Rate the Article: Modeling of Dragonfly Algorithm with Deep Learning for Skin Cancer Diagnosis on Dermoscopic Images, IJSR, Call for Papers, Online Journal
International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
Call for Papers | Fully Refereed | Open Access | Double Blind Peer Reviewed

ISSN: 2319-7064

Downloads: 2 | Views: 198 | Weekly Hits: ⮙2 | Monthly Hits: ⮙2

Research Paper | Computer Science and Information Technology | India | Volume 12 Issue 10, October 2023 | Rating: 5.2 / 10


Modeling of Dragonfly Algorithm with Deep Learning for Skin Cancer Diagnosis on Dermoscopic Images

Vijay Arumugam R., Saravanan S.


Abstract: Skin cancer, a widespread and possibly life-threatening condition, requires an earlier and accurate diagnosis for efficient involvement. Dermoscopic images, providing a window into skin lesions, gives a useful resource for medical specialists in this context. This article goals to develop a complex architecture to detect skin cancer employing dermatoscopy images, integrating cutting-edge techniques in image analysis, machine learning, and artificial intelligence. This manuscript introduces the Dragonfly Algorithm with Deep Learning for Skin Cancer Diagnoses on Dermoscopy Images (DFADL-SCDDI) method. Our technique integrates advanced methods namely feature extraction, preprocessing, classification, and parameter optimization to increase the reliability and of accuracy identification. For image preprocessing, we exploit the Gabor filter (GF), a robust tool for enriching texture and structure data in images. Feature extraction has been executed employing a Capsule Network (CapsNet). CapsNet is a deep learning (DL) model that exceeds in capturing hierarchical and in-depth features in images. Classification is conducted by a Gated Recurrent Unit (GRU), a kind of recurrent neural network (RNN) ability to model sequential patterns and dependencies within the feature representations. To additional improve the model's effectiveness; we implement the Dragonfly Algorithm (DA) for parameter tuning. The DA has a powerful optimization system stimulated by nature, developed for enhancing hyperparameters efficiently, consequently higher the model's diagnostic accuracy. The proposed architecture is assessed on a large database of dermatoscopy images, signifying its effectiveness in skin cancer detection. The outcomes exhibit substantial enhancement in reliability and accuracy compared to traditional systems.


Keywords: Dragonfly Algorithm; Dermoscopy; Gated Recurrent Unit; Deep Learning; Parameter Tuning


Edition: Volume 12 Issue 10, October 2023,


Pages: 1916 - 1922



Rate this Article


Select Rating (Lowest: 1, Highest: 10)

5

Your Comments (Only high quality comments will be accepted.)

Characters: 0

Your Full Name:


Your Valid Email Address:


Verification Code will appear in 2 Seconds ... Wait

Top