The challenges of interpreting ANOVA by dermatologists

CART-ANOVA-Based Transfer Learning Approach for Seven Distinct Tumor Classification Schemes with Generalization Capability

Background/Objectives: Deep transfer learning, leveraging convolutional neural networks (CNNs), has become a pivotal tool for brain tumor detection. However, key challenges include optimizing hyperparameter selection and enhancing the generalization capabilities of models. This study introduces a novel CART-ANOVA (Cartesian-ANOVA) hyperparameter tuning framework, which differs from traditional optimization methods by systematically integrating statistical significance testing (ANOVA) with the Cartesian product of hyperparameter values. This approach ensures robust and precise parameter tuning by evaluating the interaction effects between hyperparameters, such as batch size and learning rate, rather than relying solely on grid or random search. Additionally, it implements seven distinct classification schemes for brain tumors, aimed at improving diagnostic accuracy and robustness. Methods: The proposed framework employs a ResNet18-based knowledge transfer learning (KTL) model trained on a primary dataset, with 20% allocated for testing. Hyperparameters were optimized using CART-ANOVA analysis, and statistical validation ensured robust parameter selection. The model's generalization and robustness were evaluated on an independent second dataset. Performance metrics, including precision, accuracy, sensitivity, and F1 score, were compared against other pre-trained CNN models. Results: The framework achieved exceptional testing accuracy of 99.65% for four-class classification and 98.05% for seven-class classification on the source 1 dataset. It also maintained high generalization capabilities, achieving accuracies of 98.77% and 96.77% on the source 2 datasets for the same tasks. The incorporation of seven distinct classification schemes further enhanced variability and diagnostic capability, surpassing the performance of other pre-trained models. Conclusions: The CART-ANOVA hyperparameter tuning framework, combined with a ResNet18-based KTL approach, significantly improves brain tumor classification accuracy, robustness, and generalization. These advancements demonstrate strong potential for enhancing diagnostic precision and informing effective treatment strategies, contributing to advancements in medical imaging and AI-driven healthcare solutions.

Effect of auxiliary geometric devices on the accuracy of intraoral scans in full-arch implant-supported rehabilitations: An in vitro study

OBJECTIVES

The aim of the present in vitro study was to evaluate the effect of a novel auxiliary geometric device (AGD) on the accuracy of full-arch scans captured with 3 different intraoral scanners (IOS).

METHODS

An edentulous maxillary model with four internal connection implant replicas was scanned using 3 different IOS: iTero Element 5D (ITERO) (Align Technology, Tempe, AZ, USA), Trios 4 (TRIOS) (3Shape A/S, Copenhagen, Denmark), and Carestream 3700 (CS) (Carestream Dental, Atlanta, USA). Thirty-six scans were taken with each IOS, 18 with the AGD in place, and 18 without the AGD. A digital master model was created using an industrial optical scanner (ATOS compact Scan 5M, GOM GmbH, Braunschweig, Germany). The master and IOS models were aligned using the scan bodies as a reference area. A surface comparison was performed, and deviation labels were exported for each scan body to evaluate the linear and angular deviation. Total body, platform and angular deviations were measured.

RESULTS

The use of AGD resulted in a statistically significant increase of angular deviation: 0.87° (SD=0.21) in the AGD group versus 0.64° (SD=0.46) in the no AGD group (p-value=0.005). The difference between the AGD and no AGD groups was not statistically significant for total body and platform deviation values (p-value=0.051 and 0.302 respectively). Using AGD, ITERO showed a statistically significant increase in angular deviation (mean difference=-0.46 µm, p-value=0.002) and a decrease in mean platform deviation (mean difference=63.19 µm, p-value<0.001). No statistically significant differences were found for the other IOS.

CONCLUSIONS

The use of AGD did not add benefit on CS and TRIOS. On ITERO, there was an improvement in platform deviation, that was outweighed by the worsening of the angular deviation.

CLINICAL SIGNIFICANCE

In vitro data suggest that intraoral scans can be successfully used in full-arch cases. The use of AGD has no additional benefit on CS and TRIOS. On ITERO there was an improvement in platform deviation that was outweighed by the worsening of the angular deviation. Translational application to clinical practice deserves further investigation, taking into account patient-related and anatomical variables.

Testing Equality of Multiple Population Means under Contaminated Normal Model Using the Density Power Divergence

This paper considers the problem of comparing several means under the one-way Analysis of Variance (ANOVA) setup. In ANOVA, outliers and heavy-tailed error distribution can seriously hinder the treatment effect, leading to false positive or false negative test results. We propose a robust test of ANOVA using an M-estimator based on the density power divergence. Compared with the existing robust and non-robust approaches, the proposed testing procedure is less affected by data contamination and improves the analysis. The asymptotic properties of the proposed test are derived under some regularity conditions. The finite-sample performance of the proposed test is examined via a series of Monte-Carlo experiments and two empirical data examples-bone marrow transplant dataset and glucose level dataset. The results produced by the proposed testing procedure are favorably compared with the classical ANOVA and robust tests based on Huber's M-estimator and Tukey's MM-estimator.