Cervix Type Classification Using Convolutional Neural Networks

Autor:	Elizabeth Guevara, Daniel A. Cruz, Carmen Villar-Patiño, Marisol Martinez-Alanis
Rok vydání:	2019
Předmět:	Computer science business.industry Pattern recognition Convolutional neural network Image (mathematics) Set (abstract data type) Identification (information) medicine.anatomical_structure Region of interest medicine Preprocessor Artificial intelligence business Transfer of learning Cervix
Zdroj:	IFMBE Proceedings ISBN: 9783030306472
DOI:	10.1007/978-3-030-30648-9_49
Popis:	Cervical cancer is still a significant cause of death, especially in developing countries. The detection and correct treatment of the disease is vital in its early stages. One of the key factors in selecting appropriate treatment is the identification of the cervix type. The objective of this work is to propose a Convolutional Neural Network (CNN) architecture to perform a classification of the cervix from a set of images published by Intel and MobileODT. The proposed architecture is combined with a preprocessing algorithm based on an assembly of color models to select the region of interest in the image. The obtained model provides better results than other models in which transfer learning is used or there is no preprocessing stage.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::f5365e36044b0ea5c6f9153eb6fdd7b7 https://doi.org/10.1007/978-3-030-30648-9_49 Zobrazit plný text záznamu