Deep Learning Based Automated Diagnosis of Skin Diseases Using Dermoscopy

Biomedical image analysis has been exploited considerably by recent technology involvements,carrying about a pattern shift towards‘automation’and‘error free diagnosis’classification methods with markedly improved accurate diagnosis productivity and cost effectiveness.This paper proposes an automated deep learning model to diagnose skin disease at an early stage by using Dermoscopy images.The proposed model has four convolutional layers,two maxpool layers,one fully connected layer and three dense layers.All the convolutional layers are using the kernel size of 3∗3 whereas the maxpool layer is using the kernel size of 2∗2.The dermoscopy images are taken from the HAM10000 dataset.The proposed model is compared with the three different models of ResNet that are ResNet18,ResNet50 and ResNet101.The models are simulated with 32 batch size and Adadelta optimizer.The proposed model has obtained the best accuracy value of 0.96 whereas the ResNet101 model has obtained 0.90,the ResNet50 has obtained 0.89 and the ResNet18 model has obtained value as 0.86.Therefore,features obtained from the proposed model are more capable for improving the classification performance of multiple skin disease classes.This model can be used for early diagnosis of skin disease and can also act as a second opinion tool for dermatologists.

Hybrid Segmentation Scheme for Skin Features Extraction Using Dermoscopy Images

Objective and quantitative assessment of skin conditions is essential for cosmeceutical studies and research on skin aging and skin regeneration.Various handcraft-based image processing methods have been proposed to evaluate skin conditions objectively,but they have unavoidable disadvantages when used to analyze skin features accurately.This study proposes a hybrid segmentation scheme consisting of Deeplab v3+with an Inception-ResNet-v2 backbone,LightGBM,and morphological processing(MP)to overcome the shortcomings of handcraft-based approaches.First,we apply Deeplab v3+with an Inception-ResNet-v2 backbone for pixel segmentation of skin wrinkles and cells.Then,LightGBM and MP are used to enhance the pixel segmentation quality.Finally,we determine several skin features based on the results of wrinkle and cell segmentation.Our proposed segmentation scheme achieved a mean accuracy of 0.854,mean of intersection over union of 0.749,and mean boundary F1 score of 0.852,which achieved 1.1%,6.7%,and 14.8%improvement over the panoptic-based semantic segmentation method,respectively.

Dermoscopy in China: current status and future prospective

Objective: Dermoscopy is a useful technique for improving the diagnostic accuracy of various types of skin disorders. In China, dermoscopy has been widely accepted, and domestic researchers have made tremendous progress in the field of dermoscopy. The main purpose of this review is to summarize the current status of dermoscopy in China and identify its future directions. Data sources: Articles included in this review were obtained by searching the following databases: Wanfang, China National Knowledge Infrastructure, PubMed, and the Web of Science. We focused on research published before 2019 with keywords including dermoscopy, dermoscopic, dermoscope and trichoscopy. Study selection: A total of 50 studies were selected. Of these studies, 20 studies were in Chinese and 30 in English, research samples of all the studies were collected from Chinese populations. Results: Since 2000, more than 380 articles about dermoscopy have been published in domestic or foreign journals. Dermoscopy can improve the diagnostic accuracy of neoplastic diseases, evaluating the therapeutic effect of treatment, and determining the treatment endpoint, and it can also assist in the differential diagnosis of inflammatory diseases and in the assessment of the severity of the disease. In addition, researches about the applications of dermoscopy during surgical treatment have been published. Training courses aiming to improve the diagnostic ability of dermatologists, either face-to-face or online, have been offered. The Chinese Skin Image Database, launched in 2017 as a work platform for dermatologists, has promoted the development of dermoscopy in China. Computer-aided diagnostic systems based on the Chinese population are ready for use. In the future, cooperation, resource sharing, talent development, image management, and computer-aided diagnosis will be important directions for the development of dermoscopy in China. Conclusion: Dermoscopy has been widely used and developed in China, however, it still needs to address more challenges in the future.

Adaptive segmentation based on multi-classification model for dermoscopy images

Segmentation accuracy of dermoscopy images is important in the computer-aided diagnosis of skin cancer and a wide variety of segmentation methods for dermoscopy images have been developed. Considering that each method has its strengths and weaknesses, a novel adaptive segmentation framework based on multi-classification model is proposed for dermoscopy images. Firstly, five patterns of images are summarized according to the factors influencing segmentation. Then the matching relation is established between each image pattern and its optimal segmentationmethod. Next, the given image is classified into one of the five patterns by the multi-classification model based on BP neural network. Finaily, the optimal segmentation method for this image is selected according to the matching relation, and then the image is effectively segmented. Experiments show that the proposed method delivers better accuracy and more robust segmentation results compared with the other seven state-of-the-art methods.

Nail Psoriasis in a Child Observed Under Ultraviolet Dermoscopy Treated by a Topical Biological Agent Cream:A Case Report

Introduction:Nail psoriasis is a type of psoriasis involving nail lesions characterized by pitting,onycholysis,longitudinal ridges,and subungual hyperkeratosis.We herein describe a 9-year-old girl with nail psoriasis who presented with nail crumbling and was treated with topical cream containing 45μg/g mouse monoclonal antibody to human interleukin-8.Case presentation:A 9-year-old Chinese girl presented with a 6-month history of a rough,thickened fingernail and toenails.Nail plate crumbling,onycholysis,and fissured periungual folds were observed under dermoscopy and ultraviolet dermoscopy.The nails were soaked in warm water,then topical wrapped with Abcream cream overnight.After about 4 months of treatment,the nails significantly improved by both dermoscopy and ultraviolet-dermoscopy evaluattion.Discussion:Due to the different wavelengths of light emitted by polarized light dermatoscope and ultraviolet-dermatoscope,the characteristics of observation will be different.Abcream acts by antagonizing human interleukin-8,inhibiting leukocyte chemotaxis and neovascularization,and regulating the abnormal proliferation and differentiation of keratinocytes.Conclusion:Ultraviolet-dermoscopy is pivotal in evaluating the severity and potency of nail psoriasis.And Abcream can be regarded as a new drug for the treatment of nail psoriasis in children.

Basal Cell Carcinoma Excision Guided by Dermoscopy:A Retrospective Study in Macao

Objective:Dermoscopic evaluation of tumor margins may help to accurately detect lateral borders before surgical excision.The purpose of this study was to comprehensively evaluate the usefulness of dermoscopically detecting basal cell carcinoma tumor margins before surgical excision.Methods:The authors retrospectively analyzed the outcomes of 60 basal cell carcinomas that were excised after undergoing dermoscopic evaluation of the margins from 2016 to 2018 in a single center in Macao SAR,China.Descriptive statistical analysis was carried out by using frequencies and percentages.Results:All treated tumors were completely excised,although five had safety margins of<1 mm.No re-excisions were performed and during a medium follow-up period of 31 months,none of the 60 tumors showed any evidence of recurrence.Conclusions:Our data suggest that the preoperative dermoscopic evaluation of tumor margins increases the chances of successful and complete tumor excision,while preserving healthy adjacent tissue.Further studies comparing dermoscopy-assisted versus conventional excision are needed to better evaluate the value of this technique.

Scabies Evaluated by Dermoscopy and Fluorescence Microscopy: A Case Report

Introduction: Scabies is an infectious skin disorder caused by the mite Sarcoptes scabiei. The diagnosis of scabies is often confirmed by microscopic detection of scabies mites, eggs, or feces.We herein describe the diagnosis of scabies in a boy with the assistance of dermoscopy and fluorescence microscopy.Case presentation: A 16-year-old Chinese boy presented with a 2-month history of extensive papules and excoriations with intense pruritus on the trunk and limbs. Dermoscopy showed a sinuous burrow in the finger webs with a brown jet-shaped triangular structure at the end. A fluorescence microscopy revealed a hatching egg and a mite. The boy was diagnosed with scabies and treated with 5% sulfur ointment.Discussion: The use of dermoscopy in patients with scabies reveals a sinuous burrow with a brown jet-shaped triangular structure composed of the pigmented head and anterior legs of the mite[1]. A fluorescence microscopy revealed mites and eggs with blue fluorescence after staining, allowing us to easily identify scabies.Conlcusion: The combination of dermoscopy-guided tape testing with fluorescence staining technology may enhance the diagnostic accuracy and efficiency of scabies.

Dermoscopy Features of Trichofolliculoma: A Case Report

Introduction:Trichofolliculoma characterized clinically by the presence of acentral dilated pore with tufted hairs and corresponds histologically to a central primary follicle and many secondary vellus hair follicles originating from it.Thus far,dermoscopy dascription of trichofolliculoma has been lacking.Here,we describe a typical case of trichofolliculoma,particularly the special manifestation under the dermoscopy.Case presentation:A 29-year-old man complained a five-year history of a nodule containing a central whitish hair plug on his left tempus.Dermoscopic examination revealed a well-defined,yellow macule with a central whitish hair plug surrounded by dilated capillaries,and histopathological analysis was consistent with trichofolliculoma.Discussion:The findings of dermoscopy were also vary,which may be affected by the origin of the disease,various phases of the hair cycle,depth of the lesion,and even external stimulus.Conclusion:Diagnosis of trichofolliculoma is sometimes difficult because the presence of the central crater and visible hairs is low,and dermoscopy examination may be a potentially useful diagnostic tool.

Multi-Class Skin Cancer Detection Using Fusion of Textural Features Based CAD Tool

Skin cancer has been recognized as one of the most lethal and complex types of cancer for over a decade.The diagnosis of skin cancer is of paramount importance,yet the process is intricate and challenging.The analysis and modeling of human skin pose significant difficulties due to its asymmetrical nature,the visibility of dense hair,and the presence of various substitute characteristics.The texture of the epidermis is notably different from that of normal skin,and these differences are often evident in cases of unhealthy skin.As a consequence,the development of an effective method for monitoring skin cancer has seen little progress.Moreover,the task of diagnosing skin cancer from dermoscopic images is particularly challenging.It is crucial to diagnose skin cancer at an early stage,despite the high cost associated with the procedure,as it is an expensive process.Unfortunately,the advancement of diagnostic techniques for skin cancer has been limited.To address this issue,there is a need for a more accurate and efficient method for identifying and categorizing skin cancer cases.This involves the evaluation of specific characteristics to distinguish between benign and malignant skin cancer occurrences.We present and evaluate several techniques for segmentation,categorized into three main types:thresholding,edge-based,and region-based.These techniques are applied to a dataset of 200 benign and melanoma lesions from the Hospital Pedro Hispano(PH2)collection.The evaluation is based on twelve distinct metrics,which are designed to measure various types of errors with particular clinical significance.Additionally,we assess the effectiveness of these techniques independently for three different types of lesions:melanocytic nevi,atypical nevi,and melanomas.The first technique is capable of classifying lesions into two categories:atypical nevi and melanoma,achieving the highest accuracy score of 90.00%with the Otsu(3-level)method.The second technique also classifies lesions into two categories:common nevi and melanoma,achieving a score of 90.80%with the Binarized Sauvola method.

Intelligent Multiclass Skin Cancer Detection Using Convolution Neural Networks

The worldwide mortality rate due to cancer is second only to cardiovascular diseases.The discovery of image processing,latest artificial intelligence techniques,and upcoming algorithms can be used to effectively diagnose and prognose cancer faster and reduce the mortality rate.Efficiently applying these latest techniques has increased the survival chances during recent years.The research community is making significant continuous progress in developing automated tools to assist dermatologists in decision making.The datasets used for the experimentation and analysis are ISBI 2016,ISBI 2017,and HAM 10000.In this work pertained models are used to extract the efficient feature.The pertained models applied are ResNet,InceptionV3,and classical feature extraction techniques.Before that,efficient preprocessing is conducted on dermoscopic images by applying various data augmentation techniques.Further,for classification,convolution neural networks were implemented.To classify dermoscopic images on HAM 1000 Dataset,the maximum attained accuracy is 89.30%for the proposed technique.The other parameters for measuring the performance attained 87.34%(Sen),86.33%(Pre),88.44%(F1-S),and 11.30%false-negative rate(FNR).The class with the highest TP rate is 97.6%for Melanoma;whereas,the lowest TP rate was for the Dermatofibroma class.For dataset ISBI2016,the accuracy achieved is 97.0%with the proposed classifier,whereas the other parameters for validation are 96.12%(Sen),97.01%(Pre),96.3%(F1-S),and further 3.7%(FNR).For the experiment with the ISBI2017 dataset,Sen,Pre,F1-S,and FNR were 93.9%,94.9%,93.9%,and 5.2%,respectively.

Bowen’s disease on the palm: A case report

BACKGROUND Bowen’s disease (BD) is a persistent, progressive intraepidermal carcinoma. BD usually occurs in areas exposed to sunlight. Involvement of the dorsum of the hand is not rare, but that of the palmar aspect is very unusual. Only a few cases have been reported in the literature. CASE SUMMARY Here, we report the case of a 48-year-old male patient who presented with a history of persistent local erythema lasting for 2 years on the thenar eminence of the left palm. Initially diagnosed as hand eczema, the condition did not improve with intermittent treatment with anti-allergy medications or topical glucocorticoid ointments, among other approaches. Then, the area of erythema gradually enlarged and was accompanied by mild itching. For a definite diagnosis and treatment, the patient came to our hospital. Dermoscopic examination revealed BD, and histopathological examination confirmed the diagnosis. We performed partial resection of the skin lesion followed by photodynamic therapy. No recurrence was observed at the 6-mo follow-up. CONCLUSION For all atypical palmar lesions, early dermoscopy and/or skin biopsy are needed to avoid missed diagnosis or misdiagnosis.

Dermoscopic Appearances in the Superficial and Deep Type of Infantile Hemangioma

Infantile hemangioma is clinically classified as superficial, superficial and deep (mixed), and deep. Dermoscopy can be used for analyzing vascular structure and for classifying infantile hemangioma. Here, we focus on vascular features of the superficial and deep type of infantile hemangioma and show the three characteristic dermoscopic forms, mostly reddish, diffuse reddish, and cerebriform reddish.

Lesion presenting with a “blue amber” pattern

Atypical fibroxanthoma(AFX) is a spindle cell neoplasm with low metastatic potential but high tendency to recur after surgery. Because of the rarity of this lesion and its aspecific clinical features, AFX could be easily misdiagnosed and undertreated by many clinicians who encounter them. Dermoscopy represents a valuable tool for easily assessing skin lesions, even though histological examination is required for final diagnosis. We report a case of a cheek lesion with dermoscopic "blue amber pattern", easily recognisable and not observed in others skin tumours, which could represent an additional feature useful in differentiating this tumour from other skin neoplasms.

Mystery behind labial and oral melanotic macules: Clinical, dermoscopic and pathological aspects of Laugier-Hunziker syndrome

Labial and oral melanotic macules are commonly encountered in a broad range of conditions ranging from physiologic pigmentation to a sign of an underlying life-threatening disease. Although Laugier-Hunziker syndrome(LHS) shares some features of labial and oral pigmentation with a variety of conditions, it is a benign and acquired condition, frequently associated with longitudinal melanonychia. Herein, the demographic, clinical, dermoscopic, and pathological aspects of LHS were reviewed comprehensively. The important differential diagnoses of mucocutaneous and nail pigmentation are provided. An accurate diagnosis is crucial to design a reasonable medical strategy, including management options, malignant transformation surveillance, and psychological support. It is important that clinicians conduct long-term follow-up and surveillance due to the potential risks of malignant transformation and local severe complications in some conditions.

Arithmetic Optimization with Ensemble Deep Transfer Learning Based Melanoma Classification

Melanoma is a skin disease with high mortality rate while earlydiagnoses of the disease can increase the survival chances of patients. Itis challenging to automatically diagnose melanoma from dermoscopic skinsamples. Computer-Aided Diagnostic (CAD) tool saves time and effort indiagnosing melanoma compared to existing medical approaches. In this background,there is a need exists to design an automated classification modelfor melanoma that can utilize deep and rich feature datasets of an imagefor disease classification. The current study develops an Intelligent ArithmeticOptimization with Ensemble Deep Transfer Learning Based MelanomaClassification (IAOEDTT-MC) model. The proposed IAOEDTT-MC modelfocuses on identification and classification of melanoma from dermoscopicimages. To accomplish this, IAOEDTT-MC model applies image preprocessingat the initial stage in which Gabor Filtering (GF) technique is utilized.In addition, U-Net segmentation approach is employed to segment the lesionregions in dermoscopic images. Besides, an ensemble of DL models includingResNet50 and ElasticNet models is applied in this study. Moreover, AOalgorithm with Gated Recurrent Unit (GRU) method is utilized for identificationand classification of melanoma. The proposed IAOEDTT-MC methodwas experimentally validated with the help of benchmark datasets and theproposed model attained maximum accuracy of 92.09% on ISIC 2017 dataset.

Malignant form of hidroacanthoma simplex:A case report

BACKGROUND This paper presents a case of malignant hidroacanthoma simplex(HAS)and review the literature of previous cases to summarize the histopathological and immunohistochemical features and display the dermoscopic features of malignant HAS.CASE SUMMARY We present an 88-year-old Asian female with malignant HAS.The diagnosis was made according to the histopathological and immunohistochemical results after biopsy.Previous case reports of malignant HAS were retrieved from PubMed to characterize the histopathological and immunohistochemical features.We also display the dermoscopic features of malignant HAS that have not been reported.CONCLUSION Our findings demonstrate that prompt surgical treatment is an effective strategy for malignant HAS.Histopathology and immunohistochemistry are valuable diagnostic tools.This is the first case report to display the dermoscopic features of malignant HAS,and we speculate that dermoscopy may contribute to the diagnosis of malignant HAS.

Long term clinical and dermoscopic follow-up of a child with a Spitz nevus

Background: Spitz nevus is uncommon, benign melanocytic neoplasm that may show some clinical, dermoscopical and histological features of melanoma. It occurs often in childhood, but may appear also in early adulthood. Rare congenital cases have been reported in literature. It is frequently located on the face and the lower extremities, but in some cases may appears on the trunk. Methods: We report a case of a 9-years-old girl presented to our Dermatology Unit because of the presence of a pigmented lesion on her right leg, 4 mm in diameter, which was clinically and dermoscopically diagnosed as Spitz nevus. We described the clinical and dermoscopic features that we observed every six months, over 11 years follow-up period. Objective: Our observation show that the globular, the starburst, the homogeneous patterns and diffuse brown colour with diffuse hypopigmented areas may be the different expression that correspond to possible evolutionary phases of pigmented Spitz nevus.

Dermoscopic Features of Hyperpigmented Dots in Crista Cutis in Two Siblings in a Japanese Family with Inherited Acanthosis Nigricans

Acanthosis nigricans is characterized by papillomatous brownish lesions mainly in the intertriginous areas. We used dermoscopy to examine such lesions in a family with acanthosis nigricans. The dermoscopic images showed an aberrant skin structure of linear crista cutis and sulcus cutis, and hyperpigmented dots in crista cutis. The hyperpigmented dots, which could not be seen with the naked eyes, may contribute to the color of the pigmented skin. Dermoscopy can be useful for evaluating disorders involving the structure of area cutanea and a change in skin color.

Trichoscopy: Essentials for the dermatologist

Noninvasive in vivo imaging techniques have become an important diagnostic aid for dermatology. Dermoscopy, also known as dermatoscopy, has been shown to increase the clinician's diagnostic accuracy when evaluating cutaneous neoplasms. Dermoscope, both hand-held and videodermoscope, are nowadays a basic instrument for almost all the dermatologists around the world. Trichoscopy is the term coined for dermoscopic imaging of the scalp and hair. Routinely using dermoscopy and recognizing the structures and patterns of the different types of alopecia will likely improve the observer's sensitivity for diagnosis and followup of hair and scalp disorders. Structures which may be visualized by trichoscopy include hair shafts of different types, the number of hairs in one pilosebaceous unit, hair follicle openings(dots), the peri and interfollicular areas and the vasculature. This review summarizes the current knowledge about trichoscopic findings which may aid in the diagnosis of alopecia. Besides diagnosing alopecia, it has the potential for obviating unnecessary biopsies and when a biopsy is still needed it is helpful in choosing an ideal biopsy site. Moreover, trichoscopy can be a valuable tool for evaluating the treatment response photographically at each follow-up. Finally, we have discussed the utility of dermoscopy in inflammatory scalp disorders and infections.

AB084.Correlation between histopathology and optical coherence tomography in periocular tumors

Background:Basal cell carcinoma(BCC)accounts for 90%of the eyelid malignancies followed by sebaceous cell carcinoma(SC)and squamous cell carcinoma(SCC).Even though biopsy for histopathological(HP)evaluation remains the gold standard for confirming the diagnosis of eyelid tumors and the status of surgical margins,noninvasive techniques such as dermoscopy and optical coherence tomography(OCT)may increase diagnostic sensitivity and contribute to more accurate delineation of surgical margins.The purpose of this prospective study was to correlate the HP and OCT findings of eyelid malignancies.Methods:After ethics committee approval(BCC/2017-2608),patients with suspected eyelid carcinomas had photographs taken,tumor characteristics assessed clinically and using dermoscopy and OCT,which were subsequently compared with HP evaluation.We obtained OCT images of periorbital tumors with an anterior segment lens adjusted in an Optovue Avanti.The histopathology was evaluated using a digital pathology system.Patients under 18 years old were excluded from this study.Results:Three female and three male patients(age 69-88 years)with lesions located in the lower periorbital area(n=5)and upper eyelid(n=1)were included in this study.Two participants previously had skin cancer:1 female patient had a lentigo maligna and a BCC,and 1 male patient had multiple BCC and actinic keratosis due to immunosuppression related to a liver transplant.Clinical and dermoscopicc diagnosis were accurate in four cases:3 BCC and 1 SCC.One clinically suspicious BCC was diagnosed as SC and one patient had a previous confirmed diagnosis of SC.All OCT images showed disruption of the dermoepidermal junction and hyporeflective areas in the dermis,which were darker and well-delineated in the three confirmed BCCs.OCT also clearly demarcated the transition between tumor and normal skin in all cases.Conclusions:OCT is a fast and non-invasive exam to acquire high-resolution images of intraocular and ocular surface structures.Preliminary data from our study indicates that real-time OCT images correlate with HP findings and may also help delineate tumor margins.The inclusion of additional participants will provide further evidence of the benefits of using OCT pre-operatively to safely minimize surgical margins during eyelid tumor excision.