An automatic method for skeletal patterns classification using craniomaxillary variables on a Colombian population

Evaluation of natural mandibular shape asymmetry: an approach by using elliptical Fourier analysis

“…This is consistent because the ANB angle is a spectrum of continuous measures that discriminates the general facial form with one degree of difference between its limits. Unlike previous studies on the same research line, 14 in this case the differentiation of Class II and Class I was evident except in Class I males' right side. This means that the exploration in the shape of the classes or even different groupings other than Steiner's classifications is necessary because it could determine a practical differentiation which would also help with the processes of prediction of the mandibular morphology.…”

“…The selection criteria included a complete permanent dentition with third molars in similar position. Patients with previous orthognathic or orthodontic treatment, orthopedic and/or aesthetic surgery, severe parafunctional habits and congenital or acquired malformations were excluded; this information is available in the database of the Master of Dentistry of the National University of Colombia, described in Niño-Sandoval et al 14 All participants in this study signed an informed consent considering the principles stated in the Declaration of Helsinki "Ethical Principles for Medical Research Involving Human Subjects". Also, ethics considerations were based according to Law 84 of the 1989 Colombian constitution in its resolution 008,430 which establishes the scientific, technical, and administrative norms for the investigation in health.…”

Applications of artificial intelligence and machine learning in orthodontics

“…Simple and easy to understand [71] Order of training has no effect on training [71] It is based on statistical modeling [71] Requires small amount of data for training [ 72] Fast and can deal with discrete and continuous attributes [72] Robust to outliers [73] Accuracy is affected by redundant attributes and class frequency [71] Normal distribution is assumed for numeric attributes [71] Attributes are assumed to be conditionally independent [71] Neural network Used for classification and regression Applied in dentistry and medicine for diagnosis [37] Boolean functions (AND, OR, and NOT) can be used with neural networks Can handle noisy inputs and allows changing input features during data collection [74] Successful with complex non-linear relationships between predicted variable and input data [74] Overfitting is common especially with too many variables [75] Have limited ability to identify causal relationship [74] Require more computational resources [74] Support vector machine Used for classification and regression Applied in dentistry for classification of skeletal patterns [56] Resistant to overfitting [10] Can model nonlinear functions [10] Can be used with non-linear relationships between predicted variable and input data…”

“…Another study using support vector machines to classify normal or abnormal skeletal patterns based on craniofacial measures was correct only 74.5% of the time. [56] Classification of Class III growth patterns has also been performed. Based on longitudinal data of untreated Class III subjects, who were classified as either good or bad growers based on the changes in their sagittal relationships, a classification tree had a significantly lower rate of misclassification (12.0%) than discriminant analysis (40.7%), both of which were based based on the same 11 cephalometric variables.…”

Artificial intelligence and machine learning in orthodontics

“…12 of the studies included in the systematic review consist of those on machine learning and super vector analysis to automatically determine cephalometric points on threedimensional or two-dimensional radiography images [4][5][6][8][9][10][11][12][13][14][15]. Artificial intelligence and super vector machine were used in 3 studies that worked on facial attractiveness and perception [16][17][18].…”