Dermatoscopic Images Research Articles

Skin Cancer Detection from Dermatoscopic Images Using Hybrid Fuzzy Ensemble Learning Model

Skin Cancer Detection from Dermatoscopic Images Using Hybrid Fuzzy Ensemble Learning Model

An Interpretable Skin Cancer Classification Using Optimized Convolutional Neural Network for a Smart Healthcare

Skin cancer is a prevalent and life-threatening disease worldwide, making early and accurate diagnosis crucial for effective treatment. In the era of smart healthcare, the integration of artificial intelligence and computer-aided diagnosis systems has shown promise in improving the diagnostic accuracy and interpretability of skin cancer classification. This project presents an innovative approach for the interpretable classification of skin cancer using an optimized Convolutional Neural Network (CNN). The proposed system leverages deep learning techniques to analyze dermatoscopic images, offering a non-invasive and efficient solution for early skin cancer detection. Through careful optimization of the CNN architecture, feature extraction, and training parameters, our model achieves enhanced classification performance while maintaining a high degree of interpretability. Furthermore, the system incorporates advanced visualization and explanation techniques to provide clinicians and patients with transparent insights into the decision-making process of the CNN. This interpretability is crucial for building trust in the automated diagnostic system and facilitating effective collaboration between AI and healthcare professionals. The project's results demonstrate the potential for improved skin cancer diagnosis through a transparent and optimized CNN model, contributing to the advancement of smart healthcare. The combination of AI technology and interpretability not only enhances the accuracy of skin cancer classification but also ensures that the decision-making process is comprehensible and actionable for medical practitioners, ultimately leading to better patient outcomes. Key Words: Skin cancer, Healthcare, Artificial Intelligence, Deep Learning, OpenCV, RNN

Diagnostic Accuracy of Hyperpigmented Microcircles in Dermatoscopy of Non-Facial Non-Acral Melanomas: A Pilot Retrospective Study using a Public Image Database

Background: Grey perifollicular circles are a dermatoscopic clue to melanoma, especially in facial skin. So far, no other adnexal clues than follicular have been investigated in this diagnosis. Objectives: The study aimed to analyse the prevalence of hyperpigmented periadnexal microcircles (HMs) in melanoma and its common simulators at non-facial non-acral sites, analyse the relation between the presence of HM, regression and hypopigmentation, and evaluate the diagnostic accuracy of this structure in melanoma. Methods: International Skin Imaging Collaboration (69,445 images) was searched in April 2020 for the pathology-confirmed dermatoscopic images with metadata including sex, age bin, and declared non-acral non-facial anatomic site. The final study sample (5,408 images, 1,326 of which were melanomas) was evaluated by expert dermatoscopist blinded to the diagnosis and labelled for the presence of ≥3 HM distributed centrally (cHM) or peripherally (pHM), hypopigmentation, and classic dermatoscopic regression structures. A subset of 40 images was labelled by 7 raters (2 residents, 5 experts) to assess interobserver agreement. We compared the presence of pHM with the presence of regression as well as performed a set of independent χ2 tests to evaluate the discriminatory power and its fragility. Performance of the models was assessed using measures of discrimination and calibration. Results: HM were significantly more prevalent in melanomas than in non-melanomas and nevi. Fair/good interobserver agreement for HM was reached for all the raters, and moderate/good for experts only (single rater/average, respectively). Regardless of regression/hypopigmentation status, pHM were significantly more common in melanoma than in non-melanomas or nevi and were observed significantly more often in melanomas on sun-damaged skin (upper extremity, posterior torso). No significant differences between the groups were found for cHM. pHM proved a high odds ratio in the tests as to the classical indicators such as classic dermatoscopic regression structures. Conclusion: pHM could be considered a novel dermatoscopic clue to melanoma.

Deep ensemble learning using a demographic machine learning risk stratifier for binary classification of skin lesions using dermatoscopic images

Deep ensemble learning using a demographic machine learning risk stratifier for binary classification of skin lesions using dermatoscopic images

Early Melanoma Detection Based on a Hybrid YOLOv5 and ResNet Technique.

Skin cancer, specifically melanoma, is a serious health issue that arises from the melanocytes, the cells that produce melanin, the pigment responsible for skin color. With skin cancer on the rise, the timely identification of skin lesions is crucial for effective treatment. However, the similarity between some skin lesions can result in misclassification, which is a significant problem. It is important to note that benign skin lesions are more prevalent than malignant ones, which can lead to overly cautious algorithms and incorrect results. As a solution, researchers are developing computer-assisted diagnostic tools to detect malignant tumors early. First, a new model based on the combination of "you only look once" (YOLOv5) and "ResNet50" is proposed for melanoma detection with its degree using humans against a machine with 10,000 training images (HAM10000). Second, feature maps integrate gradient change, which allows rapid inference, boosts precision, and reduces the number of hyperparameters in the model, making it smaller. Finally, the current YOLOv5 model is changed to obtain the desired outcomes by adding new classes for dermatoscopic images of typical lesions with pigmented skin. The proposed approach improves melanoma detection with a real-time speed of 0.4 MS of non-maximum suppression (NMS) per image. The performance metrics average is 99.0%, 98.6%, 98.8%, 99.5, 98.3%, and 98.7% for the precision, recall, dice similarity coefficient (DSC), accuracy, mean average precision (MAP) from 0.0 to 0.5, and MAP from 0.5 to 0.95, respectively. Compared to current melanoma detection approaches, the provided approach is more efficient in using deep features.

SBXception: A Shallower and Broader Xception Architecture for Efficient Classification of Skin Lesions.

Skin cancer is a major public health concern around the world. Skin cancer identification is critical for effective treatment and improved results. Deep learning models have shown considerable promise in assisting dermatologists in skin cancer diagnosis. This study proposes SBXception: a shallower and broader variant of the Xception network. It uses Xception as the base model for skin cancer classification and increases its performance by reducing the depth and expanding the breadth of the architecture. We used the HAM10000 dataset, which contains 10,015 dermatoscopic images of skin lesions classified into seven categories, for training and testing the proposed model. Using the HAM10000 dataset, we fine-tuned the new model and reached an accuracy of 96.97% on a holdout test set. SBXception also achieved significant performance enhancement with 54.27% fewer training parameters and reduced training time compared to the base model. Our findings show that reducing and expanding the Xception model architecture can greatly improve its performance in skin cancer categorization.

BT01 Taking pressure off the skin cancer service: impact of primary care dermoscopy pilot, medical photographer-led clinics and non-2-week-wait skin lesion advice and guidance pathway

Abstract Since the COVID-19 pandemic, we have struggled with similar pressures and capacity issues as most dermatology departments across the country. Our number of 2-week-wait (2WW) and general referrals have increased by 26%. To manage this, we have gradually introduced strategies based around a triaging system for 2WW referrals received with photographic images from the general practitioner (GP) via the National Health Service e-Referral Service (e-RS). We triage patients either to a face-to-face 2WW clinic, a spot clinic run by supervised junior members of the team or a photography/dermoscopy clinic. If we receive dermoscopic images with the GP referral, where appropriate, we can discharge or downgrade the patient to a routine clinic. Strategy 1: in September 2021, we set up a 1-year primary care 2WW dermoscopy pilot. Dermatoscopes were delivered to interested GP practices (appropriate technical training was provided), and the GPs attended skin lesion teaching sessions every 2 months. Thirteen per cent of all our 2WW referrals came through this dedicated pilot 2WW e-RS pathway. We discharged or downgraded 26% of those that were referred with both macroscopic and dermoscopic images. All GP practices (not just the pilot practices) in our region now have dermatoscopes. Strategy 2: in July 2022, we introduced secondary care photography/dermoscopy clinics run by both dermatology nurses and medical photographers using the Pathpoint/eDERMA platform. We triage patients from 2WW e-RS referrals with macroscopic images suggestive of benign lesions to these clinics. We can discharge or downgrade directly from the platform, which immediately sends patients’ information about their diagnosis and management plan via text message. Around 15% of 2WW referrals are booked to eDERMA clinics and around 60% are discharged. Strategy 3: in October 2022, we developed a non-2WW advice and guidance (A&G) e-RS skin lesion pathway for lesions of diagnostic uncertainty and basal cell carcinomas. Dermatoscopic images are not an essential requirement as this is not a 2WW pathway. Referrals are triaged within 72 h, providing rapid advice to GPs. Initial data show that since introducing this service, 2WW referral numbers decreased by 32%. Of the A&G referrals, 28% were discharged with advice and 17% were upgraded to 2WW. Within 3 months of starting this pathway, our department’s 2WW and Faster Diagnostic Standard 28-day cancer targets were no longer breaching. The combination of GP-based photography and dermoscopy, careful triage to a 2WW photography clinic in secondary care and a non-2WW lesion A&G pathway has significantly improved our 2WW service.

BT10 Skin lesion triage teledermatology service: a review of secondary care 2-week wait upgrades

Abstract A teledermatology service for skin lesions was commenced in January 2021 initially as a pilot scheme with six general practices and one dermatology department in a metropolitan area; to date, 70% of the city practices have since been on-boarded. Many 2-week wait (2WW) referrals are made not just for suspected skin cancer, but also for cases where there is diagnostic uncertainty and secondary care opinion/reassurance sought. The aim of the service was to embed a fully integrated information technology solution to facilitate single skin lesion triage in an attempt to help to address the 2WW capacity issues in secondary care. The referring general practitioners (GPs) were provided with a mobile phone adaptor attachment and dermatoscope to obtain images of the index lesion in question. A proforma that includes a concise history of the lesion and clinical impression is completed. The GPs were encouraged to use this triage service but could also refer directly to the 2WW service if concerned. The consultant then provided advice and guidance with a menu of triage outcomes, including upgrade to a 2WW appointment. We reviewed the referrals received via this skin lesion pathway and numbers of cases converted to a 2WW appointment from 12 April to 1 December 2022 as a snapshot of the service. We aimed to determine factors from the referral that prompted the conversion. There were 1201 skin lesion referrals, and 299 cases were converted to 2WW (24.9% of the cohort). In 93 of 299 referrals, the GP’s impression was possible squamous cell carcinoma or melanoma. In 137 of 299 cases, the GP had not provided a clinical impression. In 70 of 299 referrals, the GP’s clinical impression correlated with the dermatology’s impression based on the dermatoscopic images. In 44 cases, the clinician’s impression based on the image was of a benign lesion; however, we noted that a more cautious approach was adopted by some clinicians who triaged to 2WW if there was any diagnostic uncertainty. After face-to-face review, 159 of 299 were listed for biopsy or excision. Some factors identified that limited diagnostic accuracy included poor image quality and some referrals contained multiple lesions, causing confusion. This skin lesion triage service appears to support the idea that photographic triage could reduce the burden of 2WW face-to-face referrals. Areas for improvement identified include ensuring that the referrer provides a concise history and clinical impression, the use of gel for dermatoscopic images and feedback when images are not of adequate quality. In the future, it may be necessary to consider photographic hubs to standardize referral photographic quality.

P29 Selfies: not ideal!

Abstract We sought to determine whether the quality of photographic referrals from primary care had been impacted by the COVID-19 pandemic. A total of 2411 referrals made in 2020, 2021 and 2022 were retrospectively reviewed. From this, data were collected on urgent suspected cancer (USC) referrals (n = 1174) made between 1 January and 31 March. For each referral, the photographs included were reviewed. Eighty-four per cent of the 271 USC referrals made in 2020 included photos, 90.3% of the 413 referrals in 2021 and 91.2% of the 490 referrals in 2022. We devised criteria to determine quantitatively the quality of photographs based on the presence of a locator image, a close-up image, a ruler, a dermatoscopic image and whether the photos were in focus. Each feature scored 1, generating a quality score of 0–5. The mean quality scores in 2020, 2021 and 2022 were 2.16, 2.13 and 2.30, respectively. A Kruskal–Wallis test was performed to determine whether there was statistical significance to the differences seen between the sample groups. The alternative hypothesis was that the distribution of the quality scores across the sample groups was unequal. An overall P-value of 0.075 was generated, which was below the threshold of 0.1; therefore, the alternative hypothesis was accepted. On paired comparison between the sample groups, there was a statistically significant increase from 2021 to 2022, with an adjusted P-value of 0.069. We also subjectively judged whether the photos from each referral had made a positive impact on the final diagnosis. Seventy-five per cent (n = 170) of referrals in 2020, 251 (67.3%) in 2021 and 356 (79.6%) in 2022 included photos that had a positive impact on the final diagnosis. To conclude, our results showed that there was a significant decrease in the quality of photographs included in USC referrals in 2021 when compared with the sample groups of 2020 and 2022, suggesting that the COVID-19 pandemic had a negative impact on the triage of USC referrals from primary care. This could be explained by the reduction in face-to-face consultations and an increase in the number of patient-provided images. The results highlight that a significant proportion of images sent by general practitioners and patients are of insufficient quality for effective review and diagnosis. Possible solutions proposed include further provision of medical photography services and education. This work will be used as a baseline for the introduction of a teledermatology service.

Accuracy of a Smartphone-Based Artificial Intelligence Application for Classification of Melanomas, Melanocytic Nevi, and Seborrheic Keratoses

Current artificial intelligence algorithms can classify melanomas at a level equivalent to that of experienced dermatologists. The objective of this study was to assess the accuracy of a smartphone-based “You Only Look Once” neural network model for the classification of melanomas, melanocytic nevi, and seborrheic keratoses. The algorithm was trained using 59,090 dermatoscopic images. Testing was performed on histologically confirmed lesions: 32 melanomas, 35 melanocytic nevi, and 33 seborrheic keratoses. The results of the algorithm’s decisions were compared with those of two skilled dermatologists and five beginners in dermatoscopy. The algorithm’s sensitivity and specificity for melanomas were 0.88 (0.71–0.96) and 0.87 (0.76–0.94), respectively. The algorithm surpassed the beginner dermatologists, who achieved a sensitivity of 0.83 (0.77–0.87). For melanocytic nevi, the algorithm outclassed each group of dermatologists, attaining a sensitivity of 0.77 (0.60–0.90). The algorithm’s sensitivity for seborrheic keratoses was 0.52 (0.34–0.69). The smartphone-based “You Only Look Once” neural network model achieved a high sensitivity and specificity in the classification of melanomas and melanocytic nevi with an accuracy similar to that of skilled dermatologists. However, a bigger dataset is required in order to increase the algorithm’s sensitivity for seborrheic keratoses.

Skin Melanoma Cancer Detection and Classification using Machine Learning

Skin cancer is a common and potentially life-threatening disease that affects millions of people worldwide. Early detection and accurate classification of skin lesions are critical for effective therapy and improved patient outcomes. In recent years, advances in machine learning and computer vision techniques have shown promising results in automating skin cancer detection and classification. The goal of this project is to develop an automated system for skin cancer detection and classification using machine learning algorithms. The proposed system uses a dataset of dermatoscopic images collected from different sources covering different types of skin lesions, including malignant melanoma, basal cell carcinoma, and squamous cell carcinoma. The project includes several main stages. First, preprocessing techniques including noise reduction, normalization, and feature extraction are used to improve image quality. Next, a comprehensive set of features such as color, texture, and portrait features are extracted from the preprocessed images. These features are input to various machine learning models, including Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs) and Random Forests.

Is photographic surveillance in melanoma diagnosis dispensable?

Thank you for the opportunity to respond to Melanoma diagnosis at a specialist dermatology practice without the use of photographic surveillance.1 In this study, the authors provide an interesting descriptive report on a private dermatology practice in Brisbane, Queensland, Australia, practising without photographic surveillance. They calculate a single defined metric, melanoma in situ to invasive ratio, and compare it to the same metric in 11 other studies, nine of which involve practices which utilise photographic surveillance in the form of total body photography, serial digital dermatoscopic imaging or both. They then draw a conclusion that their method of practice produces a higher melanoma in situ to invasive ratio than the alternative studies. They add that the utility of dermatoscopy historically ‘…was demonstrated in studies’, and that similarly, only a ‘prospective, randomised and controlled trial’ can demonstrate whether practice using photographic technologies is superior. The authors present their conclusion in spite of admitting that the chosen metric is unsuitable for such comparisons, stating ‘Direct comparison between studies is problematic’ and conceding that confounding factors such as ‘sun exposure, public education and access to healthcare’ and ‘variable patient and clinician characteristics’ may affect the ratios.1 They also admit the impact of 43.4% of their Queensland, Australia-derived melanomas being lentigo maligna subtype, a subtype known to be predominantly in situ. The elephant in the room not mentioned is the impact of their patients being filtered through general practitioner (GP) referrals. Of relevance to this, in a recent study that reported 637 melanomas encountered by 27 Australasian GPs in 2013, GPs definitively managed 74.9% of them, referring 17.6% to surgeons and 0.3% to dermatologists, invasive melanomas preferentially being referred to general surgeons.2 It is true that there have been studies comparing naked eye examination and dermatoscopy in the clinical setting and the suggestion that a prospective, randomised and controlled trial to validate photographic surveillance is also necessary, might seem reasonable. However, there are ethical considerations in designing a study which according to currently published literature, would put patients in the control arm at increased risk.3 It would also be unlikely that patients would agree to such a trial when they were informed, in the mandatory patient information leaflet, that the control arm was receiving treatment contrary to current guidelines,4 with a stated risk of missing up to 10% of melanomas, which are featureless, even at the time of excision, only being detectable by side-by-side examination of images taken at different times.3, 5 The practice of excising ‘suspect lesions’ or ‘lesions considered suspicious’1 would not be expected to discover such lesions, a situation the authors have not addressed. As a descriptive report, this paper an interesting presentation of an alternative practice style which is resisting technological innovations. The conclusions are at best unsupported with the authors' recommendations likely to be ethically challenging. Cliff Rosendahl conceived and wrote the manuscript. None Open access publishing facilitated by The University of Queensland, as part of the Wiley - The University of Queensland agreement via the Council of Australian University Librarians.

Design and validation of a new machine-learning-based diagnostic tool for the differentiation of dermatoscopic skin cancer images.

Skin cancer is the most common cancer in the United States. Current estimates are that one in five Americans will develop skin cancer in their lifetime. A skin cancer diagnosis is challenging for dermatologists requiring a biopsy from the lesion and histopathological examinations. In this article, we used the HAM10000 dataset to develop a web application that classifies skin cancer lesions. This article presents a methodological approach that utilizes dermoscopy images from the HAM10000 dataset, a collection of 10015 dermatoscopic images collected over 20 years from two different sites, to improve the diagnosis of pigmented skin lesions. The study design involves image pre-processing, which includes labelling, resizing, and data augmentation techniques to increase the instances of the dataset. Transfer learning, a machine learning technique, was used to create a model architecture that includes EfficientNET-B1, a variant of the baseline model EfficientNET-B0, with a global average pooling 2D layer and a softmax layer with 7 nodes added on top. The results of the study offer a promising method for dermatologists to improve their diagnosis of pigmented skin lesions. The model performs best in detecting melanocytic nevi lesions with an F1 score of 0.93. The F1 score for Actinic Keratosis, Basal Cell Carcinoma, Benign Keratosis, Dermatofibroma, Melanoma, and Vascular lesions was consecutively 0.63, 0.72, 0.70, 0.54, 0.58, and 0.80. We classified seven distinct skin lesions in the HAM10000 dataset with an EfficientNet model reaching an accuracy of 84.3%, which provides a promising outlook for further development of more accurate models.

Melanoma Clinical Decision Support System: An Artificial Intelligence-Based Tool to Diagnose and Predict Disease Outcome in Early-Stage Melanoma Patients.

This study set out to assess the performance of an artificial intelligence (AI) algorithm based on clinical data and dermatoscopic imaging for the early diagnosis of melanoma, and its capacity to define the metastatic progression of melanoma through serological and histopathological biomarkers, enabling dermatologists to make more informed decisions about patient management. Integrated analysis of demographic data, images of the skin lesions, and serum and histopathological markers were analyzed in a group of 196 patients with melanoma. The interleukins (ILs) IL-4, IL-6, IL-10, and IL-17A as well as IFNγ (interferon), GM-CSF (granulocyte and macrophage colony-stimulating factor), TGFβ (transforming growth factor), and the protein DCD (dermcidin) were quantified in the serum of melanoma patients at the time of diagnosis, and the expression of the RKIP, PIRIN, BCL2, BCL3, MITF, and ANXA5 proteins was detected by immunohistochemistry (IHC) in melanoma biopsies. An AI algorithm was used to improve the early diagnosis of melanoma and to predict the risk of metastasis and of disease-free survival. Two models were obtained to predict metastasis (including "all patients" or only patients "at early stages of melanoma"), and a series of attributes were seen to predict the progression of metastasis: Breslow thickness, infiltrating BCL-2 expressing lymphocytes, and IL-4 and IL-6 serum levels. Importantly, a decrease in serum GM-CSF seems to be a marker of poor prognosis in patients with early-stage melanomas.

Segmentation and Classification of Dermatoscopic Skin Lesion images using U-Net and MobileNet models

The 17th most prevalent cancer in the world is cutaneous melanoma. Early detection and adequate treatment are essential for skin cancer success. It might not be possible to tell benign lesions from malignant tumours just by looking at them. The histopathological study of the skin biopsy is the gold standard procedure. Skin biopsy has some drawbacks, including its invasiveness, the pain it causes, and the requirement for many samples for suspected lesions with multiple presentations. Clinical diagnosis can also be aided by noninvasive tools. Several non-invasive imaging techniques are now available to diagnose melanoma because of numerous scientific and technological developments. The most advanced network for pattern identification in medical image analysis is the convolutional neural network (CNN). Thus, utilizing these advanced techniques a Skin Lesion Classifier can be built, which performs segmentation of the Lesion area in the pre-process step to avoid extracting and remembering other additional features from the background of the dermatoscopic image. For the segmentation task U-Net Architecture is utilized on the PH2 dataset and obtained validation accuracy of 95%. And performs well on the test data. The segmented images are then used to train a lightweight CNN Architecture “MobileNet” using some pretrained weights from imagenet dataset and produced validation accuracy of 84.73% which is pretty good performance.

AI Techniques of Dermoscopy Image Analysis for the Early Detection of Skin Lesions Based on Combined CNN Features.

Melanoma is one of the deadliest types of skin cancer that leads to death if not diagnosed early. Many skin lesions are similar in the early stages, which causes an inaccurate diagnosis. Accurate diagnosis of the types of skin lesions helps dermatologists save patients' lives. In this paper, we propose hybrid systems based on the advantages of fused CNN models. CNN models receive dermoscopy images of the ISIC 2019 dataset after segmenting the area of lesions and isolating them from healthy skin through the Geometric Active Contour (GAC) algorithm. Artificial neural network (ANN) and Random Forest (Rf) receive fused CNN features and classify them with high accuracy. The first methodology involved analyzing the area of skin lesions and diagnosing their type early using the hybrid models CNN-ANN and CNN-RF. CNN models (AlexNet, GoogLeNet and VGG16) receive lesions area only and produce high depth feature maps. Thus, the deep feature maps were reduced by the PCA and then classified by ANN and RF networks. The second methodology involved analyzing the area of skin lesions and diagnosing their type early using the hybrid CNN-ANN and CNN-RF models based on the features of the fused CNN models. It is worth noting that the features of the CNN models were serially integrated after reducing their high dimensions by Principal Component Analysis (PCA). Hybrid models based on fused CNN features achieved promising results for diagnosing dermatoscopic images of the ISIC 2019 data set and distinguishing skin cancer from other skin lesions. The AlexNet-GoogLeNet-VGG16-ANN hybrid model achieved an AUC of 94.41%, sensitivity of 88.90%, accuracy of 96.10%, precision of 88.69%, and specificity of 99.44%.

Skin Cancer Prediction using Deep Learning

There are over 200 different forms of cancer. Out of 200 cases, melanoma is the most lethal form of skin cancer. The melanoma diagnostic process begins with clinical screening followed by dermatoscopy and histopathological examination. If cutaneous melanoma is detected early, the cure rate is high. The first step in diagnosing cutaneous melanoma is a visual examination of the affected areas of the skin. Dermatologists take the dermatoscopic images of the skin lesions by the high-speed camera, which have an accuracy of 65-80% in the melanoma diagnosis without any additional technical support. With further visual examination by cancer treatment specialists and dermatoscopic images, the overall prediction rate of melanoma diagnosis raised to 75-84% accuracy. The project aims to build an automated classification system based on image processing techniques to classify skin cancer using skin lesions images. There is a necessary need for early detection of skin cancer and can prevent further spread in some cases of skin cancers, such as melanoma and focal cell carcinoma. Anyhow there are several factors that have bad impacts on the detection accuracy. Recently, the use of image processing and machine vision in medical and medical applications is growing to a greater degree. In this article, convolutional neural networks are used to detect and classify cancer classes based on historical clinical image data using CNNs. Some of the goals of this study are to build a CNN model for skin cancer detection with over 80% accuracy, keep the false-negative rate below 10% in prediction, achieve over 80% accuracy, and render the data. Simulation results show that the proposed method has advantages over other comparison methods.

Registration of dermatoscopic images of skin neoplasms and detection of structural differences

The paper considers the application of an algorithm based on a morphological projector for determining structural differences for comparing dermoscopic images. This will allow to identify changes that have occurred in skin lesions over time, for a more accurate diagnosis of their malignancy. The proposed algorithm makes it possible to detect differences in images even if there is a significant difference in the brightness and color levels of the compared images, and also ignores small insignificant details, such as noise, dermatoscope optics marks, hair, etc. A method for correcting the desynchronization of images using the structural similarity index as a similarity metric, and the sinecosine algorithm as an optimization algorithm is proposed. The proposed algorithms were tested on dermatoscopic images and the possibility of their application was demonstrated.

Comparing Dermatoscopic Features With Slit Skin Smear and Histopathology in Diagnosis of Cutaneous Leishmaniasis.

To evaluate dermatoscopic features of cutaneous leishmaniasis and to compare its diagnostic accuracy against slit skin smear and skin histopathology. This cross-sectional study was conducted at the Department of Dermatology, Pakistan Navy Station (PNS) Shifa Hospital, Karachi, Pakistan, from August 2021 to August 2022. A total of 200 lesions from 70 patients of cutaneous leishmaniasis diagnosed with slit skin smear for Leishmania-Donovan (LD) bodies and skin biopsy were included via non-probability consecutive sampling technique. Dermatoscopic evaluation was performed via a handheld dermatoscope (DELTA 20T; HEINE,Gilching, Germany) on 10x magnification. All dermatoscopic images were analyzed by two different observers who had command of dermatoscopy. Data analysis was done using Statistical Package for the Social Sciences SPSS version 27 (IBM Corp., Armonk, NY, USA). Common dermatoscopic findings were erythema 200 (100%), hyperkeratosis 140 (70%), crusting 50 (25%), ulceration 42 (21%), milia-like structure 58 (29%), tear drop-like structure 46 (23%), yellow tears 70 (35%), and white starburst pattern 68 (34%). Less common findings were yellow hue 28 (14%), orange areas 26 (13%) and scar seven (3.5%). Vascular structures frequently observed were linear vessels 109 (54.5%), dotted vessels 80 (40%), and hairpin vessels 61 (30.5%); less common findings were comma-shaped vessels 52 (26%), arborizing vessels 20 (10%), crown vessels nine (4.5%). Comparison of dermatoscopic features was done with slit skin smear for LD bodies (p value = 0.003 ) and histopathology (p value = 0.001). Dermatoscopy is a non-invasive technique that is helpful in diagnosing cutaneous leishmaniasis, saving time in making rapid diagnosis and saving the need to undergo extensive invasive investigation. Yield of dermatoscopy was comparable to slit smear for LD bodies and histopathology and was found to be effective in making rapid diagnosis with significant accuracy (p value <0.05).

Detection and classification of dermatoscopic images using segmentation and transfer learning

Detection and classification of dermatoscopic images using segmentation and transfer learning