A Deep Learning Framework with an Intermediate Layer Using the Swarm Intelligence Optimizer for Diagnosing Oral Squamous Cell Carcinoma

• Published in: Diagnostics (Basel) 2023-11, Vol.13 (22), p.3461
• Main Authors: Nagarajan, Bharanidharan, Chakravarthy, Sannasi, Venkatesan, Vinoth Kumar, Ramakrishna, Mahesh Thyluru, Khan, Surbhi Bhatia, Basheer, Shakila, Albalawi, Eid
• Format: Article
• Language: English
• Subjects: Accuracy; Algorithms; Biomarkers; Biopsy; Cell growth; Classification; CNN; Decision making; Deep learning; deep learning framework; Diagnosis; Feature selection; Gorilla Troops Optimizer; Health aspects; histopathologic images; Machine learning; Medical diagnosis; Medical imaging equipment; Neural networks; Optimization; Oral cancer; Parameter identification; Spectrum analysis; Squamous cell carcinoma; Support vector machines; swarm intelligence; Tumors
• ISSN: 2075-4418; 2075-4418
• DOI: 10.3390/diagnostics13223461

One of the most prevalent cancers is oral squamous cell carcinoma, and preventing mortality from this disease primarily depends on early detection. Clinicians will greatly benefit from automated diagnostic techniques that analyze a patient’s histopathology images to identify abnormal oral lesions. A deep learning framework was designed with an intermediate layer between feature extraction layers and classification layers for classifying the histopathological images into two categories, namely, normal and oral squamous cell carcinoma. The intermediate layer is constructed using the proposed swarm intelligence technique called the Modified Gorilla Troops Optimizer. While there are many optimization algorithms used in the literature for feature selection, weight updating, and optimal parameter identification in deep learning models, this work focuses on using optimization algorithms as an intermediate layer to convert extracted features into features that are better suited for classification. Three datasets comprising 2784 normal and 3632 oral squamous cell carcinoma subjects are considered in this work. Three popular CNN architectures, namely, InceptionV2, MobileNetV3, and EfficientNetB3, are investigated as feature extraction layers. Two fully connected Neural Network layers, batch normalization, and dropout are used as classification layers. With the best accuracy of 0.89 among the examined feature extraction models, MobileNetV3 exhibits good performance. This accuracy is increased to 0.95 when the suggested Modified Gorilla Troops Optimizer is used as an intermediary layer.