Skin Disease Images Research Articles

Identification of Skin Disease Using Deep Learning

Globally, skin diseases are among the most common health problems in all humans irrespective of age. Prevention and early detection of these diseases can improve the chance of surviving. This model illustrates the identification of skin diseases providing more objective and reliable solutions using deep learning technology and convolutional neural network approach. In this model, the system design, implementation and identification of common skin diseases such as acne, blister, eczema, warts etc. are explained. The system applies deep learning technology to train itself with various images of skin diseases from the Kaggle platform. The accuracy obtained by using deep learning is 83.23%. The main objective of this system is to achieve maximum accuracy of skin disease prediction. Moreover, if the disease is identified the system provides detailed information about the diseases along with home remedies.

Successful regional teledermatological and teledermatoscopic triage networking between general practitioners and dermatologists

There is an increasing need by general practitioners (GPs) for consultations regarding skin diseases or skin lesions. Teledermatological and teledermoscopic diagnoses are possible with good results. This descriptive study analyzed whether the increasing need for dermatological consultations could be improved by regional teledermatological and teledermoscopic triage between GPs and dermatologists. Via an internet platform, six GPs sent images of unclear skin diseases or skin lesions to adermatology practice (2dermatologists) over aperiod of 9months. Possible triage answers were (1)stay at the GP, (2)urgent consultation at the dermatologist (days), or (3)later (week/months) consultation at the dermatologist, respectively, with the diagnosis and therapeutic suggestions. Afeedback (grade1-6) were given by the dermatologists about the image quality and from the GPs about the appraisal. The dermatologists' appraisals were done independently and were audited by athird dermatologist. The patients' acceptance was 100% and avery good to good image quality was achieved in 94%; 66.3% of patients could stay at their GP and 20.7% of patients should come urgently to the dermatologist. The teletriage decisions of an urgent consultation at the dermatologist were confirmed in 41.5%. The GPs' feedback about the teletriage appraisal received an overall average grade of1.1. Successful regional teledermatological and teledermoscopic triage between GPs and dermatologists is possible. Specifications on data privacy laws must be followed and an appropriate remuneration is preferable.

Image investigation using higher moment statistics and edge detection for recognizing abnormal skin

This paper discusses the recognition of skin abnormalities by investigating the images using high-level statistics and edge detection. Six images of skin diseases were analyzed using several statistical parameters, including high-level moments such as skewness and kurtosis. In comparison, some images from other categories such as animal, architecture, art, vehicle, food, people, and scenery have been analyzed as well. The results were compared to skin disease images. It is expected that the general pattern of statistical parameters can distinguish skin images against images from other categories. MatLab is used as a medium to calculate the values of statistical parameters. The mean and median of the skin disease image are much larger. Meanwhile, the standard deviation is the smallest compared to other categorical images. Almost all the analyzed images close to symmetry. Nearly all images category are distributed more leaning to the left, except for the images of the art category, which is slightly more leaning to the right. Moreover, the edge detection process has been done using the Sobel algorithm. The result, however, cannot clearly distinguish a skin's abnormality. This difficulty is because of the lack of accuracy in selection the intensity.

Discriminative Feature Learning for Skin Disease Classification Using Deep Convolutional Neural Network

Nowadays, skin disease among humans has been a common disease, especially in America millions of people are suffering from various kinds of skin disease. Usually, these diseases have hidden dangers which lead to human not only lack of self-confidence and psychological depression but also a risk of skin cancer. Diagnosis of these kinds of diseases usually required medical experts with high-level instruments due to a lack of visual resolution in skin disease images. Moreover, manual diagnosis of skin disease is often subjective, time-consuming, and required more human effort. Thus, there is a need to develop a computer-aided system that automatically diagnoses the skin disease problem. Moreover, most of the existing works in skin disease used convolutional neural networks (CNN) with classical loss functions, which limit the model to learn discriminative features from skin images. Thus to address the above mention problem we proposed a new framework by fine-tuning layers of ResNet152 and InceptionResNet-V2 models with a triplet loss function. In the proposed framework, first, we learning the embedding from input images into Euclidean space by using deep CNN ResNet152 and InceptionResNet-V2 model. Second, we compute L-2 distance among corresponding images from euclidean space to learn discriminative features of skin disease images by using triplet loss function. Finally, classify the input images using these L-2 distances. Human face skin disease images used in the proposed framework are acquired from the Hospital in Wuhan China. Experiment results and their analysis shows the effectiveness of the proposed framework which achieve better accuracy than many existing works in skin disease tasks.

Deep Learning in Skin Disease Image Recognition: A Review

The application of deep learning methods to diagnose diseases has become a new research topic in the medical field. In the field of medicine, skin disease is one of the most common diseases, and its visual representation is more prominent compared with the other types of diseases. Accordingly, the use of deep learning methods for skin disease image recognition is of great significance and has attracted the attention of researchers. In this study, we review 45 research efforts on the identification of skin disease by using deep learning technology since 2016. We analyze these studies from the aspects of disease type, data set, data processing technology, data augmentation technology, model for skin disease image recognition, deep learning framework, evaluation indicators, and model performance. Moreover, we summarize the traditional and machine learning-based skin disease diagnosis and treatment methods. We also analyze the current progress in this field and predict four directions that may become the research topic in the future. Our results show that the skin disease image recognition method based on deep learning is better than those of dermatologists and other computer-aided treatment methods in skin disease diagnosis, especially the multi deep learning model fusion method has the best recognition effect.

Aversion caused by skin diseases - a crowdsourcing study.

Disgust is a universal emotion that plays a role in reducing people's exposure to situations with risks of contamination. By the same token, skin diseases could generate aversion in observers. The present study aimed to assess the aversion triggered when viewing skin disease lesions on the face and hands. A questionnaire was applied over the Internet containing demographic questions, assessment of personal sensitivity to disgust - by the Disgust Scale-Revised (DS-R) -, and the aversion triggered when viewing eight images of skin diseases using an image aversion score (IAS) ranging from 0 to 10. Six-hundred forty-nine people responded, 49% female, with a median age of 26 (p25-p75: 22-34). A total of 2596 evaluations of images were collected. The overall IAS median was 2 (0-4), and Cronbach's alpha resulted 0.87. There was a significant correlation between DS-R and IAS (Spearman's rho = 0.42; P < 0.01). Seven-hundred twenty-four (28%) evaluations were considered as contagious diseases and received higher IAS: 5 (2-7) × 1 (0-3); P < 0.01. In multivariate analysis, the higher IAS was independently associated with male gender, with no children, not working as a health professional or caregiver, and presenting higher DS-R (P < 0.01). By analyzing the IAS according to morphological characteristics, we identified lower scores for macular dyschromic lesions (P < 0.01). The aversion triggered by skin lesions varies according to the personal characteristics of the observer and the morphology of the lesion, being lower for macular dyschromic lesions, and associated with disgust sensitivity and the assignment of contagiousness to the lesion.

Le musée de l’hôpital Saint-Louis

In 1866, Alphonse Devergie, physician supported by Armand Husson, manager of the Assistance publique created a permanent exhibition of teaching images of skin diseases at the Hôpital Saint-Louis (Paris). Twenty-five years later the initiative was made concrete by the inauguration of a building housing a museum, a medical library and rooms for outpatients. Enriched by the works of dermatologists, moulageurs - mainly Jules Baretta -, painters and photographs the museum became worldwide renown exhibiting the largest wax moulages collection of the world. Today the didactic value of the moulages is no longer used. The collection illustrates several aspects of hospital dermatology during a century. Listed as an Historical Monument in 1992, the museum has been recently renewed thanks to the support of private and public actors attracted by its patrimonial value. © 2018 Published by Elsevier Masson SAS. All rights reserved. © 2018 Elsevier Masson SAS. Tous droits réservés.

Skin Disease Recognition Method Based on Image Color and Texture Features.

Skin diseases have a serious impact on people's life and health. Current research proposes an efficient approach to identify singular type of skin diseases. It is necessary to develop automatic methods in order to increase the accuracy of diagnosis for multitype skin diseases. In this paper, three type skin diseases such as herpes, dermatitis, and psoriasis skin disease could be identified by a new recognition method. Initially, skin images were preprocessed to remove noise and irrelevant background by filtering and transformation. Then the method of grey-level co-occurrence matrix (GLCM) was introduced to segment images of skin disease. The texture and color features of different skin disease images could be obtained accurately. Finally, by using the support vector machine (SVM) classification method, three types of skin diseases were identified. The experimental results demonstrate the effectiveness and feasibility of the proposed method.

Development of Dermatological Skin Disease Detection and Classification based on Wavelet and ANN

One of the most common and augmenting health problems in the world are related to skin. The most unpredictable and one of the most difficult entities to automatically detect and evaluate is the human skin disease because of complexities of texture, tone, presence of hair and other distinctive features. Many cases of skin diseases in the world have triggered a need to develop an effective automated screening method for detection and diagnosis of the area of disease. Therefore the objective of this work is to develop a new technique for automated detection and analysis of the skin disease images based on color and texture information for skin disease screening. In this paper, system is proposed which detects the skin diseases using Wavelet Techniques and Artificial Neural Network. This paper presents a wavelet-based texture analysis method for classification of five types of skin diseases. The method applies tree-structured wavelet transform on different color channels of red, green and blue dermoscopy images, and employs various statistical measures and ratios on wavelet coefficients. In all 99 unique features are extracted from the image. By using Artificial Neural Network, the system successfully detects different types of dermatological skin diseases. It consists of mainly three phases image processing, training phase, detection and classification phase.

Review: Dermatological Skin Disease Detection and Classification Based on Wavelet Transform and Artificial Neural Network

One of the most common and augmenting health problems in the world are related to skin. The most unpredictable and one of the most difficult entities to automatically detect and evaluate is the human skin disease because of complexities of texture, tone, presence of hair and other distinctive features. Many cases of skin diseases in the world have triggered a need to develop an effective automated screening method for detection and diagnosis of the area of disease. Therefore the objective of this work is to develop a new technique for automated detection and analysis of the skin disease images based on color and texture information for skin disease screening. In this paper, system is proposed which detects the skin diseases using Wavelet Techniques and Artificial Neural Network. This paper presents a wavelet-based texture analysis method for classification of five types of skin diseases. The method applies tree-structured wavelet transform on different color channels of red, green and blue dermoscopy images, and employs various statistical measures and ratios on wavelet coefficients. In all 99 unique features are extracted from the image. By using Artificial Neural Network, the system successfully detects different types of dermatological skin diseases. It consists of mainly three phases image processing, training phase, detection and classification phase.

Content Based Image Retrieval Using Lacunarity and Color Moments Of Skin Diseases

&lt;p&gt;&lt;em&gt;The research conducted contributed in the form of CBIR application which was developed using texture and color feature extraction in searching the contents information of an object of skin disease image. The textured feature is extracted using Lacunarity, while for color feature extraction using Color Moments as well as a combination of both methods. The results of color characteristic extraction test using Color Moments Method yielded images corresponding to 100% similarity percentages and experimentation of texture characteristic extraction using Lacunarity Method yielded images corresponding to a percentage of suitability of 25%, followed by a combined test of both methods and the normalization process produces images corresponding to a percentage of conformity of 60%.&lt;/em&gt;&lt;/p&gt;

CBIR Penyakit Kulit dengan Metode Co-Occurrence Matrix dan Color Moments

Content Based Image Retrieval (CBIR) is a technique for searching images from database based on information from the image which developed because the technique based on text-based is less effective for represent an image. CBIR skin disease in this research use 12 sample of skin disease images such as Acne, Acropustulosis, Alopecia, Dermatitis, Hemangioma, Herpes, Ichtyosis, Molluscum, Nummular, Skin Tag, Urticaria, and Vitiligo. Method use for this research is for extraction texture feature and color feature from a skin disease image. Texture feature is using co-occurrence Matrix which compute energy, contrast, entropy, homogeneity, and correlation until vector texture result. Extraction color use color moments to compute color space using three moments which result color feature from color distributions such as mean, standard deviation, and skewness. Final result showed the comparison of similarity computation of two methods is the acuration of Color Moments method is more robust than Co-occurrence Matrix Method for skin disease images.

Skin image retrieval using Gabor wavelet texture feature.

Skin imaging plays a key role in many clinical studies. We have used many skin imaging techniques, including the recently developed capacitive contact skin imaging based on fingerprint sensors. The aim of this study was to develop an effective skin image retrieval technique using Gabor wavelet transform, which can be used on different types of skin images, but with a special focus on skin capacitive contact images. Content-based image retrieval (CBIR) is a useful technology to retrieve stored images from database by supplying query images. In a typical CBIR, images are retrieved based on colour, shape, texture, etc. In this study, texture feature is used for retrieving skin images, and Gabor wavelet transform is used for texture feature description and extraction. The results show that the Gabor wavelet texture features can work efficiently on different types of skin images. Although Gabor wavelet transform is slower compared with other image retrieval techniques, such as principal component analysis (PCA) and grey-level co-occurrence matrix (GLCM), Gabor wavelet transform is the best for retrieving skin capacitive contact images and facial images with different orientations. Gabor wavelet transform can also work well on facial images with different expressions and skin cancer/disease images. We have developed an effective skin image retrieval method based on Gabor wavelet transform, that it is useful for retrieving different types of images, namely digital colour face images, digital colour skin cancer and skin disease images, and particularly greyscale skin capacitive contact images. Gabor wavelet transform can also be potentially useful for face recognition (with different orientation and expressions) and skin cancer/disease diagnosis.

Using optical coherence tomography for the longitudinal non‐invasive evaluation of epidermal thickness in a murine model of chronic skin inflammation

Non-invasive methods are desirable for longitudinal studies examining drug efficacy and disease resolution defined as decreases in epidermal thickness in mouse models of psoriasiform skin disease. This would eliminate the need for either sacrificing animals or collecting serial skin biopsies to evaluate changes in disease progression during an individual study. The quantitation of epidermal thickness using optical coherence tomography (OCT) provides an alternative to traditional histology techniques. Using the KC-Tie2 doxycycline-repressible psoriasiform skin disease mouse model, OCT imaging was completed on diseased back skin of adult KC-Tie2 (n = 3-4) and control (n = 3-4) mice, followed immediately by the surgical excision of the same region for histologic analyses. Animals were then treated with doxycycline to suppress transgene expression and to reverse the skin disease and additional OCT images and tissues were collected 2 and 4 weeks following. Epidermal thickness was measured using OCT and histology. Optical coherence tomography and histology both demonstrated that KC-Tie2 mice had significantly thicker epidermis (~4-fold; P < 0.0001) than control animals. By 2 weeks following gene repression, decreases in epidermal thickness were observed using both OCT and histology, and were sustained through 4 weeks. Correlation analyses between histology and OCT values at all time points and in all animals revealed high significance (R(2) = 0.78); with correlation being highest in KC-Tie2 mice (R(2) = 0.92) compared to control animals (R(2) = 0.16). Non-invasive OCT imaging provided similar values as those collected using standard histologic measures in thick skin of KC-Tie2 mice but became less reliable in thinner control mouse skin, possibly reflecting limitations in resolution of OCT. Future advances in resolution of OCT may improve and allow greater accuracy of epidermal thickness measurements.

Optical coherence tomography of the human skin

Background: Optical coherence tomography (OCT) is a new diagnostic method for tissue characterization. Objective: We investigated normal and pathologic structures in human skin in several locations to evaluate the potential application of this technique to dermatology. Methods: Based on the principle of low-coherence interferometry, cross-sectional images of the human skin can be obtained in vivo with a high spatial resolution of about 15 μm. Within a penetration depth of 0.5 to 1.5 mm, structures of the stratum corneum, the living epidermis, and the papillary dermis can be distinguished. Results: Different layers could be detected that were differentiated by induction of experimental blisters and by comparison with corresponding histologic sections. Furthermore, OCT images of several skin diseases and tumors were obtained. Conclusion: OCT is a promising new imaging method for visualization of morphologic changes of superficial layers of the human skin. It may be useful for noninvasive diagnosis of bullous skin diseases, skin tumors, and in vivo investigation of pharmacologic effects. (J Am Acad Dermatol 1997;37:958-63.)