Classification of skin lesions using transfer learning and augmentation with Alex-net

Hosny, Khalid M.; Kassem, Mohamed A.; Foaud, Mohamed M.

doi:10.1371/journal.pone.0217293

Cited by 258 publications

(149 citation statements)

References 33 publications

Supporting

Mentioning

145

Contrasting

Unclassified

Order By: Relevance

“…Some excellent studies [15]- [18] have shown that there is considerable redundancy in a traditional VGG model, and layer-by-layer connections make the network replicate features of the front layer in the whole network. Hosny et al [19], [20] performed several experiments using different models, such as Alex-net, ResNet, VGG, and GoogleNet. Their results showed that the VGG model required substantial memory and high-configuration hardware.…”

Section: Introductionmentioning

confidence: 99%

Improving the Performance of VGG Through Different Granularity Feature Combinations

Zhou

Chang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Convolutional neural networks have achieved amazing success in many areas in recent years, and VGG is a widely used convolutional neural network model. However, it has some limitations, such as a large number of parameters, which take up significant memory and thus restrict its application in resourceconstrained scenarios such as mobile devices and embedded systems. In convolutional neural network models, the different number of convolutional layers can extract different granularity features, which represent the different levels of importance in the image recognition process. Here, we propose a new VGG architecture with different granularity feature combinations that combine different granularity features from block1, block2, block3, block4, and block5 in VGG. Each block is followed by a local fully connected layer to reduce the dimensionality of the coarse and fine features, and five different granularity features are combined as the input of the first global fully connected layer. By combining the features of different blocks, it can increase information flow from a lower layer directly to a fully connected layer and increase feature reuse without adding too many connections. The addition of five local fully connected layers means an increase in parameters, so we reduce the neuron number in two global fully connected layers to reduce the number of parameters. The well-known datasets CIFAR-10 and MNIST were used to evaluate the network's classification performance. The experimental results show that the proposed model achieves better training and testing performance than traditional VGGs and reduces the number of parameters.

show abstract

Section: Introductionmentioning

confidence: 99%

Improving the Performance of VGG Through Different Granularity Feature Combinations

Zhou

Chang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Since the image processing techniques were developed, Computer-Aided Detection (CAD) systems and approaches in the classification [ 3 , 4 , 5 , 6 , 7 ] and segmentation [ 8 ] of a Pigmented Skin Lesion (PSL) have been improved, benefiting patient diagnoses in early stages of the disease without shocking or painful medical procedures.…”

Section: Introductionmentioning

confidence: 99%

Melanoma and Nevus Skin Lesion Classification Using Handcraft and Deep Learning Feature Fusion via Mutual Information Measures

Almaraz-Damian

Ponomaryov

Sadovnychiy

et al. 2020

Entropy

130

View full text Add to dashboard Cite

In this paper, a new Computer-Aided Detection (CAD) system for the detection and classification of dangerous skin lesions (melanoma type) is presented, through a fusion of handcraft features related to the medical algorithm ABCD rule (Asymmetry Borders-Colors-Dermatoscopic Structures) and deep learning features employing Mutual Information (MI) measurements. The steps of a CAD system can be summarized as preprocessing, feature extraction, feature fusion, and classification. During the preprocessing step, a lesion image is enhanced, filtered, and segmented, with the aim to obtain the Region of Interest (ROI); in the next step, the feature extraction is performed. Handcraft features such as shape, color, and texture are used as the representation of the ABCD rule, and deep learning features are extracted using a Convolutional Neural Network (CNN) architecture, which is pre-trained on Imagenet (an ILSVRC Imagenet task). MI measurement is used as a fusion rule, gathering the most important information from both types of features. Finally, at the Classification step, several methods are employed such as Linear Regression (LR), Support Vector Machines (SVMs), and Relevant Vector Machines (RVMs). The designed framework was tested using the ISIC 2018 public dataset. The proposed framework appears to demonstrate an improved performance in comparison with other state-of-the-art methods in terms of the accuracy, specificity, and sensibility obtained in the training and test stages. Additionally, we propose and justify a novel procedure that should be used in adjusting the evaluation metrics for imbalanced datasets that are common for different kinds of skin lesions.

show abstract

“…The main point of CNN is that it is able to map important features of images that can be used to classify those images without the CNN's being explicitly programmed to do so. CNN has been proven to work with great accuracy in the classification of many medical domains like diabetic retinopathy detection and classification [4,5], Alzheimer's disease detection [6,7], and skin lesion detection [8,9], among others.…”

Section: Introductionmentioning

confidence: 99%

“…The medical field could benefit a lot from the usage of CNN for image classification, but the main drawback is the number of images available for training, and that is where transfer learning can be used. Using transfer learning for medical images has been successfully applied in References [4,6,8]. As pointed out by Chollet [15] , transfer learning uses two main techniques: fine-tuning, wherein the original weights will be re-tuned to suit the new dataset, or feature extraction, in which the original weights will be fixed and the original layers will serve for feature extraction only.…”

Section: Introductionmentioning

confidence: 99%

How Deeply to Fine-Tune a Convolutional Neural Network: A Case Study Using a Histopathology Dataset

Kandel

Castelli

2020

Applied Sciences

View full text Add to dashboard Cite

Accurate classification of medical images is of great importance for correct disease diagnosis. The automation of medical image classification is of great necessity because it can provide a second opinion or even a better classification in case of a shortage of experienced medical staff. Convolutional neural networks (CNN) were introduced to improve the image classification domain by eliminating the need to manually select which features to use to classify images. Training CNN from scratch requires very large annotated datasets that are scarce in the medical field. Transfer learning of CNN weights from another large non-medical dataset can help overcome the problem of medical image scarcity. Transfer learning consists of fine-tuning CNN layers to suit the new dataset. The main questions when using transfer learning are how deeply to fine-tune the network and what difference in generalization that will make. In this paper, all of the experiments were done on two histopathology datasets using three state-of-the-art architectures to systematically study the effect of block-wise fine-tuning of CNN. Results show that fine-tuning the entire network is not always the best option; especially for shallow networks, alternatively fine-tuning the top blocks can save both time and computational power and produce more robust classifiers.

show abstract

Classification of skin lesions using transfer learning and augmentation with Alex-net

Cited by 258 publications

References 33 publications

Improving the Performance of VGG Through Different Granularity Feature Combinations

Improving the Performance of VGG Through Different Granularity Feature Combinations

Melanoma and Nevus Skin Lesion Classification Using Handcraft and Deep Learning Feature Fusion via Mutual Information Measures

How Deeply to Fine-Tune a Convolutional Neural Network: A Case Study Using a Histopathology Dataset

Contact Info

Product

Resources

About