ArticlePloS one2025
Accuracy of AI-based binary classification for detecting malocclusion in the mixed dentition stage.
Article in PloS one, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
5 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
backgroundMalocclusion is a common anomaly and is frequently observed in children and adults. Early detection and treatment of malocclusion is necessary to prevent and minimize complications. Therefore, developing a tool to check dentition at an early stage and motivate patients themselves to visit the dentist is required.
objectiveThis study aimed to examine the feasibility of building an AI model that can detect malocclusion in children during the mixed dentition stage.
methodsThis study was conducted as a feasibility study using cross-sectional data. Subjects were recruited from panelists registered with Macromill, Inc. (approximately 1.3 million registered in 2021). A total of 519 elementary school children (275 boys and 244 girls in Grades 3-6) were included in this study. Questionnaire data and tooth alignment images of the children were collected. The dataset was created, and AI-based binary classification models for malocclusion were developed using an automated machine learning platform (DataRobot) to construct three algorithms for determining malocclusion (deep bite, maxillary protrusion, and crowding). Using a test dataset, the model's performance was assessed through sensitivity, specificity, accuracy, precision, F1 score, receiver operating characteristic (ROC) curves, and area under the ROC curves (AUC).
resultsThree dental images were used for all model building, and questionnaire data used all four questions about oral habits (Q1: mouth open during the day, Q2: sleep with mouth open, Q3: have difficulty eating hard foods, Q4: prefer soft foods) for the deep bite classification model, Q1 and Q3 for the maxillary protrusion classification model, and Q1 and Q4 for the crowding classification model. The maxillary protrusion and crowding classification models showed moderate accuracy (AUC > 0.70), and the deep bite classification model showed high accuracy (AUC > 0.90). The permutation importance showed that dental image was the highest contributing factor in each model. Furthermore, while questionnaire data on oral habits were not an important factor in determining deep bite, these questionnaire data were an important factor in determining maxillary protrusion and crowding. Also, statistical analysis of the association between malocclusion and these oral habits revealed a significant association between maxillary protrusion or crowding and the presence or absence of oral habits.
conclusionFor the detection of malocclusion in mixed dentition, AI-based binary classification models are a promising approach as a screening tool.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.