Ein innovatives Machine-Learning-Modell für die Entscheidungsfindung bei Klasse-III-Operationen

Abstract

Purpose The primary purpose of this study was to develop a new machine learning model for the surgery/non-surgery decision in class III patients and evaluate the validity and reliability of this model. Methods The sample consisted of 196 skeletal class III patients. All the cases were allocated randomly, 136 to the training set and the remaining 60 to the test set. Using the test set, the success rate of the artificial neural network model was estimated, along with a 95% confidence interval. To predict surgical cases, we trained a binary classifier using two different methods: random forest (RF) and logistic regression (LR). Results Both the RF and the LR model showed high separability when classifying each patient for surgical or non-surgical treatment. RF achieved an area under the curve (AUC) of 0.9395 on the test set. 95% confidence intervals were computed by bootstrap sampling as lower bound = 0.7908 and higher bound = 0.9799. On the other hand, LR achieved an AUC of 0.937 on the test set. 95% confidence intervals were computed by bootstrap sampling as lower bound = 0.8467 and higher bound = 0.9812. Conclusions RF and LR machine learning models can be used to generate accurate and reliable algorithms to successfully classify patients up to 90%. The features selected by the algorithms coincide with the clinical features that we as clinicians weigh heavily when determining a treatment plan. This study further supports that overjet, Wits appraisal, lower incisor angulation, and Holdaway H angle can be used as strong predictors in assessing a patient’s surgical needs.

Computation of the distribution of model accuracy statistics in machine learning: Comparison between analytically derived distributions and simulation‐based methods

“…16,17,[26][27][28][29] As a result, the vast majority of packages focus upon bootstrapping and thus so too do the vast majority of studies. 1,8,[30][31][32] Due to the increased computational power, these nonparametric methods can be completed efficiently and are F I G U R E 1 Balanced accuracy, Accuracy, F1, Sensitivity, Specificity, Positive predictive value, Negative predictive value, and area under the receiver operator characteristic curve (AUROC) for the XGboost model following bootstrap simulation.…”

“…Bootstrapping can generate a distribution based on data without any knowledge of the distribution and without violating any assumptions that are required to utilize a distribution for inference 16,17,26‐29 . As a result, the vast majority of packages focus upon bootstrapping and thus so too do the vast majority of studies 1,8,30‐32 . Due to the increased computational power, these nonparametric methods can be completed efficiently and are especially useful for distributions that cannot be quantified analytically 1,3,5,22,33‐35 .…”

Call for algorithmic fairness to mitigate amplification of racial biases in artificial intelligence models used in orthodontics and craniofacial health

“…There was also wide variation on how race was included or not included in the ML models. [33][34][35][36][37] Racial categories have been historically difficult to define and the inherent diversity within racial groups has not been well delineated and accounted for in prior research. 22 The effects of upstream factors such as social, behavioural, economical, educational and structural determinants of health that have a significant confounding effect on race have not been systematically studied and parsed out.…”

“…24,29,30 Orthodontic datasets are hampered by considerable amounts of missing data, especially on race and socio-economic factors. [34][35][36][37] Researchers should consider using imputation strategies to account for missing information. Certain mix of variables such as insurance type, health status and socioeconomic status might be predictive of race and ethnicity and a two-staged model can be developed wherein a propensity scoring approach is developed to obtain a continuous or categorical score for missing race in the first stage.…”

“…For example, the five most recent publications originating from the United States on “Machine Learning in Orthodontics” have reported race in an inconsistent manner or have not reported it at all (Table 1). 33–37 In two of the five studies, information on race was missing in over 50% of subjects and information about ethnicity was missing for over 70% of subjects. There was also wide variation on how race was included or not included in the ML models 33–37 .…”