Identificador persistente para citar o vincular este elemento: http://hdl.handle.net/10553/52534
Título: Hyperspectral database of pathological in-vitro human brain samples to detect carcinogenic tissues
Autores/as: Ortega, S. 
Callico, G. M. 
Plaza, M.L. 
Camacho, R
Fabelo, H. 
Sarmiento, R. 
Clasificación UNESCO: 3314 Tecnología médica
Palabras clave: Artificial neural network
Brain cancer detection
Data mining
Hyperspectral imaging
Support vector machine
Fecha de publicación: 2016
Publicación seriada: Proceedings - International Symposium on Biomedical Imaging
Conferencia: 13th IEEE International Symposium on Biomedical Imaging (ISBI) 
2016 IEEE 13th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2016 
Resumen: Hyperspectral imaging is an emerging technology for medical diagnosis. Some previous studies have used this type of images to detect cancer diseases. In this research work, a multidisciplinary team conformed by pathologists and engineers has created a diagnosed hyperspectral database of in-vitro human brain tissues. In order to capture the hyperspectral information from histological slides, an acquisition system based on a microscope coupled with a hyperspectral camera has been developed. Preliminary results of applying two different supervised classification algorithms (Support Vector Machines and Artificial Neural Networks) to the hyperspectral database show that an automatic discrimination between healthy and tumour brain tissues from in-vitro samples is possible using exclusively their spectral information. The sensitivity and the specificity are over 92% in all the cases.
URI: http://hdl.handle.net/10553/52534
ISBN: 9781479923502
ISSN: 1945-7928
DOI: 10.1109/ISBI.2016.7493285
Fuente: Proceedings - International Symposium on Biomedical Imaging[ISSN 1945-7928],v. 2016-June (7493285), p. 369-372
Colección:Actas de congresos
Vista completa

Google ScholarTM

Verifica

Altmetric


Comparte



Exporta metadatos



Los elementos en ULPGC accedaCRIS están protegidos por derechos de autor con todos los derechos reservados, a menos que se indique lo contrario.