An efficient dermatological lesion diagnosis method based on U-shape architecture

Scabies has been called ‘at once the easiest and most difficult diagnosis in dermatology’. In its quintessential form, it is not hard to recognize. Patients and their family members present with severe itch, often worse at night, with burrows predominantly on the hands, wrists, and often genitals. There is no other skin disease quite like it in this regard. Dermatologist and nondermatologist alike, as well as medical student or other medical professional, can often quickly recognize scabies. Even the lay public can diagnose scabies when the presentation is classic. However, when all the clues are not in place or there is contradictory or misleading information, the picture can be much more confusing. This chapter describes the classical features of scabies and the method of diagnosis. The chapter also covers a variety of complex scabies scenarios including scabies in institutional settings (such as nursing homes) as well as scabies in immunocompromised individuals.

Melanoma arises from cancerous growth in pigmented skin lesion and t is the most deadliest form of skin cancer. Melanoma forms 4% from all skin cancer cases and it accounts for 75% of all skin cancer deaths. Even when the expert dermatologists uses the dermoscopy for diagnosis, the accuracy of melanoma diagnosis is estimated to be about 75–84%. The aim of this work classify skin lesions like normally, atypical and melanoma using artificial intelligence techniques and help to decide of the expert dermatologists in diagnosis for melanoma. Decision support system, which will be held improve both the speed and the accuracy of diagnosis. In this study that done for the classification of skin lesions with ANN, were correctly classified 100% normal skin lesions according to data from the data set PH2. Abnormal and melanoma skin cancers are correctly classified %93.3 with neural network that performed. As a result, the findings that were obtained have indicated the decision support system will be help to the dermatologists in the diagnosis of skin lesions.

In addition to dermoscopy, there are other imaging and biophysical methods for the noninvasive diagnosis of skin lesions. Confocal laser microscopy allows for high-resolution imaging of the epidermis and upper dermis. It is particularly suitable in the differential diagnosis of melanocytic lesions. Optical coherence tomography (OCT) has a lower resolution compared to confocal laser microscopy but a greater depth of penetration. It is primarily used for imaging epithelial skin cancer, especially in the context of monitoring the effectiveness of nonsurgical therapies. Electrical impedance spectroscopy does not yield cutaneous images but rather provides a score based on the cellular irregularity of the skin. Multispectral analysis involves illumination of the skin with different wavelengths and likewise results in the computation of a score. Both methods are used in the differentiation of dysplastic nevi from melanoma. Other diagnostic imaging and biophysical methods are currently still in the developmental stages. By increasing the sensitivity and specificity of clinical and dermoscopic findings, the aforementioned methods bring about an improvement in noninvasive diagnosis. They allow for skin lesions to be monitored over time and therapeutic effects to be quantified. Finally, they facilitate early diagnosis of skin cancer, and help avoid unnecessary surgery of benign lesions.

An efficient multi-class dermatological lesion diagnosis using adaptive hybrid segmentation and residual graph CNN with attention mechanism

Diagnosis in dermatology is largely based on contextual factors going far beyond the visual and dermoscopic inspection of a lesion. Diagnostic tools such as the different types of dermoscopy, confocal microscopy and optical coherence tomography (OCT) are available and all of these have shown their importance in improving the dermatologist's ability, especially in the diagnosis of skin cancer. Their use, however, remains limited and time consuming, and optimizing their practice appears to be difficult, requiring extensive pre-processing, lesion segmentation and extraction of domain-specific visual features before classification. Over the last two decades, image recognition has been a matter of interest in a large part of our society and in industry, leading to the development of several techniques such as convolutional processing combined with artificial intelligence or neural networks (CNN/ANN). The aim of the present manuscript is to provide a short overview of the most recent data about CNN in the field of dermatology, mainly in skin cancer detection and its diagnosis.

Raman spectroscopy has strong potential for providing noninvasive dermatological diagnosis of skin cancer. In this study, confocal Raman microscopy was applied to the dermatological diagnosis for one of the most common skin cancers, basal cell carcinoma (BCC). BCC tissues were obtained from 10 BCC patients using a routine biopsy and used for confocal Raman measurements. Autofluorescence signals from tissues, which interfere with the Raman signals, were greatly reduced using a confocal slit adjustment. Distinct Raman band differences between normal and BCC tissues for the amide I mode and the PO2- symmetric stretching mode showed that this technique has strong potential for use as a dermatological diagnostic tool without the need for statistical treatment of spectral data. It was also possible to precisely differentiate BCC tissue from surrounding noncancerous tissue using the confocal Raman depth profiling technique. We propose that confocal Raman microscopy provides a novel method for dermatological diagnosis since direct observations of spectral differences between normal and BCC tissues are possible.

The efficacy of light therapeutic and diagnostic applications can be enhanced by employing optical tissue clearing (OTC) agents to minimize light scattering in tissue. This study aimed to investigate the optimal concentration of glycerol, so that it can be efficiently used as an OTC agent in dermatology. Glycerol was topically applied to avoid the possibility of edema that could be caused by dermal injection. The efficacy of glycerol was quantitatively evaluated for various concentrations using optical coherence tomography (OCT) to evaluate light scattering and ultrasound imaging modality to evaluate collagen dissociation. The intensity in the OCT images in the deeper regions increased over time after glycerol application owing to enhanced light penetration caused by glycerol permeation into the sample. A comparable decrease over time in the collagen distribution was observed in the ultrasound images after glycerol application. The optimal concentration of glycerol to maximize OTC was found to be 70%. The finding of this study may provide a guideline regarding the use of glycerol for efficient light diagnosis and therapy in dermatology.

Diagnosis of benign and malign skin lesions is currently mostly relying on visual assessment and frequent biopsies performed by dermatologists. As the timely and correct diagnosis of these skin lesions is one of the most important factors in the therapeutic outcome, leveraging new technologies to assist the dermatologist seems natural. Complicating matters is a blood cancer called Cutaneous T-Cell Lymphoma, which also exhibits symptoms as skin lesions. We propose a framework using optical spectroscopy and a multi-spectral classification scheme using support vector machines to assist dermatologists in diagnosis of normal, benign and malign skin lesions. As a first step we show successful classification (94.9%) of skin lesions from regular skin in 48 patients based on 436 measurements. This forms the basis for future automated classification of different skin lesions in diseased patients.

There are at least 100 entities that present a significative conjunction between skin signs and skeletal or other bony changes that can be detected by radiological means (skin and bones and others). The aim of this article is to underline that sometimes a radiological examination could guide to a correct diagnosis in dermatology. In addition, bloody diagnostic procedures could be avoided in ambiguous skeletal lesions if the radiologist remembers that skin and bone diseases exist. We review the literature and suggest a practical classification, including skin markers of occult radiological changes, skin diseases often associated with skeletal or other bony changes that require a radiological evaluation for a diagnosis, radiological changes mimicking cutaneous diseases, skeletal side effects of dermatological therapies (topical and systemic) and skin lesions occurring as a consequence of bone injury.

Optical coherence tomography (OCT) has become established in routine diagnosis in dermatology only in the last few years. The reason is that the skin is achallenge for OCT as astrong scattering medium, because only avery small proportion of photons is reflected and can be used for imaging. In addition, in most cases a visual assessment or a biopsy is sufficient. Nevertheless, the main field of application in dermatology is the diagnostics of epithelial skin tumors. The OCT is suitable for the early recognition of small, clinically and light microscopically unspecific basal cell carcinomas as well as for the differential diagnosis of other tumors and precancerous lesions. Using OCT, the preoperative measurement of tumor spread, observation of the course and treatment control of non-surgical procedures are possible; therefore, in many cases a biopsy or treatment control can be avoided. Dynamic OCT is anewly developed add on technique to visualize and quantify the superficial blood vessels of the skin. First studies are focused on the evaluation of tumor vessels, wound healing and monitoring of laser therapy. In ophthalmology, OCT diagnostics of basal cell carcinomas on the eyelids as well as for planning and control of eyelid interventions can be of interest.

Background: Ultrasound and colour Doppler are adjuvant techniques widely used in clinical settings in obstetrics, cardiology, and others. Its use in dermatology is more incipient although it presents potential for clinical use namely in dermo-oncology. Objective: This study explores the usefulness of the combination of cutaneous ultrasound with Doppler after digital dermatoscopy in distinguishing between most common benign and malignant skin lesions, focusing on the importance of different vascular patterns. To streamline the diagnostic process, we propose a combined imaging workflow that integrates dermoscopic findings with vascular and structural data obtained via Doppler ultrasound. Methods: In total, 42 benign and malignant skin tumours were analysed in a population of 42 patients using a Fotofinder digital dermatoscopy device and a GE ultrasound machine with a high-frequency probe (20 MHz). Doppler was applied to assess lesion vascularization and identify distinct blood flow patterns. Results: Cutaneous ultrasound revealed that malignant lesions often exhibited intense and disorganized vascularization, while benign lesions displayed more ordered and peripheral blood flow patterns. In all of our cases, ultrasound with Doppler imaging clarified the uncertainties raised by dermatoscopy. Conclusions: The use of Doppler cutaneous ultrasound after digital dermatoscopy proved to be a valuable tool to aid the diagnosis in dermatology, as it improved the differential diagnosis between benign and malignant lesions, contributing to the establishment of the final diagnosis in the studied cases.

Artificial intelligence (AI) is increasingly present in dermatology, demonstrating accuracy levels comparable to, or even superior to, those of dermatologists in diagnosing skin lesions from clinical and dermoscopic images. This review provides an overview of AI's role in the automated classification and monitoring of skin lesions. To explore and map the existing literature on the use and benefits of AI in dermatology. This narrative review focused on exploring the use and benefits of AI in dermatology, utilizing MeSH/DeCS terms such as "Dermatology," "Artificial Intelligence," "Diagnosis," and "Computer Aided Diagnosis." Three databases (PubMed, Lillacs, and Scopus) from 2008 to 2024. We excluded articles that did not focus on dermatology, lacked the topic of Artificial Intelligence, or presented theoretical designs without practical application or evidence, resulting in a final selection of forty-four articles. The results strongly support AI's effectiveness, displaying its precision in diagnosis, comparable to or exceeding that of human dermatologists across diverse tasks. The evolution of AI in dermatology implies a substantial transformation in care, extending its applications from skin cancer to various dermatological pathologies. While emphasizing the vital collaboration between AI and healthcare professionals, a critical gap remains in the real-world clinical validation of AI. Ethical considerations, especially in automated decision-making, need careful attention. This narrative review highlights the crucial role of AI in dermatology, emphasizing its potential to enhance diagnostic accuracy for skin lesions.

Differential diagnosis of round or discoid lesions

Pediatric dermatology is a new topic and no epidemiologic data exist from Switzerland. Therefore we performed a survey of the pediatric population referred to the hospital of Aarau, Switzerland, between 1998 and 2001. All inpatients and outpatients less than 16 years old with a dermatologic diagnosis were included prospectively in our study. Demographic data (age, mean age, sex distribution), referral method, pattern and frequency of the different diagnoses in various age groups, diagnostic pattern, and therapy were analyzed. A total of 1105 children were included, with a slightly higher proportion of girls (53.8% versus 46.2%). The average age was 6.8 years and infants and school children represented 60% of the study population. Half of the patients (51%) were external referrals, almost one-third (29%) presented spontaneously, and the remaining 20% were sent from other hospital departments. With a frequency of 25.9%, atopic dermatitis was the most frequent diagnosis, followed by pigmented nevi (9.1%) and warts (5.0%). Local therapy was prescribed in 66% of patients and systemic therapy in 18.6%. Other treatments such as curettage, surgery, cryotherapy, ultraviolet therapy, and electrotherapy were rarely performed (2%). We found that atopic dermatitis was the most frequent skin disorder seen in all age groups. As this was a dermatologic subspecialty clinic, higher frequencies of chronic and uncommon dermatoses such as genetic and autoimmune diseases were seen, whereas frequent diagnoses such as diaper rash and miliaria were rarely seen and the frequencies of other common skin disorders such as scabies, pediculosis, impetigo contagiosa, warts, and molluscum contagiosum were expected to be higher compared with the figures in the literature. In our study these dermatoses are underreported, as most patients are treated by general practitioners and pediatricians. Our survey documents the most common skin diseases in childhood primarily seen by pediatricians. We emphasize that dermatologic education of medical students, primary care physicians, and pediatricians should focus on allergic skin diseases, skin infections, pigmentary disorders, and vascular lesions.

The use of Computer-Aided Diagnosis in dermatology raises the necessity of integrating Content-Based Image Retrieval (CBIR) technologies. The latter could be helpful to untrained users as a decision support system for skin lesion diagnosis. However, classical CBIR systems perform poorly due to semantic gap. To alleviate this problem, we propose in this paper an intelligent Content-Based Dermoscopic Image Retrieval (CBDIR) system with Relevance Feedback (RF) for melanoma diagnosis that exhibits: efficient and accurate image retrieval as well as visual features extraction that is independent of any specific diagnostic method. After submitting a query image, the proposed system uses linear kernel-based active SVM, combined with histogram intersection-based similarity measure to retrieve the K most similar skin lesion images. The dominant (melanoma, benign) class in this set will be identified as the image query diagnosis. Extensive experiments conducted on our system using a 1097 image database show that the proposed scheme is more effective than CBDIR without the assistance of RF.