Feature extraction , selection and classification in Dermoscopy : an experimental comparison of intelligent methods to support decision making in medicine

Melanoma, the most dangerous skin cancer, is sometimes associated with nevus, a relatively common skin lesion. To early find melanoma, nevus and other lesions, dermoscopy is often used. In this context, intelligent methods have been applied in dermoscopic images to support decision making. A typical computer-aided diagnosis method comprises three steps: (1) extraction of features that describe image properties, (2) selection of important features previously extracted, (3) classification of images based on the selected features. In this work, traditional data mining approaches underexploited in dermoscopy were applied: information gain for feature selection and an ensemble classification method based on gradient boosting. The former technique ranks image features according to data entropy, while the latter one combines the outputs of single classifiers to predict the image class. After evaluating these approaches in a public dataset with 104 dermoscopic images, we found that they are competitive with a state of the art approach.

[1]  S. Robbins,et al.  Pathologic basis of disease , 1974 .

[2]  Igor Kononenko,et al.  Estimating Attributes: Analysis and Extensions of RELIEF , 1994, ECML.

[3]  M. G. Fleming,et al.  Dermoscopy of pigmented skin lesions: results of a consensus meeting via the Internet. , 2003, Journal of the American Academy of Dermatology.

[4]  P. Carli,et al.  Improvement of malignant/benign ratio in excised melanocytic lesions in the ‘dermoscopy era’: a retrospective study 1997–2001 , 2004, The British journal of dermatology.

[5]  Hiroshi Motoda,et al.  Computational Methods of Feature Selection , 2022 .

[6]  R. Suganya,et al.  Data Mining Concepts and Techniques , 2010 .

[7]  Zhi-Hua Zhou,et al.  Ensemble Methods: Foundations and Algorithms , 2012 .

[8]  Jorge S. Marques,et al.  Bag-of-Features Classification Model for the Diagnose of Melanoma in Dermoscopy Images Using Color and Texture Descriptors , 2013, ICIAR.

[9]  Jorge S. Marques,et al.  The Role of Keypoint Sampling on the Classification of Melanomas in Dermoscopy Images Using Bag-of-Features , 2013, IbPRIA.

[10]  Jorge S. Marques,et al.  Evaluation of Color Based Keypoints and Features for the Classification of Melanomas Using the Bag-of-Features Model , 2013, ISVC.

[11]  Jorge S. Marques,et al.  Two Systems for the Detection of Melanomas in Dermoscopy Images Using Texture and Color Features , 2014, IEEE Systems Journal.

[12]  อนิรุธ สืบสิงห์,et al.  Data Mining Practical Machine Learning Tools and Techniques , 2014 .

[13]  Jorge S. Marques,et al.  Improving dermoscopy image analysis using color constancy , 2014, 2014 IEEE International Conference on Image Processing (ICIP).

[14]  M. Faezipour,et al.  A comparison of feature sets for an automated skin lesion analysis system for melanoma early detection and prevention , 2015, 2015 Long Island Systems, Applications and Technology.

[15]  R. H. Moss,et al.  Real‐time supervised detection of pink areas in dermoscopic images of melanoma: importance of color shades, texture and location , 2015, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[16]  Jorge S. Marques,et al.  Improving Dermoscopy Image Classification Using Color Constancy , 2015, IEEE Journal of Biomedical and Health Informatics.

[17]  Franck Marzani,et al.  Automatic differentiation of melanoma from dysplastic nevi , 2015, Comput. Medical Imaging Graph..

[18]  Rui Fonseca-Pinto,et al.  Classification of reticular pattern and streaks in dermoscopic images based on texture analysis , 2015, Journal of medical imaging.

[19]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[20]  María Pérez-Ortiz,et al.  Classification of Melanoma Presence and Thickness Based on Computational Image Analysis , 2016, HAIS.

[21]  Pedro Antonio Gutiérrez,et al.  Machine Learning Methods for Binary and Multiclass Classification of Melanoma Thickness From Dermoscopic Images , 2016, IEEE Transactions on Medical Imaging.

[22]  María Pérez-Ortiz,et al.  Tackling the ordinal and imbalance nature of a melanoma image classification problem , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[23]  Sejung Yang,et al.  Ridge and furrow pattern classification for acral lentiginous melanoma using dermoscopic images , 2017, Biomed. Signal Process. Control..

[24]  Newton Spolaôr,et al.  Dermoscopic assisted diagnosis in melanoma: Reviewing results, optimizing methodologies and quantifying empirical guidelines , 2018, Knowl. Based Syst..

[25]  Hiroki Sayama,et al.  A network-based classification framework for predicting treatment response of schizophrenia patients , 2018, Expert Syst. Appl..

[26]  Daniel Riccio,et al.  Learning-based approach to segment pigment signs in fundus images for Retinitis Pigmentosa analysis , 2018, Neurocomputing.

[27]  Verónica Bolón-Canedo,et al.  Recent Advances in Ensembles for Feature Selection , 2018, Intelligent Systems Reference Library.