Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

Computer vision adjunctive technology for oral lesion diagnoses has been developed to detect and identify Oral Potentially Malignant Disorders (OPMDs) and non-OPMDs. The early detection of OPMDs can reduce the risk of oral cancer development, improving the survival rate of the patients. This study aims to evaluate the computer vision technique in the white oral lesion domain within the scope of photographic images. Deep learning techniques for the classification of Convolution Neural Networks (CNNs) and transformer neural networks, and one-stage models of YOLOv7 and YOLOv8 were utilized to classify and detect five classes of OPMDs and non-OPMDs oral white lesions including oral leukoplakia, oral lichen planus, pseudomembranous candidiasis, oral ulcers covered with pseudomembrane and other white benign oral lesions. From the evaluation results of classification, the IFormerBase model achieves overperformance compared to CNN models with accuracy, precision, and F1 score of more than 80% on the test set. The best model for object detection is YOLOv7 with 84.5% mean Average Precision (mAP) at Intersection over Union (IoU) threshold of 0.3 and 74.5% at IoU of 0.5 on the test set. Object detection results reveal promising automatic oral lesion identification, which can be further developed to enhance the lesion screening system.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12371035PMC
http://dx.doi.org/10.1038/s41598-025-14450-wDOI Listing

Publication Analysis

Top Keywords

object detection
12
oral lesion
12
oral
11
deep learning
8
oral white
8
white lesions
8
photographic images
8
computer vision
8
opmds non-opmds
8
neural networks
8

Similar Publications

The aim of the study was to validate a new method for semiautomatic subtraction of [Tc]Tc-sestamibi and [Tc]NaTcO SPECT 3-dimensional datasets using principal component analysis (PCA) against the results of parathyroid surgery and to compare its performance with an interactive method for visual comparison of images. We also sought to identify factors that affect the accuracy of lesion detection using the two methods. Scintigraphic data from [Tc]Tc-sestamibi and [Tc]NaTcO SPECT were analyzed using semiautomatic subtraction of the 2 registered datasets based on PCA applied to the region of interest including the thyroid and an interactive method for visual comparison of the 2 image datasets.

View Article and Find Full Text PDF

ALYREF stabilizes CREPT mRNA to accelerate the development of nasopharyngeal carcinoma through dependence on m5C modification.

Exp Cell Res

September 2025

The Department of Hematology, The First Affiliated Hospital of Hainan Medical University, No.31 Longhua Road, Haikou City, Hainan Province, 570000, P.R. China. Electronic address:

Background: Nasopharyngeal carcinoma (NPC) is a kind of tumor disease with high malignant degree. CREPT expression was elevated abnormally in multi-cancers. However, the role and regulatory mechanism of CREPT in NPC remains unknown.

View Article and Find Full Text PDF

Passive Wake Differentiation by Seal Vibrissae in Response to Independently Oscillating Upstream Objects.

Bioinspir Biomim

September 2025

Mechanical Engineering, University of Massachusetts Dartmouth, 285 Old Westport Road, Dartmouth, Massachusetts, 02747-2300, UNITED STATES.

Harbor seals possess a remarkable ability to detect hydrodynamic footprints left by moving objects, even long after the objects have passed, through interactions between wake flows and their uniquely shaped whiskers. While the flow-induced vibration (FIV) of harbor seal whisker models has been extensively studied, their response to unsteady wakes generated by upstream moving bodies remains poorly understood. This study investigates the wake-induced vibration (WIV) of a flexibly mounted harbor seal-inspired whisker positioned downstream of a forced-oscillating circular cylinder, simulating the hydrodynamic footprint of a moving object.

View Article and Find Full Text PDF

Detection and pharmacokinetics of licochalcone A in brains of neuroinflammatory mouse model.

Naunyn Schmiedebergs Arch Pharmacol

September 2025

Pharmacology and Toxicology Department, Faculty of Pharmacy and Biotechnology, German University in Cairo, Gamal Abdel Nasser, 11835, New Cairo, Egypt.

Licochalcone A (LCA), a natural flavonoid with potent anti-inflammatory properties, has shown promise as a neuroprotective agent. However, its ability to cross the blood-brain barrier (BBB) and exert central effects remains underexplored. In this study, we demonstrate for the first time that LCA enhances cognitive function in a lipopolysaccharide (LPS)-induced neuroinflammatory mouse model and effectively penetrates the BBB.

View Article and Find Full Text PDF

Sulforaphane Repairs Oxidative Stress Damage Induced by Oxidized Fish Oil by Activating in .

Aquac Nutr

August 2025

Guangdong Provincial Key Laboratory of Aquatic Animal Disease Control and Healthy Culture and Key Laboratory of Control for Disease of Aquatic Animals of Guangdong Higher Education Institutes, Fisheries College, Guangdong Ocean University, Zhanjiang, China.

Nuclear factor erythroid 2-related factor 2 (Nrf2) is an essential component in regulating oxidative stress. Sulforaphane (SFN) is a natural antioxidant and gene agonist that can increase the antioxidant capacity of the organism and reduce oxidative stress. However, research on the repair of oxidative stress damage by SFN in aquatic animals remains extremely scarce.

View Article and Find Full Text PDF