in results.

Therefore, the efficiency of outcome measurements as well as the predictability and reliability of the results will be considerably improved by the use of artificial intelligence-based models for automatic skeletal age detection in lateral cephalograms. This may be helpful where there is a lack of medical resources because automated technologies can speed up medical care and provide accurate and prompt diagnosis.