Paper Details
Review of Identification and categorization of skin cancer using a Convolutional Neural Network
Authors
Archana Prajapati, Arpita Dash, Priya
Abstract
Skin cancer, a prevalent and potentially lethal condition, necessitates accurate diagnosis for effective treatment and management. Recent advancements in deep learning, particularly the use of Convolutional Neural Networks (CNNs), have revolutionized the identification and categorization of skin cancer, providing a potent alternative to traditional diagnostic methods. This review paper systematically explores various CNN architectures and methodologies employed in the detection and classification of skin cancer. We evaluate the performance of well-known models such as VGG-16, VGG-19, and ResNet, alongside custom-built CNNs specifically tailored for dermatological imagery. This review discusses the datasets typically used in this field, including the International Skin Imaging Collaboration (ISIC) dataset, highlighting the challenges and successes in applying CNNs to these complex image sets. We also examine preprocessing techniques, feature extraction methods, and the effectiveness of different activation and optimization functions. Moreover, the review delves into comparative studies that demonstrate the accuracy, sensitivity, and specificity of CNN models in distinguishing between malignant and benign lesions, offering insights into their clinical applicability and potential to enhance early detection rates. Finally, future directions and potential improvements in the algorithmic approach to skin cancer diagnostics are proposed, aiming to bridge the gap between technical advancements and clinical practice. This comprehensive review aims to underscore the transformative impact of CNNs in dermatology, paving the way for more personalized and precise skin cancer treatments.
Keywords
Skin Cancer , International Skin Imaging Collaboration , Convolutional Neural Network,VGG-16, VGG-19, ResNet
Citation
Review of Identification and categorization of skin cancer using a Convolutional Neural Network. Archana Prajapati, Arpita Dash, Priya. 2024. IJIRCT, Volume 10, Issue 3. Pages 1-5. https://www.ijirct.org/viewPaper.php?paperId=2406072