Skin Lesion Classification Research Articles

Integrating Explainable AI for Skin Lesion Classifications: A Systematic Literature Review

Skin cancer, particularly melanoma, poses a significant global health challenge due to its prevalence and mortality rate. Early detection is critical to improving outcomes, as advanced cases become increasingly difficult to treat. The advent of Artificial Intelligence (AI) and Explainable AI (XAI) techniques has revolutionized dermatological diagnostics by offering accurate and interpretable solutions. This systematic review investigates the integration of XAI in skin lesion classification, analyzing 22 recent studies published between 2019 and 2023. The studies encompass diverse approaches, including deep learning models like CNNs, ResNet, DenseNet, and MobileNet, as well as explainability techniques such as Grad-CAM, SHAP, and saliency maps. Results highlight significant advancements in accuracy and interpretability, with some models achieving over 99% accuracy on datasets like ISIC 2018 and HAM10000. However, challenges persist, including dataset imbalances, limited diversity in patient metadata, and generalizability across different skin types and imaging conditions. XAI methods, by visualizing model decision pathways, enhance transparency, fostering trust among clinicians and enabling seamless AI integration into clinical practice. This review underscores the potential of combining state-of-the-art AI models with explainable frameworks to address the complexities of skin lesion diagnostics. It advocates for future research to prioritize diverse, metadata-rich datasets, innovative optimization techniques, and robust architectures to develop reliable, interpretable diagnostic tools. By bridging the gap between advanced AI and user understanding, this work contributes to the creation of clinically applicable, trustable AI-driven healthcare solutions.

Bilinear Interpolation Augmented Deep Feature Extraction with an Improved Remora Optimization-Based Deep Convolutional Neural Network for Skin Lesion Classification

Bilinear Interpolation Augmented Deep Feature Extraction with an Improved Remora Optimization-Based Deep Convolutional Neural Network for Skin Lesion Classification

Hierarchical skin lesion image classification with prototypical decision tree

Traditional disease classification models often disregard the clinical significance of misclassifications and lack interpretability. To overcome these challenges, we propose a hierarchical prototypical decision tree (HPDT) for skin lesion classification. HPDT combines prototypical networks and decision trees, leveraging a class hierarchy to guide interpretable predictions from general to specific categories. By incorporating a hierarchy-based distance matrix, the model prioritizes less severe misclassifications while maintaining diagnostic accuracy. Evaluated on a dataset of 235,268 dermoscopic images across 65 conditions, HPDT outperforms flat classifiers and existing hierarchical methods in accuracy, error severity reduction, and interpretability. It also generalizes effectively to unseen classes. These results highlight the value of integrating clinical hierarchies into model design and training to improve diagnostic reliability and decision transparency, demonstrating HPDT’s potential for clinical decision support.

Skin Lesion Classification Through Test Time Augmentation and Explainable Artificial Intelligence.

Despite significant advancements in the automatic classification of skin lesions using artificial intelligence (AI) algorithms, skepticism among physicians persists. This reluctance is primarily due to the lack of transparency and explainability inherent in these models, which hinders their widespread acceptance in clinical settings. The primary objective of this study is to develop a highly accurate AI-based algorithm for skin lesion classification that also provides visual explanations to foster trust and confidence in these novel diagnostic tools. By improving transparency, the study seeks to contribute to earlier and more reliable diagnoses. Additionally, the research investigates the impact of Test Time Augmentation (TTA) on the performance of six Convolutional Neural Network (CNN) architectures, which include models from the EfficientNet, ResNet (Residual Network), and ResNeXt (an enhanced variant of ResNet) families. To improve the interpretability of the models' decision-making processes, techniques such as t-distributed Stochastic Neighbor Embedding (t-SNE) and Gradient-weighted Class Activation Mapping (Grad-CAM) are employed. t-SNE is utilized to visualize the high-dimensional latent features of the CNNs in a two-dimensional space, providing insights into how the models group different skin lesion classes. Grad-CAM is used to generate heatmaps that highlight the regions of input images that influence the model's predictions. Our findings reveal that Test Time Augmentation enhances the balanced multi-class accuracy of CNN models by up to 0.3%, achieving a balanced accuracy rate of 97.58% on the International Skin Imaging Collaboration (ISIC 2019) dataset. This performance is comparable to, or marginally better than, more complex approaches such as Vision Transformers (ViTs), demonstrating the efficacy of our methodology.

SL-R-CNN-HHO: multi-class skin lesion classification using region-based convolutional neural networks and harris hawk optimization on the HAM dataset

SL-R-CNN-HHO: multi-class skin lesion classification using region-based convolutional neural networks and harris hawk optimization on the HAM dataset

Deep learning-based skin lesion analysis using hybrid ResUNet++ and modified AlexNet-Random Forest for enhanced segmentation and classification.

Skin cancer is considered globally as the most fatal disease. Most likely all the patients who received wrong diagnosis and low-quality treatment die early. Though if it is detected in the early stages the patient has fairly good chance and the aforementioned diseases can be cured. Consequently, diagnostic identification and management of the patient at this level becomes a rather enormous task. This paper offers a cutting-edge hybrid deep learning approach of better segmentation and classification of skin lesions. The proposed method incorporates three key stages: preprocessing, segmentation of lesions, and classification of lesions. By the stage of preprocessing, a morphology-based technique takes out hair so as to enhance the segmentation precision to use the cleansed images for subsequent analysis. Segmentation cuts off the lesion from the surrounding skin, giving the classification phase a dedicated area of interest and the ability to clear the background noise that may affect classification rates. The isolation enables the model to better analyze anatomical lesion features in order to achieve accurate benign and malignant classifications. Using ResUNet++, the cutting-edge deep learning architecture, we achieved accurate lesion segmentation. Next, we will modify and use an AlexNet-Random Forest (AlexNet-RF) based classifier for robust lesion classification. The proposed hybrid deep learning model is intensively validated on the Ham10000 data set which is one of the most popular datasets for skin lesions analysis. The obtained results show that the utilized approach, compared to the previous ones, is more effective, giving better segmentation and classification results. This method takes advantage of ResUNet++ strong classification skill and modified AlexNet-Random Forest robustness for more accurate segmentation. There is a high probability that ResUNet++, which is highly proficient at medical image segmentation, can produce better segmentation of lesions than the simpler models. The composition of AlexNet's extraction of features with Random Forest ability to reduce overfitting possibly may be more precise in the classification when compared to using only one model.

PoxDetect: Advancing Monkeypox Diagnosis with Machine Learning for Skin Lesion Classification

PoxDetect: Advancing Monkeypox Diagnosis with Machine Learning for Skin Lesion Classification

Temporal Integration of ResNet Features with LSTM for Enhanced Skin Lesion Classification

Temporal Integration of ResNet Features with LSTM for Enhanced Skin Lesion Classification

Artificial Intelligence-Aided Diagnosis System for the Detection and Classification of Private-Part Skin Diseases: Decision Analytical Modeling Study.

Private-part skin diseases (PPSDs) can cause a patient's stigma, which may hinder the early diagnosis of these diseases. Artificial intelligence (AI) is an effective tool to improve the early diagnosis of PPSDs, especially in preventing the deterioration of skin tumors in private parts such as Paget disease. However, to our knowledge, there is currently no research on using AI to identify PPSDs due to the complex backgrounds of the lesion areas and the challenges in data collection. This study aimed to develop and evaluate an AI-aided diagnosis system for the detection and classification of PPSDs: aiding patients in self-screening and supporting dermatologists' diagnostic enhancement. In this decision analytical modeling study, a 2-stage AI-aided diagnosis system was developed to classify PPSDs. In the first stage, a multitask detection network was trained to automatically detect and classify skin lesions (type, color, and shape). In the second stage, we proposed a knowledge graph based on dermatology expertise and constructed a decision network to classify seven PPSDs (condyloma acuminatum, Paget disease, eczema, pearly penile papules, genital herpes, syphilis, and Bowen disease). A reader study with 13 dermatologists of different experience levels was conducted. Dermatologists were asked to classify the testing cohort under reading room conditions, first without and then with system support. This AI-aided diagnostic study used the data of 635 patients from two institutes between July 2019 and April 2022. The data of Institute 1 contained 2701 skin lesion samples from 520 patients, which were used for the training of the multitask detection network in the first stage. In addition, the data of Institute 2 consisted of 115 clinical images and the corresponding medical records, which were used for the test of the whole 2-stage AI-aided diagnosis system. On the test data of Institute 2, the proposed system achieved the average precision, recall, and F1-score of 0.81, 0.86, and 0.83, respectively, better than existing advanced algorithms. For the reader performance test, our system improved the average F1-score of the junior, intermediate, and senior dermatologists by 16%, 7%, and 4%, respectively. In this study, we constructed the first skin-lesion-based dataset and developed the first AI-aided diagnosis system for PPSDs. This system provides the final diagnosis result by simulating the diagnostic process of dermatologists. Compared with existing advanced algorithms, this system is more accurate in identifying PPSDs. Overall, our system can not only help patients achieve self-screening and alleviate their stigma but also assist dermatologists in diagnosing PPSDs.

Beyond binary: multi-class skin lesion classification with AlexNet transfer learning-towards enhanced dermatological diagnosis

ProblemSkin lesions are the major indicator for diagnosing different skin diseases, which are caused by the abnormal growth of skin cells. Skin cancer, one of the most fatal types of cancer in the world, relies on the proper diagnosis of skin lesions and other relevant disease indicators for early detection, which can enhance treatment and increase life expectancy by up to 92%. Diagnosing skin lesions accurately is challenging, even for experienced practitioners.ChallengeArtificial Intelligence (AI) techniques based on deep learning have proved promising in the field of medical image processing and diagnosis. However, there is a lack of comprehensive medical image datasets. This paper will present a skin lesion classification model that employs a transfer learning approach based on Alex Net and classifies eight different skin diseases. This proposed model was trained, validated, and tested using ISIC 2019 challenge data with a very abnormal class distribution. This variation in class size has been overcome using data augmentation and resizing in the preprocessing phase. Preprocessing is performed in such a way that no colour detail or rich identification features will be lost from the images.ResultsThe proposed model achieved a testing accuracy of 90.9%, with a precision of 90.8%, a sensitivity of 90.9%, and an F1-Score of 90.8%. The values for precision and sensitivity are among the highest reported, specifically for models using the ISIC 2019 dataset.

ADVANCED SKIN CANCER DETECTION USING CONVOLUTIONAL NEURAL NETWORKS AND TRANSFER LEARNING

This study investigates the effectiveness of MobileNetV2 transfer learning method and a deep learning based Convolutional Neural Network (CNN) model for the classification of malignant and benign skin lesions in skin cancer diagnosis. Since skin cancer is a disease that can be cured with early detection but can be fatal if delayed, accurate diagnosis is of great importance. The model was trained with MobileNetV2 architecture and performed the classification task with high accuracy on images of skin lesions. Metrics such as accuracy, recall, precision and F1 score obtained during the training and validation processes support the high performance of the model. The accuracy of the model was 92.97%, Recall 92.71%, Precision 94.70% and F1 score 93.47%. The results show that the CNN-based MobileNetV2 model is a reliable and effective tool for skin cancer diagnosis, but small fluctuations in the validation phase require further data and hyperparameter optimization to further improve the generalization ability of the model. This study demonstrates that CNN-based models enhanced with MobileNetV2 transfer learning offer a powerful solution to medical image classification problems and have the potential to contribute to the development of early detection systems in the healthcare field.

Deep Learning-Based Classification of Skin Lesions: Enhancing Melanoma Detection through Automated Preprocessing and Data Augmentation

Skin cancer of the most dangerous type, melanoma, requires an early and accurate diagnosis for its treatment to reduce mortality and increase the number of positive outcomes. Even with the availability of better imaging and diagnostic techniques, it is still difficult to differentiate between benign lesions and malignant melanoma because of overlapping features, noisy images and images with artefacts such as hair and glare. To overcome these challenges, this research adopts deep learning models to classify skin lesions based on images from the ISIC Archive dataset. The study establishes a strong two-stage classification framework. Therefore, noise reduction, ROI cutting, and data enhancement techniques are used for data pre-processing. Second, lesion classification is performed using a ResNet-based convolutional neural network (CNN) architecture. The model is trained and validated on a balanced dataset that contains an equal number of benign and malignant lesion categories. Using accuracy, precision, recall, F1 score and AUC, the system can be assessed and compared to other state-of-art approaches. The findings show that the proposed model has a high level of classification performance and a high level of discriminative ability between melanoma and benign lesions. The ROC curve effectively exhibits the model’s performance and accuracy, and the confusion matrix reveals tendencies to misclassify and where it should be improved. The application of sophisticated preprocessing methods improves model performance, responding to the issues arising from the presence of noise in data. This research is valuable to the field of dermatological diagnostics as it offers a scalable, automated means of skin lesion classification. The proposed framework can be applied clinically to assist dermatologists in the detection of early melanoma and, therefore benefit patients. Subsequent studies will address the development of combined approaches and the improvement of interpretability aids in order to increase the diagnostic accuracy and practical applicability of the methods in clinical practice.

Moving Healthcare AI Support Systems for Visually Detectable Diseases to Constrained Devices

Image classification usually requires connectivity and access to the cloud, which is often limited in many parts of the world, including hard-to-reach rural areas. Tiny machine learning (tinyML) aims to solve this problem by hosting artificial intelligence (AI) assistants on constrained devices, eliminating connectivity issues by processing data within the device itself, without Internet or cloud access. This study explores the use of tinyML to provide healthcare support with low-spec devices in low-connectivity environments, focusing on the diagnosis of skin diseases and the ethical use of AI assistants in a healthcare setting. To investigate this, images of skin lesions were used to train a model for classifying visually detectable diseases (VDDs). The model weights were then offloaded to a Raspberry Pi with a webcam attached, to be used for the classification of skin lesions without Internet access. It was found that the developed prototype achieved a test accuracy of 78% when trained on the HAM10000 dataset, and a test accuracy of 85% when trained on the ISIC 2020 Challenge dataset.

Dual scale light weight cross attention transformer for skin lesion classification.

Skin cancer is rapidly growing globally. In the past decade, an automated diagnosis system has been developed using image processing and machine learning. The machine learning methods require hand-crafted features, which may affect performance. Recently, a convolution neural network (CNN) was applied to dermoscopic images to diagnose skin cancer. The CNN improved its performance through its high-dimension feature extraction capability. However, these methods lack global co-relation of the spatial features. In this study, we design a dual-scale lightweight cross-attention vision transformer network (DSCATNet) that provides global attention to high-dimensional spatial features. In the DSCATNet, we extracted features from different patch sizes and performed cross-attention. The attention from different scales improved the spatial features by focusing on the different parts of the skin lesion. Furthermore, we applied a fusion strategy for the different scale spatial features. After that, enhanced features are fed to the lightweight transformer encoder for global attention. We validated the model superiority on the HAM 10000 and PAD datasets. Furthermore, the model's performance is compared with CNN and ViT-based methods. Our DSCATNet achieved an average kappa and accuracy of 95.84% and 97.80% on the HAM 10000 dataset, respectively. Moreover,the model obtained 94.56% and 95.81% kappa and precision values on the PAD dataset.

FC20 Automating psoriasis area and severity index assessment from real-world images using Vision Transformers

Abstract The psoriasis area and severity index (PASI) is the gold standard for assessing psoriasis severity but is time-consuming, subjective, poorly reproducible, and requires face-to-face interaction between clinician and patient. Automating PASI assessments using deep learning-based image analysis may provide a more objective, efficient measure of disease severity in trial and real-world settings. Most studies have used Convolutional Neural Networks.1 However, Vision Transformers (ViTs) are sophisticated deep learning models that have shown unprecedented capabilities in image analysis, including classifying skin lesions from dermoscopic images and 3D image (computed tomography/magnetic resonance imaging scan) analysis. We sought to develop and evaluate a ViT model for automated PASI using images. Following consent, adults with chronic plaque psoriasis were recruited via a UK national specialized psoriasis service (04/2021–05/2024). Professional studio photographs (five standardized views) and self-taken photographs (three views) were captured. Two blinded, independent face-to-face physician PASI measures were recorded, with the first rater used as the primary reference value. Fitzpatrick Skin Type and Physician Global Assessment (PGA) were also noted. We trained our ‘MultiViT’ model over 300 epochs and tested using an 80:20 split of the participants. MultiViT advances beyond standard ViTs by handling multiple images and resolutions simultaneously. Mean absolute error (MAE) was used to assess prediction accuracy and compare the standard ViT models with the MultiViT. We captured 1109 images from 152 participants; 63% were male, median age of 45.5 years [interquartile range (IQR) 33.75–57.0], median body mass index of 29.68, 84% Fitzpatrick Skin Types I–IV, and 16% V and VI. The majority (76%) had mild-to-moderate disease (PGA ≤3); median PASI was 8.2 (IQR 5.7–12.7). The MultiViT model demonstrated an MAE of 3.88 for predicting PASI from skin images vs. 5.13 for the standard ViT. When stratified by skin type, I–IV performed better than V and VI (MAE of 2.80 and 8.17, respectively), and by disease severity, PGA ≤3 performed better than PGA >3 (MAE of 1.97 and 9.13, respectively). Our findings suggest the MultiViT model offers potential for automated psoriasis severity assessments from skin images, with favourable prediction accuracy despite a small dataset. An expanded and more diverse (Fitzpatrick Skin Type, severity) dataset will enable further refinement, validation and testing of the MultiViT model in trial and real-world settings. Future integration into a smartphone application will facilitate self-monitoring of psoriasis and healthcare efficiency for improved health and cost outcomes in psoriasis.

AI model for accurate skin lesion classification during telemedicine triage of suspicious skin lesions

AI model for accurate skin lesion classification during telemedicine triage of suspicious skin lesions

Skin Lesion Classification: A Deep Learning Approach with Local Interpretable Model-Agnostic Explanations (LIME) for Explainable Artificial Intelligence (XAI)

The classification of skin cancer is crucial as the chance of survival increases significantly with timely and accurate treatment. Convolution Neural Networks (CNNs) have proven effective in classifying skin cancer. However, CNN models are often regarded as "black boxes”, due to the lack of transparency in the decision-making. Therefore, explainable artificial intelligence (XAI) has emerged as a tool for understanding AI decisions. This study employed a CNN model, VGG16, to classify five skin lesion classes. The hyperparameters were adjusted to optimize its classification performance. The best hyperparameter settings were 50 epochs, a 0.1 dropout rate, and the Adam optimizer with a 0.001 learning rate. The VGG16 model demonstrated satisfactory classification performance. The Local Interpretable Model-Agnostic Explanations (LIME) method was implemented as the XAI tool to justify the predictions made by VGG16. The LIME explanation revealed that the correct predictions made by VGG16 were owing to its truthful extraction of the cancer or lesion area, especially for the “vascular lesion” class. Meanwhile, inaccurate classifications were attributed to VGG16 extraction of the background and insignificant parts of the skin as core features. In conclusion, The LIME model allowed visual inspection of the features selected by VGG16, paving the way for improving the CNN model for better feature extraction and classification of skin lesions, offering a promising direction for future research.

DXDSENet-CM model: an ensemble learning model based on depthwise Squeeze-and-Excitation ConvMixer architecture for the classification of multi-class skin lesions

DXDSENet-CM model: an ensemble learning model based on depthwise Squeeze-and-Excitation ConvMixer architecture for the classification of multi-class skin lesions

A Unified Multi-Task Learning Model with Joint Reverse Optimization for Simultaneous Skin Lesion Segmentation and Diagnosis.

Classifying and segmenting skin cancer represent pivotal objectives for automated diagnostic systems that utilize dermoscopy images. However, these tasks present significant challenges due to the diverse shape variations of skin lesions and the inherently fuzzy nature of dermoscopy images, including low contrast and the presence of artifacts. Given the robust correlation between the classification of skin lesions and their segmentation, we propose that employing a combined learning method holds the promise of considerably enhancing the performance of both tasks. In this paper, we present a unified multi-task learning strategy that concurrently classifies abnormalities of skin lesions and allows for the joint segmentation of lesion boundaries. This approach integrates an optimization technique known as joint reverse learning, which fosters mutual enhancement through extracting shared features and limiting task dominance across the two tasks. The effectiveness of the proposed method was assessed using two publicly available datasets, ISIC 2016 and PH2, which included melanoma and benign skin cancers. In contrast to the single-task learning strategy, which solely focuses on either classification or segmentation, the experimental findings demonstrated that the proposed network improves the diagnostic capability of skin tumor screening and analysis. The proposed method achieves a significant segmentation performance on skin lesion boundaries, with Dice Similarity Coefficients (DSC) of 89.48% and 88.81% on the ISIC 2016 and PH2 datasets, respectively. Additionally, our multi-task learning approach enhances classification, increasing the F1 score from 78.26% (baseline ResNet50) to 82.07% on ISIC 2016 and from 82.38% to 85.50% on PH2. This work showcases its potential applicability across varied clinical scenarios.

Enhanced Multi-Class Skin Lesion Classification of Dermoscopic Images Using an Ensemble of Deep Learning Models

This study presents an advanced approach to multi-class skin lesion classification by leveraging an ensemble model comprising the Inception-V3, ResNet-50, and VGG16 architectures. The classification task focuses on categorizing skin lesions into distinct classes, including Melanoma, basal cell carcinoma (BCC), and squamous cell carcinoma (SCC), using the ISIC dataset, a comprehensive collection of dermoscopic images. In order to properly balance the dataset, the oversampling strategy is utilized, as some lesion types are underrepresented due to inherent imbalances in the dataset. By ensuring that the model is trained on a more representative dataset, this balancing improves the algorithm's capacity to categorize all lesion types properly and impartially. By combining the complementary features of ResNet-50, Inception-V3, and VGG16, the ensemble technique improves the overall classification performance. ResNet-50 is chosen for its deep feature extraction capabilities, which help capture fine details in lesion patterns. Inception-V3 is selected for its multi-scale processing, allowing it to effectively analyze lesions at varying resolutions and sizes. VGG16 is included due to its simple yet highly effective architecture for image classification tasks. The ensemble model with data augmentation significantly outperforms individual models in skin lesion classification for both the original and balanced ISIC datasets regarding accuracy, precision, recall, and F1-score. This method offers a robust solution for skin lesion classification, contributing to more accurate and reliable diagnostic tools in dermatology.