Optimized Convolutional Neural Network Models for Skin Lesion Classification

Villa-Pulgarin, Juan Pablo; Ruales-Torres, Anderson Alberto; Arias-Garz髇, Daniel; Bravo-Ortíz, Mario Alejandro; Arteaga-Arteaga, Harold Brayan; Mora-Rubio, Alejandro; Alzate-Grisales, Jesús Alejandro; Mercado-Ruiz, Esteban; Hassaballah, M.; Orozco-Arias, Simón; Cardona-Morales, Oscar; Tabares-Soto, Reinel

doi:10.32604/cmc.2022.019529

Cited by 23 publications

(20 citation statements)

References 35 publications

(54 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model achieved an accuracy of 94.92%, a sensitivity of 79.8%, and a specificity of 97% for classifying eight classes of the ISIC 2019 dataset. Juan et al [ 25 ] propose three CNN models to classify dermoscopy images for the HAM10000 and ISIC 2018 datasets, and their results are compared. Inception-V3 achieved 96% and 93% accuracy for the HAM10000 and ISIC 2018 datasets, respectively.…”

Section: Related Workmentioning

confidence: 99%

AI Techniques of Dermoscopy Image Analysis for the Early Detection of Skin Lesions Based on Combined CNN Features

et al. 2023

View full text Add to dashboard Cite

Melanoma is one of the deadliest types of skin cancer that leads to death if not diagnosed early. Many skin lesions are similar in the early stages, which causes an inaccurate diagnosis. Accurate diagnosis of the types of skin lesions helps dermatologists save patients’ lives. In this paper, we propose hybrid systems based on the advantages of fused CNN models. CNN models receive dermoscopy images of the ISIC 2019 dataset after segmenting the area of lesions and isolating them from healthy skin through the Geometric Active Contour (GAC) algorithm. Artificial neural network (ANN) and Random Forest (Rf) receive fused CNN features and classify them with high accuracy. The first methodology involved analyzing the area of skin lesions and diagnosing their type early using the hybrid models CNN-ANN and CNN-RF. CNN models (AlexNet, GoogLeNet and VGG16) receive lesions area only and produce high depth feature maps. Thus, the deep feature maps were reduced by the PCA and then classified by ANN and RF networks. The second methodology involved analyzing the area of skin lesions and diagnosing their type early using the hybrid CNN-ANN and CNN-RF models based on the features of the fused CNN models. It is worth noting that the features of the CNN models were serially integrated after reducing their high dimensions by Principal Component Analysis (PCA). Hybrid models based on fused CNN features achieved promising results for diagnosing dermatoscopic images of the ISIC 2019 data set and distinguishing skin cancer from other skin lesions. The AlexNet-GoogLeNet-VGG16-ANN hybrid model achieved an AUC of 94.41%, sensitivity of 88.90%, accuracy of 96.10%, precision of 88.69%, and specificity of 99.44%.

show abstract

Section: Related Workmentioning

confidence: 99%

AI Techniques of Dermoscopy Image Analysis for the Early Detection of Skin Lesions Based on Combined CNN Features

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Villa-Pulgarin et al. ( 44 ) developed the DenseNet-201, Inception-V3, and Inception-ResNet-V2 deep models using TL. The dataset HAM10000 was used to evaluate these models in which the DenseNet-201 model performed best on the International Skin Imaging Collaboration (ISIC) 2019 dataset with an achieved accuracy of 93%.…”

Section: Related Workmentioning

confidence: 99%

A novel framework of multiclass skin lesion recognition from dermoscopic images using deep learning and explainable AI

et al. 2023

View full text Add to dashboard Cite

Skin cancer is a serious disease that affects people all over the world. Melanoma is an aggressive form of skin cancer, and early detection can significantly reduce human mortality. In the United States, approximately 97,610 new cases of melanoma will be diagnosed in 2023. However, challenges such as lesion irregularities, low-contrast lesions, intraclass color similarity, redundant features, and imbalanced datasets make improved recognition accuracy using computerized techniques extremely difficult. This work presented a new framework for skin lesion recognition using data augmentation, deep learning, and explainable artificial intelligence. In the proposed framework, data augmentation is performed at the initial step to increase the dataset size, and then two pretrained deep learning models are employed. Both models have been fine-tuned and trained using deep transfer learning. Both models (Xception and ShuffleNet) utilize the global average pooling layer for deep feature extraction. The analysis of this step shows that some important information is missing; therefore, we performed the fusion. After the fusion process, the computational time was increased; therefore, we developed an improved Butterfly Optimization Algorithm. Using this algorithm, only the best features are selected and classified using machine learning classifiers. In addition, a GradCAM-based visualization is performed to analyze the important region in the image. Two publicly available datasets—ISIC2018 and HAM10000—have been utilized and obtained improved accuracy of 99.3% and 91.5%, respectively. Comparing the proposed framework accuracy with state-of-the-art methods reveals improved and less computational time.

show abstract

“…The created models are implent and the learning process takes place. Figure 2 shows a simplified diagram of the example architecture of the created DenseNet-201 network model, which was used in [43] to classify dermatoscopic images. It was created on the basis of the Keras library [33], which is used in transfer learning.…”

Section: Training Networkmentioning

confidence: 99%

“…Currently developed models allow the classification of 5, 6 and even 7 different skin diseases [28,29]. The use of more dermatoscopic images and more complex models of convolutional networks allowed to achieve great success in the diagnostic process [25,43,44]. Even papers are being created comparing the diagnostic capabilities of neural networks and dermatologists [13].…”

Section: Introductionmentioning

confidence: 99%

Multiclass Skin Less Ions Classification Based on Deep Neural Networks

Michalska

2022

IAPGOS

View full text Add to dashboard Cite

Skin diseases diagnosed with dermatoscopy are becoming more and more common. The use of computerized diagnostic systems becomes extremely effective. Non-invasive methods of diagnostics, such as deep neural networks, are an increasingly common tool studied by scientists. The article presents an overview of selected main issues related to the multi-class classification process: the stage of database selection, initial image processing, selection of the learning data set, classification tools, network training stage and obtaining final results. The described actions were implemented using available deep neural networks. The article pay attention to the final results of available models, such as effectiveness, specificity, classification accuracy for different numbers of classes and available data sets.

show abstract

Optimized Convolutional Neural Network Models for Skin Lesion Classification

Cited by 23 publications

References 35 publications

AI Techniques of Dermoscopy Image Analysis for the Early Detection of Skin Lesions Based on Combined CNN Features

AI Techniques of Dermoscopy Image Analysis for the Early Detection of Skin Lesions Based on Combined CNN Features

A novel framework of multiclass skin lesion recognition from dermoscopic images using deep learning and explainable AI

Multiclass Skin Less Ions Classification Based on Deep Neural Networks

Contact Info

Product

Resources

About