Search

BackgroundTimely and accurate outcome prediction plays a vital role in guiding clinical decisions on acute ischemic stroke. Early condition deterioration and severity after the acute stage are determinants for long-term outcomes. Therefore, predicting early outcomes is crucial in acute stroke management. However, interpreting the predictions and transforming them into clinically explainable concepts are as important as the predictions themselves. ObjectiveThis work focused on machine learning model analysis in predicting the early outcomes of ischemic stroke and used model explanation skills in interpreting the results. MethodsAcute ischemic stroke patients registered on the Stroke Registry of the Chang Gung Healthcare System (SRICHS) in 2009 were enrolled for machine learning predictions of the two primary outcomes: modified Rankin Scale (mRS) at hospital discharge and in-hospital deterioration. We compared 4 machine learning models, namely support vector machine (SVM), random forest (RF), light gradient boosting machine (LGBM), and deep neural network (DNN), with the area under the curve (AUC) of the receiver operating characteristic curve. Further, 3 resampling methods, random under sampling (RUS), random over sampling, and the synthetic minority over-sampling technique, dealt with the imbalanced data. The models were explained based on the ranking of feature importance and the SHapley Additive exPlanations (SHAP). ResultsRF performed well in both outcomes (discharge mRS: mean AUC 0.829, SD 0.018; in-hospital deterioration: mean AUC 0.710, SD 0.023 on original data and 0.728, SD 0.036 on resampled data with RUS for imbalanced data). In addition, DNN outperformed other models in predicting in-hospital deterioration on data without resampling (mean AUC 0.732, SD 0.064). In general, resampling contributed to the limited improvement of model performance in predicting in-hospital deterioration using imbalanced data. The features obtained from the National Institutes of Health Stroke Scale (NIHSS), white blood cell differential counts, and age were the key features for predicting discharge mRS. In contrast, the NIHSS total score, initial blood pressure, having diabetes mellitus, and features from hemograms were the most important features in predicting in-hospital deterioration. The SHAP summary described the impacts of the feature values on each outcome prediction. ConclusionsMachine learning models are feasible in predicting early stroke outcomes. An enriched feature bank could improve model performance. Initial neurological levels and age determined the activity independence at hospital discharge. In addition, physiological and laboratory surveillance aided in predicting in-hospital deterioration. The use of the SHAP explanatory method successfully transformed machine learning predictions into clinically meaningful results.

Background/purpose: Ultrasonic instruments are commonly used in dentistry because of their safety, acceptance by patients, ease of viewing the surgical area, and highly precise cutting. However, they do not efficiently remove enamel. The aim of this study is to optimize the surface coating parameters of an ultrasonic diamond grinding tip for enamel removal. Materials and methods: The experiments were conducted using a triple-axle precision moving platform. The ultrasonic handpiece was mounted on the platform column, and powered by an ultrasonic device with a voice coil motor providing a uniform down force of 2 N to grind adult molar specimens. The commercially available diamond tips UL3A (Cur scaler, Bonart, Taiwan) and EX1 (Newtron, Acteon, France) were used in this study. The Taguchi method and analysis of variance were used to determine the optimal surface parameters (mesh size, crystal shape, protrusion, and crystal density) of dental diamond tips. By using the optimal surface characteristics, new tips N1 and N2, having the same geometrical shape and dimension of UL3A and EX1, respectively, were fabricated and their grinding performance was tested. Results: The optimal enamel removal parameters of the grinding tip are diamonds of 100/120 mesh size, shape of well-formed crystal facet, 75 μm protrusion, and density of 45 crystals per square millimeter surface area. The material removal rate (MRR) of enamel for N1 (UL3A with new characteristics) is 0.064 mm3/s, 3.4 times that of the original UL3A tips. For N2 (EX1 with new characteristics) it is 0.126 mm3/s, which is 3.2 times that of the original EX1 tips. Conclusion: Protrusion, shape, and density of diamonds of an ultrasonic dental tip are significantly related to the MRR of enamel, and the optimal combination of these parameters is obtained. Knowledge of the importance of these variables will help in more effective use of the ultrasonic technology in dentistry.