Skin Lesions In Dermoscopic Images Research Articles

Segmentation of dermatoscopic images of skin lesions. Comparison of methods

The work discusses a number of techniques for segmenting dermoscopic images of skin lesions to identify the areas occupied by these lesions. Segmentation is necessary as the first stage of most methods of computer diagnostics of malignancy of neoplasms. A number of techniques, such as ABCDE, use the shape of the tumor as one of the criteria for making a diagnosis; for others, such as the use of convolutional neural networks, identifying the tumor allows one to increase the accuracy of the results obtained. The work discusses three methods of segmentation: thresholding using Otsu's method to calculate the threshold value, a convolutional neural network built on the U-net architecture, and a similar convolutional neural network with an added attention mechanism. The advantages and disadvantages of each method are considered, as well as the possibility of using them together to obtain the best segmentation results.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 sine-cosine algorithm as an optimization algorithm is proposed. The proposed algorithms were tested on dermatoscopic images and the possibility of their application was demonstrated.

Classification of Melanoma Skin Cancer using Transfer Learning

Abstract: Melanoma, a fatal variety of skin cancer, is a significant public health issue on a global scale. For patients to have better results, melanoma must be detected accurately and early. In this study, transfer learning techniques are used in deep learning to provide a novel method of melanoma cancer detection. By using pre-trained convolutional neural networks (CNNs), such as VGG, MobileNet, and EfficientNet, as feature extractors, the research takes advantage of the capability of transfer learning. These networks, which were first created for tasks involving generic image identification, are refined using a large dataset of dermoscopic images of skin lesions in order to adapt them to the particular needs of melanoma detection. Essential preprocessing techniques, including image resizing, image normalization, and lesion segmentation using the Otsu segmentation method, are used to improve the dataset's quality and consistency. Image normalization has been used to reduce the complexity and processing time of the image. Skin lesions have been segmented using Otsu segmentation. These procedures increase the robustness of the melanoma detection model and help to prepare the images for efficient training. The findings of this study highlight the possibility for an earlier and more precise diagnosis and provide encouraging insights into the viability of transfer learning as a formidable tool for melanoma detection. The overall objective of this work is to improve the diagnosis and patient care for melanoma cancer. This work contributes to ongoing efforts to provide practical and effective tools for dermatologists and healthcare professionals.

Residual cosine similar attention and bidirectional convolution in dual-branch network for skin lesion image classification

Skin cancer is one of the most serious threats to human health among skin lesions. Computer-aided diagnosis methods can assist patients in identifying and detecting skin lesion types early, thereby enabling corresponding treatments. In this paper, we propose a dual-branch neural network model Conformer with Residual Cosine Similarity Attention and Bidirectional Convolutional fusion strategy, named RCSABC-Conformer. The core of this network structure comprises three parts: a Convolutional Neural Network (CNN) branch with Residual Cosine Similarity Attention (RCSA), a Transformer branch, and a Feature Couple Unit with Bidirectional Convolutional strategy (BC-FCU). The RCSA module calculates the cosine similarity value between the feature map generated by the convolutional operation and the feature map of the residual edge to assess whether their semantic information is similar. The semantic information of similar parts is weighted by exponential normalization to enhance the network's memory of similar features of the same type of skin lesion. The BC-FCU module interactively fuses local features and global representations of skin lesion images with different resolutions in the two branches. Specifically, when the global representations is integrated into local features, we introduce a new bidirectional convolution strategy to extract the feature map from both forward and backward directions, and then select the element with the smaller feature value from the two directions to fuse into local features. In this way, we can minimize the interference of the artifact features extracted by the Transformer branch on the CNN branch. In addition, taking advantage of the Transformer branch's capacity to construct global representations, our model can learn contextual semantic information of normal skin and lesion areas to enhance model robustness. We conducted experiments on three datasets, consisting of clinical and dermoscopic skin lesion images, as well as a hybrid of both. The experimental results show that RCSABC-Conformer outperforms both advanced and classical classification methods in terms of classification accuracy across all three datasets, without requiring an increase in the number of parameters and computational complexity. Compared with the baseline model, the classification accuracy of our proposed method improves by 2.40%, 5.39%, and 4.44% on the three datasets, respectively. To the best of our knowledge, this is the first study to apply an interactive fusion dual-branch network for multi-disease classification on different modalities of skin lesion databases. Code will be available at https://github.com/AlenLi817/RCSABC-Conformer.

Deep Learning for Automatic Diagnosis of Skin Cancer Using Dermoscopic Images

Over the past few years, the healthcare industry has seen a dramatic increase in the use of intelligent automation enabled by AI technology. These developments are made to better the standard of medical decision making and the standard of treatment given to patients. Fuzzy boundaries, shifting sizes, and aberrations like hair or ruler lines all provide difficulties for automatic detection of skin lesions in dermoscopic images, slowing down the otherwise efficient process of diagnosing skin cancer. However, these difficulties may be conquered by employing image processing software. To address these issues, the authors of this paper provide a novel IMLT-DL model for intelligent dermoscopic image processing. Multi-level thresholding and deep learning are brought together in this model. Top hat filtering and inpainting have been included into IMLT-DL for use in image processing. In addition, Mayfly Optimization has been used in tandem with multilayer Kapur's thresholding to identify specific trouble spots. For further investigation, it uses an Inception v3-based feature extractor, and for data classification, it makes use of gradient boosting trees (GBTs). On the ISIC dataset, this model was shown to outperform state-of-the-art alternatives by a margin of 0.992% over the duration of trial iterations. These advances are a major step forwards in the quest for faster and more accurate skin lesion detection.

Crow search algorithm with deep transfer learning driven skin lesion detection on dermoscopic images

The early diagnosis of skin cancer is of paramount importance for effective patient treatment. Dermoscopy, a non-surgical technique, utilizes precise equipment to examine the skin and plays a crucial role in identifying specific features and patterns that may indicate the presence of skin carcinoma. In recent times, machine learning (ML) methods have been developed to recognize and classify dermoscopic images as either malignant or benign. Deep learning (DL) systems, including Convolutional Neural Networks (CNNs), as well as various ML models like Random Forest (RF) classifiers and Support Vector Machine (SVM), are employed to extract relevant features from these images. This study introduces the Crow Search Algorithm with Deep Transfer Learning Driven Skin Lesion Detection on Dermoscopic Images (CSADTL-SLD) technique. The CSADTL-SLD method starts with the application of a median filter (MF) to remove noise from the images and utilizes the GoogleNet model for feature extraction. GoogleNet is well-regarded for its capacity to capture intricate and meaningful patterns within the data, which are essential for accurate lesion characterization. Furthermore, the CSADTL-SLD technique applies the Crow Search Algorithm (CSA) for parameter tuning of the GoogleNet model. After feature selection, the system employs the MLP classification model for precise lesion categorization. The comprehensive results of this research demonstrate the superiority of the CSADTL-SLD algorithm, showing significant enhancements in skin lesion detection accuracy and robustness when compared to existing methods. This approach holds promise as an effective solution for automating the detection and classification of skin lesions in dermoscopic images.

Evaluation of an artificial intelligence-based decision support for detection of cutaneous melanoma in primary care - a prospective, real-life, clinical trial.

Use of artificial intelligence, or machine learning, to assess dermoscopic images of skin lesions to detect melanoma has in several retrospective studies shown high levels of diagnostic accuracy on par with, or even outperforming, experienced dermatologists. However, the enthusiasm around these algorithms has not yet been matched by prospective clinical trials performed in authentic clinical settings. In several European countries, including Sweden, the initial clinical assessment of suspected skin cancer is principally conducted in the primary health care setting by primary care physicians; with or without access to teledermoscopic support from dermatology clinics. To determine the diagnostic performance of an artificial intelligence-based clinical decision support tool for cutaneous melanoma detection, operated by a smartphone application (app), when used prospectively by primary care physicians to assess skin lesions of concern due to some degree of melanoma suspicion. This prospective, multicentre, clinical trial was conducted at 36 primary care centres in Sweden. The physicians used the smartphone app on skin lesions of concern by photographing them dermoscopically, which resulted in a dichotomous decision support text regarding evidence for melanoma. Regardless of the app outcome, all lesions underwent standard diagnostic procedure, by surgical excision or referral to dermatologist. After completed investigation, lesion diagnoses were collected from the patients' medical records and compared to app outcome and other lesion data. In total, 253 lesions of concern in 228 patients were included, of which 21 proved to be melanomas, with 11 thin invasive melanomas and 10 melanomas in situ. The app's accuracy (95% confidence interval) in identifying melanomas was reflected in an area under the receiver operating characteristic (AUROC) curve of 0.960 (0.928-0.980), corresponding to a maximum sensitivity and specificity of 95.2% and 84.5%, respectively. For invasive melanomas alone, the AUROC was 0.988 (0.965-0.997), corresponding to a maximum sensitivity and specificity of 100% and 92.6%, respectively. The clinical decision support tool evaluated in this investigation showed high diagnostic accuracy when used prospectively on primary care patients, which could add significant clinical value for primary care physicians in assessing skin lesions to detect melanoma. ClinicalTrials.gov Identifier: NCT05172232.

Artificial Intelligence Algorithms for Benign vs. Malignant Dermoscopic Skin Lesion Image Classification.

In recent decades, the incidence of melanoma has grown rapidly. Hence, early diagnosis is crucial to improving clinical outcomes. Here, we propose and compare a classical image analysis-based machine learning method with a deep learning one to automatically classify benign vs. malignant dermoscopic skin lesion images. The same dataset of 25,122 publicly available dermoscopic images was used to train both models, while a disjointed test set of 200 images was used for the evaluation phase. The training dataset was randomly divided into 10 datasets of 19,932 images to obtain an equal distribution between the two classes. By testing both models on the disjoint set, the deep learning-based method returned accuracy of 85.4 ± 3.2% and specificity of 75.5 ± 7.6%, while the machine learning one showed accuracy and specificity of 73.8 ± 1.1% and 44.5 ± 4.7%, respectively. Although both approaches performed well in the validation phase, the convolutional neural network outperformed the ensemble boosted tree classifier on the disjoint test set, showing better generalization ability. The integration of new melanoma detection algorithms with digital dermoscopic devices could enable a faster screening of the population, improve patient management, and achieve better survival rates.

Dermatological Detection and Classification using Machine Learning Techniques

Dermatology is the medical field that focuses on the study and treatment of skin conditions. It is a specialized branch of medicine that encompasses both diagnostic and surgical procedures related to the skin. It is a widespread disease among them. The researchers have shows a lot of attention to the early detection of lesions. Because of their proliferation ability to other parts of the body, death rates are quite high. A system that can distinguish between benign and malignant lesions is essential because melanoma can be cured with an early and accurate diagnosis. Dermoscopic skin lesion images are first segmented using data mining techniques, to identify the area of interest of the lesion part. When compared to individual classifier algorithms, dermatology datasets benefit from the various data mining techniques and feature selection methods. The SVM provides more accurate and effective skin disease prediction in terms of accuracy, precision, and Specificity.

A deep convolutional structure-based approach for accurate recognition of skin lesions in dermoscopy images

<span lang="EN-US">One-third of all cancer diagnoses worldwide are skin malignancies. One of the most common tumors, skin cancer can develop from a variety of dermatological conditions and is subdivided into different categories based on its textile, color, body, and other morphological characteristics. The most effective strategy to lower the mortality rate of melanoma is early identification because skin cancer incidence has been on the rise recently. In order to categorize dermoscopy images into the four diagnosis classifications of melanoma, benign, malignant, and human against machine (HAM) not melanoma, this research suggests a computer-aided diagnosis (CAD) system. Experimental results show that the suggested approach enabled 97.25% classification accuracy. In order to automate the identification of skin cancer and expedite the diagnosis process in order to save a life, the proposed technique offers a less complex and cutting-edge framework.</span>

M2CE: Multi‐convolutional neural network ensemble approach for improved multiclass classification of skin lesion

AbstractDue to inter‐class homogeneity and intra‐class variability, the classification of skin lesions in dermoscopy images has remained difficult. Although deep convolutional neural networks (DCNNs) have achieved satisfactory performance for binary skin cancer classification, multiclass skin lesion classification is still an open problem due to the limited training samples and class imbalance issues. To tackle these issues, in this article, we propose a multi‐CNN ensemble approach dubbed for multiclass skin lesion classification. The includes three individual CNN models, each helping in extracting different high‐level features from skin images and thereby yielding different prediction results. First, we design a lightweight CNN model to extract prominent features and train it from scratch, which primarily aims at avoiding the data scarcity problem. Then, we ensemble two different pre‐trained CNN models with the lightweight model to improve the performance and generalization capability. The proposed ensemble approach can effectively fuse the predictions of each individual CNN model using the averaging method. The approach is validated using a benchmark data set, HAM10000, which contains skin lesion images of seven different classes. The results demonstrate that the outperforms base CNN models and state‐of‐the‐art approaches without using any external data.

Improving Skin Cancer Diagnostics Through a Mobile App With a Large Interactive Image Repository: Randomized Controlled Trial.

Skin cancer diagnostics is challenging, and mastery requires extended periods of dedicated practice. The aim of the study was to determine if self-paced pattern recognition training in skin cancer diagnostics with clinical and dermoscopic images of skin lesions using a large-scale interactive image repository (LIIR) with patient cases improves primary care physicians' (PCPs') diagnostic skills and confidence. A total of 115 PCPs were randomized (allocation ratio 3:1) to receive or not receive self-paced pattern recognition training in skin cancer diagnostics using an LIIR with patient cases through a quiz-based smartphone app during an 8-day period. The participants' ability to diagnose skin cancer was evaluated using a 12-item multiple-choice questionnaire prior to and 8 days after the educational intervention period. Their thoughts on the use of dermoscopy were assessed using a study-specific questionnaire. A learning curve was calculated through the analysis of data from the mobile app. On average, participants in the intervention group spent 2 hours 26 minutes quizzing digital patient cases and 41 minutes reading the educational material. They had an average preintervention multiple choice questionnaire score of 52.0% of correct answers, which increased to 66.4% on the postintervention test; a statistically significant improvement of 14.3 percentage points (P<.001; 95% CI 9.8-18.9) with intention-to-treat analysis. Analysis of participants who received the intervention as per protocol (500 patient cases in 8 days) showed an average increase of 16.7 percentage points (P<.001; 95% CI 11.3-22.0) from 53.9% to 70.5%. Their overall ability to correctly recognize malignant lesions in the LIIR patient cases improved over the intervention period by 6.6 percentage points from 67.1% (95% CI 65.2-69.3) to 73.7% (95% CI 72.5-75.0) and their ability to set the correct diagnosis improved by 10.5 percentage points from 42.5% (95% CI 40.2%-44.8%) to 53.0% (95% CI 51.3-54.9). The diagnostic confidence of participants in the intervention group increased on a scale from 1 to 4 by 32.9% from 1.6 to 2.1 (P<.001). Participants in the control group did not increase their postintervention score or their diagnostic confidence during the same period. Self-paced pattern recognition training in skin cancer diagnostics through the use of a digital LIIR with patient cases delivered by a quiz-based mobile app improves the diagnostic accuracy of PCPs. ClinicalTrials.gov NCT05661370; https://classic.clinicaltrials.gov/ct2/show/NCT05661370.

Skin Cancer Detection

Abstract: Skin Cancer Detection using VGG-16 in CNN. Skin cancer is one of the most common and potentially life-threatening forms of cancer, and early detection plays a crucial role in improving patient outcomes. In recent years, Convolutional Neural Networks (CNNs) have shown remarkable success in various computer vision tasks, including medical image analysis. This abstract presents a study on the application of the VGG-16 architecture, a widely used CNN, for the detection of skin cancer. The proposed method involves training a VGG-16 model using a large dataset of dermoscopic images of skin lesions. The dataset comprises both malignant (cancerous) and benign (non-cancerous) cases, providing a diverse range of examples for the network to learn from. Through a process of supervised learning, the VGG-16 model is trained to recognize patterns and features indicative of skin cancer in these images

Melanoma Skin Cancer Segmentation using Robust Multi-View Fuzzy C- Means Algorithm

Skin cancer has become a significant issue and the primary cause of death for people all over the world. Melanoma is one of the most serious types of skin cancer since it can spread to any area of the body. So, the prognosis of melanoma in its early stages is very important for figuring out how likely it is that the patient will get better. Because of this, clinical imaging science is a very important part of finding certain types of skin lesions quickly and correctly. In this study, we showed a new way to use digital image processing to separate skin lesions in dermoscopic images. First, the image is preprocessed with a median filter to eliminate extra hair, noise, and artefacts in order to create a better image for segmentation. To improve the accuracy of dermoscopic images, the preprocessed images are segmented based on an innovative method built on an improved robust Multiview Fuzzy C-Means clustering algorithm (RMV-FCM). Following segmentation, features are extracted using the ABCD rule to produce optimal features that may be used as input for classification. Lastly, the Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forest (RF), and Nave Bayes classifiers use machine learning techniques to characterize skin lesions as malignant and benign (NB). The RMV-FCM algorithm exhibits more adaptability and higher clustering performance when compared to a variety of related clustering techniques. In terms of detection accuracy, as noted in the conclusion of this research, the novelty of the work shows that RMV-FCM is exceptional in comparison to numerous related clustering algorithms, and SVM is exceptional in comparison to other standard classifiers. The simulated outcome indicates that the suggested strategy accurately diagnoses skin tumors with a 97.5% accuracy rate. The results of the suggested segmentation are extremely accurate when compared to other algorithms in the same domain

BiDFDC-Net: a dense connection network based on bi-directional feedback for skin image segmentation.

Accurate segmentation of skin lesions in dermoscopic images plays an important role in improving the survival rate of patients. However, due to the blurred boundaries of pigment regions, the diversity of lesion features, and the mutations and metastases of diseased cells, the effectiveness and robustness of skin image segmentation algorithms are still a challenging subject. For this reason, we proposed a bi-directional feedback dense connection network framework (called BiDFDC-Net), which can perform skin lesions accurately. Firstly, under the framework of U-Net, we integrated the edge modules into each layer of the encoder which can solve the problem of gradient vanishing and network information loss caused by network deepening. Then, each layer of our model takes input from the previous layer and passes its feature map to the densely connected network of subsequent layers to achieve information interaction and enhance feature propagation and reuse. Finally, in the decoder stage, a two-branch module was used to feed the dense feedback branch and the ordinary feedback branch back to the same layer of coding, to realize the fusion of multi-scale features and multi-level context information. By testing on the two datasets of ISIC-2018 and PH2, the accuracy on the two datasets was given by 93.51% and 94.58%, respectively.

An IoMT-Based Melanoma Lesion Segmentation Using Conditional Generative Adversarial Networks

Currently, Internet of medical things-based technologies provide a foundation for remote data collection and medical assistance for various diseases. Along with developments in computer vision, the application of Artificial Intelligence and Deep Learning in IOMT devices aids in the design of effective CAD systems for various diseases such as melanoma cancer even in the absence of experts. However, accurate segmentation of melanoma skin lesions from images by CAD systems is necessary to carry out an effective diagnosis. Nevertheless, the visual similarity between normal and melanoma lesions is very high, which leads to less accuracy of various traditional, parametric, and deep learning-based methods. Hence, as a solution to the challenge of accurate segmentation, we propose an advanced generative deep learning model called the Conditional Generative Adversarial Network (cGAN) for lesion segmentation. In the suggested technique, the generation of segmented images is conditional on dermoscopic images of skin lesions to generate accurate segmentation. We assessed the proposed model using three distinct datasets including DermQuest, DermIS, and ISCI2016, and attained optimal segmentation results of 99%, 97%, and 95% performance accuracy, respectively.

Leveraging Vision Attention Transformers for Detection of Artificially Synthesized Dermoscopic Lesion Deepfakes Using Derm-CGAN

Synthesized multimedia is an open concern that has received much too little attention in the scientific community. In recent years, generative models have been utilized in maneuvering deepfakes in medical imaging modalities. We investigate the synthesized generation and detection of dermoscopic skin lesion images by leveraging the conceptual aspects of Conditional Generative Adversarial Networks and state-of-the-art Vision Transformers (ViT). The Derm-CGAN is architectured for the realistic generation of six different dermoscopic skin lesions. Analysis of the similarity between real and synthesized fakes revealed a high correlation. Further, several ViT variations were investigated to distinguish between actual and fake lesions. The best-performing model achieved an accuracy of 97.18% which has over 7% marginal gain over the second best-performing network. The trade-off of the proposed model compared to other networks, as well as a benchmark face dataset, was critically analyzed in terms of computational complexity. This technology is capable of harming laymen through medical misdiagnosis or insurance scams. Further research in this domain would be able to assist physicians and the general public in countering and resisting deepfake threats.

Skin Lesion Segmentation Using K-means Clustering with Removal Unwanted Regions

The segmentation of skin lesions is crucial to the early and accurate identification of skin cancer by computerized systems. It is difficult to automatically divide skin lesions in dermoscopic images because of challenges such as hairs, gel bubbles, ruler marks, fuzzy boundaries, and low contrast. We proposed an effective method based on K-means and a trainable machine learning system to segment regions of interest (ROI) in skin cancer images. The proposed method was implemented in several stages, including grayscale image conversion, contrast image enhancement, artifact removal with noise reduction, skin lesion segmentation from image using K-means clustering, and ROI segmentation from unwanted objects using a trainable machine learning system. The proposed model has been evaluated using the ISIC 2017 publicly available dataset. The proposed method obtained a 90.09 accuracy rate, outperforming several methods in the literature.

A customized deep learning framework for skin lesion classification using dermoscopic images

AbstractAutomated analysis of skin lesions in dermoscopy images has gained much attention due to its medical importance in the early detection of melanoma. Detection of lesions has become a challenge due to the strong visual similarity between benign and malignant skin lesions. In this research, a customized deep convolutional neural network (CNN) architecture has been designed to discriminate between benign and malignant lesions. The model is designed carefully with lesser convolution layers, fewer filters, and parameters to achieve better classification performance compared to pretrained VGG16, ResNet50, InceptionV3 models and, ensures state‐of‐the‐art performance. The proposed model is composed of nine trainable layers: eight convolution layers and one fully connected layer. The suggested framework is extensively evaluated on the benchmark ISIC 2016 challenge dataset. The effect of different input transformations over the dataset has been studied. For fair comparison, standard deep learning models such as VGG16, ResNet50, and InceptionV3 have been used for lesion classification using transfer learning approach. The memory requirement of the proposed model is reduced by 388, 68, and 63 times and FLOPs needed are lowered by 95%, 85%, and 84% compared to VGG16‐TrL, ResNet50‐TrL, and InceptionV3‐TrL, respectively. Results show that class balancing with external images improves classification performance.

Computer Aided Diagnosis of Melanoma Using Deep Neural Networks and Game Theory: Application on Dermoscopic Images of Skin Lesions.

Early detection of melanoma remains a daily challenge due to the increasing number of cases and the lack of dermatologists. Thus, AI-assisted diagnosis is considered as a possible solution for this issue. Despite the great advances brought by deep learning and especially convolutional neural networks (CNNs), computer-aided diagnosis (CAD) systems are still not used in clinical practice. This may be explained by the dermatologist's fear of being misled by a false negative and the assimilation of CNNs to a "black box", making their decision process difficult to understand by a non-expert. Decision theory, especially game theory, is a potential solution as it focuses on identifying the best decision option that maximizes the decision-maker's expected utility. This study presents a new framework for automated melanoma diagnosis. Pursuing the goal of improving the performance of existing systems, our approach also attempts to bring more transparency in the decision process. The proposed framework includes a multi-class CNN and six binary CNNs assimilated to players. The players' strategies is to first cluster the pigmented lesions (melanoma, nevus, and benign keratosis), using the introduced method of evaluating the confidence of the predictions, into confidence level (confident, medium, uncertain). Then, a subset of players has the strategy to refine the diagnosis for difficult lesions with medium and uncertain prediction. We used EfficientNetB5 as the backbone of our networks and evaluated our approach on the public ISIC dataset consisting of 8917 lesions: melanoma (1113), nevi (6705) and benign keratosis (1099). The proposed framework achieved an area under the receiver operating curve (AUROC) of 0.93 for melanoma, 0.96 for nevus and 0.97 for benign keratosis. Furthermore, our approach outperformed existing methods in this task, improving the balanced accuracy (BACC) of the best compared method from 77% to 86%. These results suggest that our framework provides an effective and explainable decision-making strategy. This approach could help dermatologists in their clinical practice for patients with atypical and difficult-to-diagnose pigmented lesions. We also believe that our system could serve as a didactic tool for less experienced dermatologists.

Generalizability and usefulness of artificial intelligence for skin cancer diagnostics: An algorithm validation study

AbstractBackgroundArtificial intelligence can be trained to outperform dermatologists in image‐based skin cancer diagnostics. However, the networks' sensitivity to biases and overfitting may hamper their clinical applicability.ObjectivesThe aim of this study was to explain the potential consequences of implementing convolutional neural networks for stand‐alone melanoma diagnostics and skin lesion triage.MethodsIn this algorithm validation study on retrospective data, we reproduced and evaluated the performance of state‐of‐the‐art artificial intelligence (convolutional neural networks) for skin cancer diagnostics. The networks were trained on 25,331 annotated dermoscopic skin lesion images from an open‐source data set (ISIC‐2019) and tested using a novel data set (AISC‐2021) consisting of 26,591 annotated dermoscopic skin lesion images. We tested the trained algorithms' ability to generalize to new data and their diagnostic performance in two simulations (melanoma diagnostics and skin lesion triage).ResultsThe trained algorithms performed significantly less accurate diagnostics on images of nevi, melanomas and actinic keratoses from the AISC‐2021 data set than the ISIC‐2019 data set (p &lt; 0.003). Almost one‐third (31.1%) of the melanomas were misclassified during the melanoma diagnostics simulation, irrespective of their Breslow thickness. Furthermore, the algorithms marked 92.7% of the lesions ‘suspicious’ during the triage simulation, which yielded a triage sensitivity and specificity of 99.7% and 8.2%, respectively.ConclusionsAlthough state‐of‐the‐art artificial intelligence outperforms dermatologists on image‐based skin lesion classification within an artificial setting, additional data and technological advances are needed before clinical implementation.