Smart Skin Care: Deep Learning in Skin Cancer Detection
KAVITHA.G
Head & Professor Computer Science
and Engineering Muthayammal Engineering
College Rasipuram,Namakkal,
Tamilnadu,India
hod.cse@mec.edu.in
HARIPPRIYA.S
Computer Science and Engineering
Muthayammal Engineering College Rasipuram
,Namakkal,Tamilnadu,India
harippriyasiva@gmail.com
LAKSHANA.R
Computer Science and Engineering
Muthayammal Engineering College
Rasipuram,Namakkal,Tamilnadu,India
lakshanar029@gmail.com
ABSTRACT
Skin cancer is one of the ten most common cancer types worldwide, primarily caused by abnormal skin cell growth due to DNA mutations from sun exposure. Early detection of melanoma, the most dangerous type of skin cancer, is crucial to prevent mortality and financial burdens. This study introduces a deep learning-based methodology for diagnosing skin cancer by analyzing dermatologic spot images. The proposed system utilizes Convolutional Neural Networks (CNNs) to automatically extract lesion characteristics such as color, area, perimeter, diameter, texture, and shape, classifying them as melanoma or non- melanoma. Various training techniques, including data augmentation and transfer learning, are implemented to improve model generalization and performance.
Additionally, ensemble learning techniques such as Max Voting (Majority Voting) are applied to increase the reliability of the classification process. The system is designed to efficiently analyze skin lesion images, ensuring accurate and early diagnosis. It reduces dependency on traditional diagnostic methods while enhancing medical decision-making. Based on experimental evaluations, deep learning models, particularly CNN-based architectures, demonstrated superior accuracy in distinguishing melanoma from non-melanoma cases. By leveraging deep learning and advanced image processing techniques, this system aims to contribute to the medical field by providing a reliable and accessible solution for skin cancer detection.
Key Words: Skin Cancer Detection, Deep Learning, Convolutional Neural Networks, EfficientNet, Image Processing, Medical Image Analysis, Early Diagnosis
