Evaluation Methodology between Globalization and Localization Features Approaches for Skin Cancer Lesions Classification

Ahmed, Hasan; Al-Azawi, Razi J.; Abdulhameed, Abbas Abdulazeez

doi:10.1088/1742-6596/1003/1/012029

Cited by 5 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2013, another study [7] was conducted to examine skin cancer diagnosis by both national and local methods. This research pursued two objectives: determining the best diagnostic method (global or local) and determining the best color or texture features used in the designed experiments of three different classifications of Knn, Svm, and Adaboost.…”

Section: Relevant Studiesmentioning

confidence: 99%

See 1 more Smart Citation

Classification of Melanoma Images Using Empirical Wavelet Transform

Fadaeian¹,

Rahmani²,

Javid³

2021

RCES

View full text Add to dashboard Cite

Skin cancer is the most common cancer, accounting for 75% of all skin cancers worldwide. Malignant melanoma is the most invasive type of skin cancer, which is deadly. Some techniques have been investigated to diagnose skin diseases using skin tissue classification and diagnosis models and skin recognition approaches using colors based on image retrieval methods. In this regard, image processing techniques and classification methods are intelligent. The purpose of this method, diagnosing melanoma skin cancer using image processing. In the proposed method, after collecting the dataset, the boundary to separate the skin lesion from the background was specified. Then in the next step, the analysis was performed using Empirical wavelet transform (EWT). Then the color, texture, and shape features were extracted. In the next step, the feature was selected by Gray Wolf meta-heuristic algorithm using ranking models and the disease was classified into two categories, namely normal and abnormal. The database used in this study contains 594 dermatoscopic images with a resolution of 512 × 768 pixels, 476 images with normal spots, and 88 images with abnormal spots caused by melanoma. The evaluation results revealed that the proposed method had an accuracy of 97.25, indicating its significant performance compared to other methods. The contribution of the results of the proposed method can be very useful and valuable in the future for early detection of skin cancer.

show abstract

Section: Relevant Studiesmentioning

confidence: 99%

“…In another study in 2013 [7] the role of shape features in the classification of images in dermoscopy was investigated. The proposed algorithm consists of three steps.…”

Section: Relevant Studiesmentioning

confidence: 99%

Classification of Melanoma Images Using Empirical Wavelet Transform

Fadaeian¹,

Rahmani²,

Javid³

2021

RCES

View full text Add to dashboard Cite

show abstract

“…Malignant Melanoma is a deadly form of skin cancer that is responsible for over 20,000 deaths in Europe with close to 100,000 new cases recorded each year [1]. World Health Organization (WHO) claimed that approximately 13% of the global mortality rate is associated with cancer-related diseases and this is still on the rise as future estimate predict that 12 million more are at risk of death by the end of 2030 [2]. Recent studies have shown that early detection of this disease is the most critical prognostic factor for survival [3].…”

Section: Introductionmentioning

confidence: 99%

“…2 dataset[35] collected by the Dermatology Service of Hospital Pedro Hispano, Portugal. 200 images are comprising of 3 classes of Clinical Diagnosis, namely Common Nevus, Atypical Nevus, and Melanoma with a sample ratio of 80:80:40 samples, respectively.…”

mentioning

confidence: 99%

Malignant skin melanoma detection using image augmentation by oversampling in nonlinear lower-dimensional embedding manifold

Abayomi‐Alli¹,

Damaševičius²,

Misra³

et al. 2021

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

The continuous rise in skin cancer cases, especially in malignant melanoma, has resulted in a high mortality rate of the affected patients due to late detection. Some challenges affecting the success of skin cancer detection include small datasets or data scarcity problem, noisy data, imbalanced data, inconsistency in image sizes and resolutions, unavailability of data, reliability of labeled data (ground truth), and imbalance of skin cancer datasets. This study presents a novel data augmentation technique based on covariant Synthetic Minority Oversampling TEchnique (SMOTE) to address the data scarcity and class imbalance problem. We propose an improved data augmentation model for effective detection of melanoma skin cancer. Our method is based on data oversampling in a non-linear lower-dimensional embedding manifold for creating synthetic melanoma images. The proposed data augmentation technique is used to generate a new skin melanoma dataset using dermoscopic images from the publicly available P H 2 dataset. The augmented images were used to train the SqueezeNet deep learning model. The experimental results in binary classification scenario show a significant improvement in detection of melanoma with respect to accuracy (92.18%), sensitivity (80.77%), specificity (95.1%), and F1-score (80.84%). We also improved the multi-class classification results in melanoma detection to 89.2% (sensitivity), 96.2% (specificity) for atypical nevus detection, 65.4% (sensitivity), 72.2% (specificity), and for common nevus detection 66% (sensitivity), 77.2% (specificity). The proposed classification framework outperforms some of the state-of-the-art methods in detecting skin melanoma.

show abstract

Automatic Skin Lesion Segmentation and Melanoma Detection: Transfer Learning Approach with U-Net and DCNN-SVM

Nazi

Abir

2019

Algorithms for Intelligent Systems

View full text Add to dashboard Cite

Evaluation Methodology between Globalization and Localization Features Approaches for Skin Cancer Lesions Classification

Cited by 5 publications

References 9 publications

Classification of Melanoma Images Using Empirical Wavelet Transform

Classification of Melanoma Images Using Empirical Wavelet Transform

Malignant skin melanoma detection using image augmentation by oversampling in nonlinear lower-dimensional embedding manifold

Automatic Skin Lesion Segmentation and Melanoma Detection: Transfer Learning Approach with U-Net and DCNN-SVM

Contact Info

Product

Resources

About