Skin Lesion Classification towards Melanoma Diagnosis using Convolutional Neural Network and Image Enhancement Methods

다국어 초록 (Multilingual Abstract)

Melanoma is the deadliest form of skin lesion which is a severe disease globally. Early detection of melanoma using medical images significantly increases the survival rate. However, the accurate recognition of melanoma is extremely challenging task. Since the joint use of image enhancement techniques and deep convolutional neural network (DCNN) has demonstrated proven success, the joint techniques could have discriminatory power on skin lesion diagnosis as well. To this hypothesis, we propose the aggregation algorithm for skin lesion diagnosis that utilize a DCNN to extract the local features classify medical images for melanoma disease. All experiments are performed using the data provided in International Skin Imaging Collaboration (ISIC) 2018 Skin Lesion Analysis towards Melanoma Detection. Experimental results show that our algorithm achieves excellent classification results for melanoma diagnosis.

목차 (Table of Contents)

• Abstract
• Ⅰ. Introduction
• Ⅱ. Related Work
• Ⅲ. Proposed Approach
• Ⅳ. Experiments and Results
• Ⅴ. Conclusion
• References

참고문헌 (Reference)

1 Chollet. F, "Xception Deep learning with depthwise separable convolutions" 84 (84): 1063-1076, 2017

2 Glorot, Xavier, "Understanding the difficulty of training deep feedforward neural networks" 63 (63): 249-256, 2010

3 Tschandl P, "The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions" 47 (47): 52-61, 2018

4 "Source of the image"

5 Noel C. F. Codella, "Skin Lesion Analysis Toward Melanoma Detection, Hosted by the International Skin Imaging Collaboration (ISIC)" 28 (28): 182-195, 2017

6 K. A. Freedberg, "Screening for malignant melanoma : A cost-effectiveness analysis" 41 (41): 738-745, 1999

7 Szegedy. C, "Rethinking the inception architecture for computer vision" 18 (18): 2818-2826, 2016

8 Cireşan Dan, "Multi-column deep neural networks for image classification" 56 (56): 124-131, 2015

9 Szegedy. C, "Inception-v4, inception-resnet and the impact of residual connections on learning" 4 : 12-, 2017

10 Namozov, "Efficient Deep Learning Algorithm for Fire and Smoke Detection with Limited Data" 18 (18): 121-128, 2018

11 Zou, Jinyi, "Dermoscopic Image Analysis for ISIC Challenge 2018"

12 Iandola. F, "Densenet Implementing efficient convnet descriptor pyramids"

13 He. K, "Deep residual learning for image recognition" 164 (164): 770-778, 2016

14 LeCun, Yann, "Deep learning Nature" 16 (16): 75-83, 2015

15 W. Zhang, "Computerized detection of clustered micro calcifications in digital mammograms using a shift-invariant artificial neural network" 42 (42): 254-361, 2001

16 R. L. Siegel, "Cancer Statistics" 68 (68): 7-30, 2018

17 L. Yu, "Automated Melanoma Recognition in Dermoscopy Images via Very Deep Residual Networks" 36 (36): 994-1004, 2017

18 Deng, Jia, "A large-scale hierarchical image database" 32 (32): 248-255, 2009