Tackling the Problem of Data Imbalancing for Melanoma Classification

Abstract : Malignant melanoma is the most dangerous type of skin cancer, yet melanoma is the most treatable kind of cancer when diagnosed at an early stage. In this regard, Computer-Aided Diagnosis systems based on machine learning have been developed to discern melanoma lesions from benign and dysplastic nevi in dermoscopic images. Similar to a large range of real world applications encountered in machine learning, melanoma classification faces the challenge of imbalanced data, where the percentage of melanoma cases in comparison with benign and dysplastic cases is far less. This article analyzes the impact of data balancing strategies at the training step. Subsequently, Over-Sampling (OS) and Under-Sampling (US) are extensively compared in both feature and data space, revealing that NearMiss-2 (NM2) outperform other methods achieving Sensitivity (SE) and Specificity (SP) of 91.2% and 81.7%, respectively. More generally, the reported results highlight that methods based on US or combination of OS and US in feature space outperform the others.
Type de document :
Communication dans un congrès
Bioimaging, Feb 2016, Rome, Italy
Liste complète des métadonnées

Littérature citée [19 références]  Voir  Masquer  Télécharger

https://hal-univ-bourgogne.archives-ouvertes.fr/hal-01250949
Contributeur : Guillaume Lemaitre <>
Soumis le : mardi 5 janvier 2016 - 14:29:26
Dernière modification le : jeudi 7 janvier 2016 - 01:01:58
Document(s) archivé(s) le : jeudi 7 avril 2016 - 15:23:42

Fichier

master(5).pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01250949, version 1

Collections

Citation

Mojdeh Rastgoo, Guillaume Lemaitre, Joan Massich, Olivier Morel, Franck Marzani, et al.. Tackling the Problem of Data Imbalancing for Melanoma Classification. Bioimaging, Feb 2016, Rome, Italy. 〈hal-01250949〉

Partager

Métriques

Consultations de
la notice

226

Téléchargements du document

180