Computer-aided diagnosis (CAD) system based on multi-layer feature fusion network for skin lesion recognition in dermoscopy images

Bakkouri, Ibtissam; Afdel, Karim

doi:10.1007/s11042-019-07988-1

Cited by 90 publications

(31 citation statements)

References 93 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As indicated in our previous work [9], pre-processing of the digital dermoscopic data is performed in four stages, including data size normalization technique, hair artifact removal, intensity correction process and class balancing to facilitate application of DermoNet algorithm. The determinant of the Hessian combined with pyramidal REDUCE decomposition was carried out to normalize data size.…”

Section: Pre-processing Of Dermoscopic Datamentioning

confidence: 99%

“…As the dermoscopic image volumes of different categories vary widely, data augmentation is needed to balance the image volumes of different classes. In this paper, we balanced the dataset by augmenting the data for the minority classes, applying geometric transformation techniques as indicated in [9]. Finally, after data preparation process, the number of dermoscopic images is expanded at least by a factor of 20.…”

Section: Pre-processing Of Dermoscopic Datamentioning

confidence: 99%

“…The dataset consists of 10015 dermoscopic images with size of 600 × 450 showing skin lesions which have been diagnosed based on expert consensus, serial imaging and histopathology. The distribution is completely imbalanced, and then we applied the geometric transformation techniques as described in [9] to overcome class prevalence. After balancing process, we obtained 2000 images per class.…”

Section: Data Acquisitionmentioning

confidence: 99%

“…The proposed method was evaluated on ISIC 2017 challenge dataset, yielding satisfactory results. In this context, the similar approach based on convolutional multi-layer feature fusion network and continuous fine-tuning of CNNs using VGG-16, ResNet-18 and DenseNet-121 architectures were suggested by authors of [9] in order to classify seven skin diseases. The proposed method was evaluated on the testing subset of the HAM10000, and it reached promising results.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

DermoNet: A Computer-Aided Diagnosis System for Dermoscopic Disease Recognition

Bakkouri

Afdel

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

The research of skin lesion diseases is currently one of the hottest topics in the medical research fields, and has gained a lot of attention on the last few years. However, the existing skin lesion methods are mainly relying on conventional Convolutional Neural Network (CNN) and the performance of skin lesion recognition is far from satisfactory. Therefore, to overcome the aforementioned drawbacks of traditional methods, we propose a novel Computer-Aided Diagnosis (CAD) system, named DermoNet, based on Multi-Scale Feature Level (MSFL) blocks and Multi-Level Feature Fusion (MLFF). Further, the DermoNet approach yields a significant enhancement in terms of dealing with the challenge of small training data sizes in the dermoscopic domain and avoiding high similarity between classes and overfitting issue. Extensive experiments are conducted on the public dermoscopic dataset, and the results demonstrate that DermoNet outperforms the state-of-the-art approaches. Hence, DermoNet can achieve an excellent diagnostic efficiency in the auxiliary diagnosis of skin lesions.

show abstract

Section: Pre-processing Of Dermoscopic Datamentioning

confidence: 99%

Section: Pre-processing Of Dermoscopic Datamentioning

confidence: 99%

Section: Data Acquisitionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

DermoNet: A Computer-Aided Diagnosis System for Dermoscopic Disease Recognition

Bakkouri

Afdel

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the preprocessing step, hair removal stands out as one of the most useful and used methods. However, traditional approaches are still used for this task in more advanced systems in which the main model uses deep learning techniques [11], [12]. Thus, we face the task of developing a deep learning model for the detection and removal of hairs.…”

Section: Introductionmentioning

confidence: 99%

Hair Segmentation and Removal in Dermoscopic Images Using Deep Learning

2021

View full text Add to dashboard Cite

Comparison of the impacts of dermoscopy image augmentation methods on skin cancer classification and a new augmentation method with wavelet packets

Göçeri

2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

This work aims to determine the most suitable technique for dermoscopy image augmentation to improve the performance of lesion classifications. Also, a new augmentation technique based on wavelet packet transformations has been developed. The contribution of this work is five‐fold. First, a comprehensive review of the methods used for dermoscopy image augmentation has been presented. Second, a new augmentation method has been developed. Third, the augmentation methods have been implemented with the same images for meaningful comparisons. Fourth, three network architectures have been implemented to see the effects of the augmented images obtained from each augmentation method on classifications. Fifth, the results of the same classifier trained separately using expanded data sets have been compared with five different metrics. The proposed augmentation method increases the classification accuracy by at least 4.77% compared with the accuracy values obtained from the same classifier with other augmentation methods.

show abstract

Computer-aided diagnosis (CAD) system based on multi-layer feature fusion network for skin lesion recognition in dermoscopy images

Cited by 90 publications

References 93 publications

DermoNet: A Computer-Aided Diagnosis System for Dermoscopic Disease Recognition

DermoNet: A Computer-Aided Diagnosis System for Dermoscopic Disease Recognition

Hair Segmentation and Removal in Dermoscopic Images Using Deep Learning

Comparison of the impacts of dermoscopy image augmentation methods on skin cancer classification and a new augmentation method with wavelet packets

Contact Info

Product

Resources

About