Performance Comparison of Machine Learning Algorithms Using EfficientNetB0 Feature Extraction on Dental Disease Classification - Read online free

Aug 3, 2026

Upload by Ophelia

Oral health problems such as dental caries, calculus, gingivitis, and ulcers remain widespread and require accurate early detection to prevent worsening outcomes. Traditional visual inspection and manual radiograph analysis introduce subjectivity, inconsistencies, and limited access, especially in underserved settings. The study builds an AI-based dental disease detection framework using intraoral images, extracting deep features with EfficientNetB0 and classifying them with eleven machine learning models, achieving up to 92.9% accuracy with SVM.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Article history

Abstract

Introduction

Methodology

Dataset

Image preprocessing and augmentation

Performance Comparison of Machine Learning Algorithms Using EfficientNetB0 Feature Extraction on Dental Disease Classification Mohammad Fa’iq Ruliff Mustafa 1*, Ajie Kusuma Wardhana 2* * Informatics, Faculty of Computer Science, Universitas Amikom Yogyakarta [mohammadfaiq03@students.amikom.ac.id](mohammadfaiq03@students.amikom.ac.id1)[1](mohammadfaiq03@students.amikom.ac.id1), [ajiekusuma@amikom.ac.id](ajiekusuma@amikom.ac.id2)[2](ajiekusuma@amikom.ac.id2) | Article history:
Received 2025-07-17 Revised 2025-08-01 Accepted 2025-08-10 | Oral health conditions such as dental caries, calculus, gingivitis, and ulcers are prevalent globally and require accurate early detection to prevent further complications. Traditional diagnostic methods such as visual inspection and manual radiograph analysis often rely on subjective judgment, leading to inconsistencies, delayed treatment, and limited accessibility, particularly in underserved areas. This study proposes an intelligent classification framework for dental disease detection based on intraoral images. Deep features were extracted using EfficientNetB0, followed by classification through eleven machine learning algorithms, including SVM, XGBoost, and K-Nearest Neighbors. Preprocessing steps included image augmentation, SMOTE for class balancing, and feature normalization. Among all models, SVM achieved the highest accuracy of 92,9%, while XGBoost and LightGBM followed closely at 91.3% . Using K-Fold Cross Validation, KNN algorithm has an increasing value with accuracy of 91,24% . This indicate the KNN algorithm able to tackle generalization problem towards the classification. The results demonstrate that features extracted using CNNs, when classified using machine learning algorithms, can provide a scalable and effective alternative to conventional diagnostic practices. Hence, Machine Learning algorithms provide a promising result towards dental disease classification.

This is an open access article under the CC–BY-SA license. | | --- | --- | | Keyword:
Convolutional Neural Networks, Dental Disease Classification,
EfficientNetB0, Machine Learning, SMOTE. | | Article Info ABSTRACT I. INTRODUCTION Global oral health faces significant challenges, affecting approximately 3.5 billion individuals worldwide who suffer from untreated dental conditions, constituting a major public health concern as highlighted by the World Health Organization [1] . Dental caries, one of the most widespread yet preventable oral diseases, continues to be inadequately addressed due to traditional detection methods and limited early diagnostic interventions [2] . It impacts up to 95.6% of adolescents in certain populations, particularly among those with lower socioeconomic status [3] . Similarly, dental calculus is prevalent, affecting over 73% of immunocompromised children, while gingivitis has been reported in nearly all adolescents in some studies [4] . Mouth ulcers, though less frequently discussed in population studies, are also common in vulnerable groups [5] . Before the integration of artificial intelligence (AI) into dental diagnostics, practitioners relied heavily on manual inspections and radiographic interpretation, which introduced variability and limited early detection [6] . These traditional approaches often lacked consistency and standardization, resulting in delayed diagnosis and treatment. The demand for more accurate and timely identification of dental conditions has spurred interest in AI technologies, which have shown potential in addressing these diagnostic limitations by providing consistent and automated solutions [7][8] . Conventional diagnostic methods are constrained by subjectivity and limited sensitivity, particularly in detecting subtle or early-stage conditions [9] . This has driven the development of computational approaches that leverage image-based data, where convolutional neural networks (CNNs) have proven especially effective [10] . CNNs are

FAQs

What problem does the study address in dental diagnostics?

It targets inaccurate and inconsistent early detection caused by traditional manual inspection and radiograph interpretation, which can delay diagnosis and treatment.

How are features extracted and which classifier family is evaluated?

The study extracts deep features from intraoral images using EfficientNetB0, then evaluates classification using eleven supervised machine learning algorithms.

What role do SMOTE and normalization play in the workflow?

SMOTE is used for class balancing, and feature normalization is applied to support stable training and fair model evaluation.

Aug 3, 2026

Upload by Ophelia