Evaluation of state-of-the-art deep learning algorithms for multispectral skin cancer images classification

It is a present need to diagnose skin cancer sooner, faster and non-invasively. With recent achievements in the use of non-invasive imaging techniques and learning algorithms for Computer-Aided Diagnosis, the present work focuses on evaluating and adapting the different state-of-the-art artificial i...

Descripción completa

Detalles Bibliográficos
Autor: Robredo Del Rey, Diego
Tipo de recurso: tesis de maestría
Fecha de publicación:2021
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/363291
Acceso en línea:https://hdl.handle.net/2117/363291
Access Level:acceso abierto
Palabra clave:Machine learning
Skin--Cancer
Multispectral imaging
multispectral imaging
machine learning
cnn
skin cancer
Aprenentatge automàtic
Pell--Càncer
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo
Descripción
Sumario:It is a present need to diagnose skin cancer sooner, faster and non-invasively. With recent achievements in the use of non-invasive imaging techniques and learning algorithms for Computer-Aided Diagnosis, the present work focuses on evaluating and adapting the different state-of-the-art artificial intelligence algorithms to classify a set of multispectral images of skin lesions acquired in +400 patients with a multispectral system that works in the visible and near infrared range. It is intended that in the future, biopsies, which are the gold standard to identify equivocal lesions, will be carried out in a very unusual way