Authors:
B. Santosh Kumar, Thirumalraj Karthikeyan, R. J. Anandhi, S. Venkatasubramanian, Chou Yi Hsu
Addresses:
Department of Computer Science and Engineering, New Horizon College of Engineering, Bengaluru, Karnataka, India. Department of Artificial Intelligence, Quest Technologies, Trichy Research Labs, Tiruchirappalli, Tamil Nadu, India. Department of Information Science and Engineering, New Horizon College of Engineering, Bengaluru, Karnataka, India. Department of Computer Science and Business Systems, Saranathan College of Engineering, Trichy, Tamil Nadu, India. Department of Pharmacy, Chia Nan University of Pharmacy and Science, Rende, Tainan, Taiwan.
Abstract:
Among women, cervical cancer is a widespread and treatable cancer that is a long-term disease. The most effective method for quickly diagnosing cervical cancer is Pap smear testing. The approach used to detect cervical cancer focuses on using new image-processing techniques and better classification parameters. The process is initialised with Gabor preprocessing, which improves crucial features in cervical images using the SIPaKMeD dataset. GhostNet uses Artificial Ecosystem-based Optimisation (AEO) to hyperparameter-tune and ensure better model performance. The classification phase is empowered by an Enhanced DenseNet architecture, specifically AdaBound-Squeeze and Excitation-DenseNet. It integrates adaptive optimisation and spatial squeeze-and-excitation mechanisms for improved learning and feature representation. The proposed framework achieves 99.41% accuracy, 99.41% precision, 99.43% recall, and 99.45% F-score, which are better results than those of other existing models. The method has been proposed, tested, and its efficacy has been confirmed. The technique can be applied in clinical settings to help detect cervical cancer early.
Keywords: Gabor Filter; Ghost Network; Squeeze and Excitation; Dense Network; Artificial Ecosystem Optimisation; Cancer Diagnoses; Cervical Cancer; Identification and Treatment; Health Care Services.
Received: 20/10/2024, Revised: 30/01/2025, Accepted: 13/04/2025, Published: 07/09/2025
DOI: 10.64091/ATIHL.2025.000173
AVE Trends in Intelligent Health Letters, 2025 Vol. 2 No. 3 , Pages: 150-162