Skin cancer screening using deep learning and convolution Neural Network

Author

Mafuva, Accuman A.

Title

Skin cancer screening using deep learning and convolution Neural Network

Abstract

Skin lesion is a severe disease in world-wide extent. Early detection of melanoma in dermoscopy images significantly increases the survival rate. However, the accurate recognition of melanoma is extremely challenging due to premature assumption and human error emerging from fatigue and inconsistent frame mind. Hence, reliable automatic detection and classification detection skin tumours is very useful to increase the accuracy and efficiency of pathologists. On the other hand, Artificial intelligence is becoming the major thing of this high-tech industries. The innovation directed by the best tech companies are affecting industry verticals such as auto, finance, retail, manufacturing and to be precise, healthcare. This has called for medical industry to adopt Artificial Intelligence in automating medical operations. In this paper, the researcher proposed a deep learning system to address lesion dermoscopic feature extraction and lesion classification. The system entailing of convolutional neural networks is proposed to produce the classify skin lesions. Although there are some screening systems already, they are still restrictive and use specialised and expensive tools. The proposed system enables the users (pathologists) to take a picture of the lesion from a dermatoscope, make image classification print the class of the lesion and probability of likeliness.

Date

June 2024

Publisher

BUSE

Keywords

Skin lesion,

Melanoma in dermoscopy,

Artificial Intelligence,

Supervisor

Dr. Gombiro,