“…Late diagnosis is sometimes caused by a combination of medical professionals and patients' ignorance about oral lesions. The screening program has focused mostly on diagnosing oral potentially malignant disorders (OPMD) due to the danger of cancerous transformation, its enormous value in reducing mortality and morbidity from oral malignancies, and its prevalence [5]. However, using this program in real-time situations has been challenging since it relies on visual examination, and healthcare workers often lack the knowledge or training to correctly detect this lesion [6,7].…”
Oral cancer is a deadly form of cancerous tumor that is widely spread in low and middleincome countries. An early and affordable oral cancer diagnosis might be achieved by automating the detection of precancerous and malignant lesions in the mouth. There are many research attempts to develop a robust machine-learning model that can detect oral cancer from images. However, these are still lacking high precision in oral cancer detection. Therefore, this work aims to propose a new approach capable of detecting oral cancer in medical images with higher accuracy. In this work, a novel and robust oral cancer detection based on a convolutional neural network (CNN) and optimized deep belief network (DBN). The design parameters of CNN and DBN are optimized using a new optimization algorithm, which is developed as a hybrid of Particle Swarm Optimization (PSO) and Al-Biruni Earth Radius (BER) Optimization algorithms and is denoted by (PSOBER). Using a standard biomedical images dataset available on the Kaggle repository, the proposed approach shows promising results outperforming various competing approaches with an accuracy of 97.35%. In addition, a set of statistical tests, such as One-way analysis-of-variance (ANOVA) and Wilcoxon signed-rank tests, are conducted to prove the significance and stability of the proposed approach. The proposed methodology is solid and efficient, and specialists can adopt it. However, additional research on a larger scale dataset is required to confirm the findings and highlight other oral features that can be utilized for cancer detection.
“…Late diagnosis is sometimes caused by a combination of medical professionals and patients' ignorance about oral lesions. The screening program has focused mostly on diagnosing oral potentially malignant disorders (OPMD) due to the danger of cancerous transformation, its enormous value in reducing mortality and morbidity from oral malignancies, and its prevalence [5]. However, using this program in real-time situations has been challenging since it relies on visual examination, and healthcare workers often lack the knowledge or training to correctly detect this lesion [6,7].…”
Oral cancer is a deadly form of cancerous tumor that is widely spread in low and middleincome countries. An early and affordable oral cancer diagnosis might be achieved by automating the detection of precancerous and malignant lesions in the mouth. There are many research attempts to develop a robust machine-learning model that can detect oral cancer from images. However, these are still lacking high precision in oral cancer detection. Therefore, this work aims to propose a new approach capable of detecting oral cancer in medical images with higher accuracy. In this work, a novel and robust oral cancer detection based on a convolutional neural network (CNN) and optimized deep belief network (DBN). The design parameters of CNN and DBN are optimized using a new optimization algorithm, which is developed as a hybrid of Particle Swarm Optimization (PSO) and Al-Biruni Earth Radius (BER) Optimization algorithms and is denoted by (PSOBER). Using a standard biomedical images dataset available on the Kaggle repository, the proposed approach shows promising results outperforming various competing approaches with an accuracy of 97.35%. In addition, a set of statistical tests, such as One-way analysis-of-variance (ANOVA) and Wilcoxon signed-rank tests, are conducted to prove the significance and stability of the proposed approach. The proposed methodology is solid and efficient, and specialists can adopt it. However, additional research on a larger scale dataset is required to confirm the findings and highlight other oral features that can be utilized for cancer detection.
“…Earlier identification of OSCC gains significant importance for improved diagnosis, treatment, and survival [ 5 , 6 ]. Late diagnosis has hampered the quest for precision medicine in spite of the advancements in the understanding of the molecular mechanism of cancer.…”
Section: Introductionmentioning
confidence: 99%
“…The lack of knowledge of health professionals and lack of public awareness concerning oral lesions are major reasons for late diagnosis. The OPMD diagnosis has a risk of malignant transformation, is of great significance to reduce mortality and morbidity from oral tumors, and has been the major emphasis of the screening program [ 6 ]. But the application of this program depends on visual inspection has turned out to be challenging in real-time settings as they depend on healthcare professionals, who are not experienced or adequately trained to identify this lesion [ 7 , 8 ].…”
Oral cancer is one of the lethal diseases among the available malignant tumors globally, and it has become a challenging health issue in developing and low-to-middle income countries. The prognosis of oral cancer remains poor because over 50% of patients are recognized at advanced stages. Earlier detection and screening models for oral cancer are mainly based on experts’ knowledge, and it necessitates an automated tool for oral cancer detection. The recent developments of computational intelligence (CI) and computer vision-based approaches help to accomplish enhanced performance in medical-image-related tasks. This article develops an intelligent deep learning enabled oral squamous cell carcinoma detection and classification (IDL-OSCDC) technique using biomedical images. The presented IDL-OSCDC model involves the recognition and classification of oral cancer on biomedical images. The proposed IDL-OSCDC model employs Gabor filtering (GF) as a preprocessing step to eliminate noise content. In addition, the NasNet model is exploited for the generation of high-level deep features from the input images. Moreover, an enhanced grasshopper optimization algorithm (EGOA)-based deep belief network (DBN) model is employed for oral cancer detection and classification. The hyperparameter tuning of the DBN model is performed using the EGOA algorithm which in turn boosts the classification outcomes. The experimentation outcomes of the IDL-OSCDC model using a benchmark biomedical imaging dataset highlighted its promising performance over the other methods with maximum
accu
y
,
prec
n
,
reca
l
, and
F
score
of 95%, 96.15%, 93.75%, and 94.67% correspondingly.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.