Deep‐learning approach in the study of skin lesions

Filipescu, Stefan-Gabriel; Butacu, Alexandra-Irina; Ţiplica, George‐Sorin; Nastac, Dumitru Iulian

doi:10.1111/srt.13045

Cited by 7 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the most common Lyme disease symptoms is a rash known as erythema migrans, which can appear at the tick bite site. The common symptoms of Lyme disease include cardiovascular problems, neurological disorders, musculoskeletal troubles, and skin disorders 7,8 . Globally, Lyme disease is increasing in prevalence, and rising temperatures may contribute.…”

Section: Introductionmentioning

confidence: 99%

“…The common symptoms of Lyme disease include cardiovascular problems, neurological disorders, musculoskeletal troubles, and skin disorders. 7 , 8 Globally, Lyme disease is increasing in prevalence, and rising temperatures may contribute. British Medical Journal Global Health reports that 14.5% of the world's population has been infected with Lyme disease.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Lyme rashes disease classification using deep feature fusion technique

Ali,

Anwar,

Nauman

et al. 2023

Skin Research and Technology

View full text Add to dashboard Cite

Automatic classification of Lyme disease rashes on the skin helps clinicians and dermatologists’ probe and investigate Lyme skin rashes effectively. This paper proposes a new in‐depth features fusion system to classify Lyme disease rashes. The proposed method consists of two main steps. First, three different deep learning models, Densenet201, InceptionV3, and Exception, were trained independently to extract the deep features from the erythema migrans (EM) images. Second, a deep feature fusion mechanism (meta classifier) is developed to integrate the deep features before the final classification output layer. The meta classifier is a basic deep convolutional neural network trained on original images and features extracted from base level three deep learning models. In the feature fusion mechanism, the last three layers of base models are dropped out and connected to the meta classifier. The proposed deep feature fusion method significantly improved the classification process, where the classification accuracy was 98.97%, which is particularly impressive than the other state‐of‐the‐art models.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Lyme rashes disease classification using deep feature fusion technique

Ali,

Anwar,

Nauman

et al. 2023

Skin Research and Technology

View full text Add to dashboard Cite

show abstract

“…Filipescu et al 27 . employed a pretrained CNN VGG‐16 to classify PSLs in the ISIC‐2019 dataset, comprising eight classes.…”

Section: Related Workmentioning

confidence: 99%

“…Filipescu et al27 employed a pretrained CNN VGG-16 to classify PSLs in the ISIC-2019 dataset, comprising eight classes. The researchers conducted preprocessing by resizing images to match the pre-trained model and optimizing hyper-parameters.…”

mentioning

confidence: 99%

Boosting the performance of pretrained CNN architecture on dermoscopic pigmented skin lesion classification

Nugroho,

Ardiyanto,

Nugroho

2023

Skin Research and Technology

View full text Add to dashboard Cite

BackgroundPigmented skin lesions (PSLs) pose medical and esthetic challenges for those affected. PSLs can cause skin cancers, particularly melanoma, which can be life‐threatening. Detecting and treating melanoma early can reduce mortality rates. Dermoscopic imaging offers a noninvasive and cost‐effective technique for examining PSLs. However, the lack of standardized colors, image capture settings, and artifacts makes accurate analysis challenging. Computer‐aided diagnosis (CAD) using deep learning models, such as convolutional neural networks (CNNs), has shown promise by automatically extracting features from medical images. Nevertheless, enhancing the CNN models' performance remains challenging, notably concerning sensitivity.Materials and methodsIn this study, we aim to enhance the classification performance of selected pretrained CNNs. We use the 2019 ISIC dataset, which presents eight disease classes. To achieve this goal, two methods are applied: resolution of the dataset imbalance challenge through augmentation and optimization of the training hyperparameters via Bayesian tuning.ResultsThe performance improvement was observed for all tested pretrained CNNs. The Inception‐V3 model achieved the best performance compared to similar results, with an accuracy of 96.40% and an AUC of 0.98.ConclusionAccording to the study, classification performance was significantly enhanced by augmentation and Bayesian hyperparameter tuning.

show abstract

“…Moreover, the diagnosis of skin lesions yields higher management predictions accuracy. Filipescu et al 51 presented an application of precision medicine—a deep learning approach to computer‐aided diagnosis in the field of dermatology. Over the ISIC 2019 challenge dataset, they achieved overall accuracies of 78.11%.…”

Section: Related Workmentioning

confidence: 99%

HS‐UNET‐ID: An approach for human skin classification integrating between UNET and improved dense convolutional network

Nguyen

Luong

Phan

et al. 2022

Int J Imaging Syst Tech

View full text Add to dashboard Cite

The Sun's ultraviolet radiation, toxic chemicals from industry, and other factors can cause problems to human skin. These factors have boosted the risk of humans contracting skin cancer. In identifying and detecting signals of skin cancer diseases, recent advancements in deep learning techniques significantly improve the accuracy of diagnosis. This work delves into processing techniques for human skin images with MorphologyEx (blackhat), thresholding, semantic segmentation with UNET, and several famous convolutional neural network architectures to improve skin classification performance. If skin cancer can be stopped in the first stage, the survival rate of the patient could be more than 95%, while it drops to just 5% if the detection is carried lately in Stage IV4. Due to its seriousness, the healthcare and research community has received considerable attention, and its most important goal is to be able to diagnose it at an early stage. However, the challenge due to the similarity in melanocytic and non-melanoma skin lesions is inevitable. We achieved skin classification performances of 0.8560, 0.9672, 0.8063, and 0.8560 measured by overall accuracy, area under curve, Matthews correlation coefficient, and F1-score, respectively. The proposed approach has been deployed into applications on Mobile and Web platforms.

show abstract

Deep‐learning approach in the study of skin lesions

Cited by 7 publications

References 10 publications

Lyme rashes disease classification using deep feature fusion technique

Lyme rashes disease classification using deep feature fusion technique

Boosting the performance of pretrained CNN architecture on dermoscopic pigmented skin lesion classification

HS‐UNET‐ID: An approach for human skin classification integrating between UNET and improved dense convolutional network

Contact Info

Product

Resources

About