Lesion Classification Using Convolutional Neural Network

Malignant melanoma is uncommon in India as compared to the Western nations. However, its growth in recent years has been significant. Early detection of malignant skin lesions can help in proper cure. All recent works on automated classification of skin lesions generate a set of features based on the lesion segment such as lesion diameter and texture. The lesions are then classified into malignant and benign classes based on these features. In our work, we use convolutional neural networks (CNNs) with LeNet architecture in order to automate the feature extraction and selection process. We classify skin lesions in binary class of malignant and benign using ISBI 2016 and PH2 data set with an accuracy of 75% and 97.91%, respectively.

[1]  Omar Abuzaghleh,et al.  Noninvasive Real-Time Automated Skin Lesion Analysis System for Melanoma Early Detection and Prevention , 2015, IEEE Journal of Translational Engineering in Health and Medicine.

[2]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[3]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[4]  Jacob Scharcanski,et al.  Automated prescreening of pigmented skin lesions using standard cameras , 2011, Comput. Medical Imaging Graph..

[5]  Isabelle Guyon,et al.  An Introduction to Feature Extraction , 2006, Feature Extraction.

[6]  P. Carli,et al.  Malignant melanoma in Europe: changes in mortality rates (1970-90) in European Community countries , 2004, Cancer Causes & Control.

[7]  Dagan Feng,et al.  Dermoscopic Image Segmentation via Multistage Fully Convolutional Networks , 2017, IEEE Transactions on Biomedical Engineering.

[8]  R. H. Moss,et al.  A fuzzy-based histogram analysis technique for skin lesion discrimination in dermatology clinical images. , 2003, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[9]  Nitesh V. Chawla,et al.  Data Mining for Imbalanced Datasets: An Overview , 2005, The Data Mining and Knowledge Discovery Handbook.

[10]  David Dagan Feng,et al.  Automated skin lesion segmentation via image-wise supervised learning and multi-scale superpixel based cellular automata , 2016, 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI).

[11]  A. Jemal,et al.  Cancer statistics, 2018 , 2018, CA: a cancer journal for clinicians.

[12]  Randy H. Moss,et al.  A methodological approach to the classification of dermoscopy images , 2007, Comput. Medical Imaging Graph..

[13]  B. Cox,et al.  A comparison of trends in melanoma mortality in New Zealand and Australia: the two countries with the highest melanoma incidence and mortality in the world , 2013, BMC Cancer.

[14]  Lucia Ballerini,et al.  Fuzzy description of skin lesions , 2010, Medical Imaging.

[15]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[16]  Rebecca L. Siegel Mph,et al.  Cancer statistics, 2018 , 2018 .

[17]  G. Argenziano,et al.  Epiluminescence microscopy for the diagnosis of doubtful melanocytic skin lesions. Comparison of the ABCD rule of dermatoscopy and a new 7-point checklist based on pattern analysis. , 1998, Archives of dermatology.

[18]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[19]  Pedro M. Ferreira,et al.  PH2 - A dermoscopic image database for research and benchmarking , 2013, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[20]  David A. Clausi,et al.  High-Level Intuitive Features (HLIFs) for Intuitive Skin Lesion Description , 2015, IEEE Transactions on Biomedical Engineering.

[21]  Enoch Peserico,et al.  Simpler, Faster, More Accurate Melanocytic Lesion Segmentation Through MEDS , 2014, IEEE Transactions on Biomedical Engineering.

[22]  Stephen W Dusza,et al.  The CASH (color, architecture, symmetry, and homogeneity) algorithm for dermoscopy. , 2007, Journal of the American Academy of Dermatology.

[23]  Xavier Giro-i-Nieto,et al.  Skin lesion classification from dermoscopic images using deep learning techniques , 2017, 2017 13th IASTED International Conference on Biomedical Engineering (BioMed).

[24]  Qiang Chen,et al.  Accurate and Scalable System for Automatic Detection of Malignant Melanoma , 2015 .

[25]  Wolfgang M. Hartmann,et al.  Dimension Reduction vs. Variable Selection , 2004, PARA.

[26]  Robert B. Fisher,et al.  A Color and Texture Based Hierarchical K-NN Approach to the Classification of Non-melanoma Skin Lesions , 2013 .

[27]  Pierre Baldi,et al.  Learning Activation Functions to Improve Deep Neural Networks , 2014, ICLR.