Autism is a condition that restricts brain development. Early diagnostic criteria are as follows: less responsive to social stimuli, inability to develop natural speaking skills, lack of communication skills, and limited and repetitive behaviors. Orthodontic treatment is a challenging process for doctors in autistic patients. The clinical and cephalometric examination of a 17-year and 7-month-old autistic patient revealed class I malocclusion, increased vertical dimensions, proclined upper and lower incisors, and inconsistent lip closure. The treatment plan involved four premolar extractions from each quadrant. Upper and lower 1. premolar extraction with fixed treatment caused reduction in vertical dimensions and significant improvement in lip closure and incisor inclination. Due to the increased metabolic activity in these patients, treatment was completed in 13 months. As a result, autistic patients can be successfully treated. In this process, it is important to include communication as a major part of treatment.
BACKGROUND: Pedodontists and general practitioners may need support in planning the early orthodontic treatment of patients with mixed dentition, especially in borderline cases. The use of machine learning algorithms is required to be able to consistently make treatment decisions for such cases. OBJECTIVE: This study aimed to use machine learning algorithms to facilitate the process of deciding whether to choose serial extraction or expansion of maxillary and mandibular dental arches for early treatment of borderline patients suffering from moderate to severe crowding. METHODS: The dataset of 116 patients who were previously treated by senior orthodontists and divided into two groups according to their treatment modalities were examined. Machine Learning algorithms including Multilayer Perceptron, Linear Logistic Regression, k-nearest Neighbors, Naïve Bayes, and Random Forest were trained on this dataset. Several metrics were used for the evaluation of accuracy, precision, recall, and kappa statistic. RESULTS: The most important 12 features were determined with the feature selection algorithm. While all algorithms achieved over 90% accuracy, Random Forest yielded 95% accuracy, with high reliability values (kappa = 0.90). CONCLUSION: The employment of machine learning methods for the treatment decision with or without extraction in the early treatment of patients in the mixed dentition can be particularly useful for pedodontists and general practitioners.
Objective: To define the dental and skeletal characteristics of Class III surgery patients with ideal final soft-tissue profiles, and to compare them with those of Class I subjects. Also, to show how soft-tissues respond to surgical jaw movements and contribute to the outcome. Methods: This short-term, retrospective study was conducted using pre-treatment (T0), pre-surgery (T1), and post-treatment (T2) records of 50 double-jaw Class III surgery patients who presented with ideal cephalometric characteristics in sagittal (Holdaway and soft-tissue convexity angles) and vertical dimensions (GoGn. SN angle and upper-to-lower face harmony) at the end of treatment, and 50 control subjects. Results: At T2, the horizontal distance between the vertical reference plane (a perpendicular plane to the horizontal reference plane that is angulated 7° clockwise to the SN plane) and hard-tissue A, B and Pog points, lower lip, soft-tissue B, and pogonion points were greater, Wits appraisal was more negative, U1.PP was higher, IMPA was lower, and soft-tissue chin (Pog-Pog’) was thicker in Group 1 when compared to Group 2 (p<0.05). Moreover, upper lip and subnasal (A-A’) thicknesses were decreased, and chin thickness (Pog-Pog’) was increased significantly (p<0.05). Conclusion: Dentoskeletal characteristics of an ideally-treated Class III surgery patient differed from a Class I subject concerning a protrusive maxilla and soft-tissue pogonion, and incisors that were not fully-decompensated. Soft-tissues hindered the actual surgical correction to 66% and 73% in the mid- and lower-faces, respectively.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.