Medical data classification is typically a challenging task due to imbalance between classes. In this paper, we propose an approach to classify dermatoscopic images from HAM10000 (Human Against Machine with 10000 training images) dataset, consisting of seven imbalanced types of skin lesions, with good precision and low resources requirements. Classification is done by using a pretrained convolutional neural network. We evaluate the accuracy and performance of the proposal and illustrate possible extensions.

Skin Cancer Classification Using Inception Network and Transfer Learning / Priscilla Benedetti; Damiano Perri; Marco Simonetti; Osvaldo Gervasi; Gianluca Reali; Mauro Femminella. - ELETTRONICO. - LNTCS 12249:(2020), pp. 536-545. (Intervento presentato al convegno International Conference on Computational Science and Its Applications tenutosi a Cagliari nel 01/07/2020 - 04/07/2020) [10.1007/978-3-030-58799-4_39].

Skin Cancer Classification Using Inception Network and Transfer Learning

Damiano Perri
;
Marco Simonetti
;
2020

Abstract

Medical data classification is typically a challenging task due to imbalance between classes. In this paper, we propose an approach to classify dermatoscopic images from HAM10000 (Human Against Machine with 10000 training images) dataset, consisting of seven imbalanced types of skin lesions, with good precision and low resources requirements. Classification is done by using a pretrained convolutional neural network. We evaluate the accuracy and performance of the proposal and illustrate possible extensions.
2020
Computational Science and Its Applications - ICCSA 2020
International Conference on Computational Science and Its Applications
Cagliari
01/07/2020 - 04/07/2020
Priscilla Benedetti; Damiano Perri; Marco Simonetti; Osvaldo Gervasi; Gianluca Reali; Mauro Femminella
File in questo prodotto:
File Dimensione Formato  
2111.02402.pdf

accesso aperto

Descrizione: Skin Cancer
Tipologia: Versione finale referata (Postprint, Accepted manuscript)
Licenza: Creative commons
Dimensione 1.02 MB
Formato Adobe PDF
1.02 MB Adobe PDF

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1293224
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 12
  • ???jsp.display-item.citation.isi??? 10
social impact