Construction and evaluation of a risk prediction model for dental cracks based on imaging and clinical features
CHEN Nianjie
OU Yuejian
Abstract：Objective To construct and evaluate a risk prediction model for cracked teeth based on CBCT images and clinical features.Methods A total of 130 molar positions from 130 patients treated in Huzhou central hospital between January 2024 and June 2025 were selected,including 80 cracked teeth in the experimental group and 50 normal teeth in the control group.Electronic medical records and CBCT measurement data were collected.Variables were screened through univariate analysis,ROC curves,and least absolute shrinkage and selection operator regression to construct and validate a nomogram model.Results Univariate analysis showed that gender,preference for hard food,unilateral chewing,bruxism,mesial buccolingual cusp slope angle,distal buccolingual cusp slope angle,tooth length,and root length were associated with cracked teeth(P＜0.05).A model constructed with 8 variables screened by LASSO regression showed an AUC of 0.91(95%CI:0.85-0.97)and accuracy of 0.87 in the training set,and an AUC of 0.922(95%CI:0.84-1.00)and accuracy of 0.85 in the validation set,with a mean calibration curve error of 0.042.Conclusion The model,combining CBCT parameters and clinical features,can effectively identify high-risk populations for cracked teeth and support clinical diagnosis and treatment.
Keywords：cracked toothCBCTrisk prediction modelnomogram
Publication Date：2026-01-28
Online Publishing Date：2026-03-11(First online date of this platform, not the publication date of the document)
Pages：6( 27-32 )