International Skin Imaging Collaboration Archive Research Articles

Claude 3 Opus and ChatGPT With GPT-4 in Dermoscopic Image Analysis for Melanoma Diagnosis: Comparative Performance Analysis.

Recent advancements in artificial intelligence (AI) and large language models (LLMs) have shown potential in medical fields, including dermatology. With the introduction of image analysis capabilities in LLMs, their application in dermatological diagnostics has garnered significant interest. These capabilities are enabled by the integration of computer vision techniques into the underlying architecture of LLMs. This study aimed to compare the diagnostic performance of Claude 3 Opus and ChatGPT with GPT-4 in analyzing dermoscopic images for melanoma detection, providing insights into their strengths and limitations. We randomly selected 100 histopathology-confirmed dermoscopic images (50 malignant, 50 benign) from the International Skin Imaging Collaboration (ISIC) archive using a computer-generated randomization process. The ISIC archive was chosen due to its comprehensive and well-annotated collection of dermoscopic images, ensuring a diverse and representative sample. Images were included if they were dermoscopic images of melanocytic lesions with histopathologically confirmed diagnoses. Each model was given the same prompt, instructing it to provide the top 3 differential diagnoses for each image, ranked by likelihood. Primary diagnosis accuracy, accuracy of the top 3 differential diagnoses, and malignancy discrimination ability were assessed. The McNemar test was chosen to compare the diagnostic performance of the 2 models, as it is suitable for analyzing paired nominal data. In the primary diagnosis, Claude 3 Opus achieved 54.9% sensitivity (95% CI 44.08%-65.37%), 57.14% specificity (95% CI 46.31%-67.46%), and 56% accuracy (95% CI 46.22%-65.42%), while ChatGPT demonstrated 56.86% sensitivity (95% CI 45.99%-67.21%), 38.78% specificity (95% CI 28.77%-49.59%), and 48% accuracy (95% CI 38.37%-57.75%). The McNemar test showed no significant difference between the 2 models (P=.17). For the top 3 differential diagnoses, Claude 3 Opus and ChatGPT included the correct diagnosis in 76% (95% CI 66.33%-83.77%) and 78% (95% CI 68.46%-85.45%) of cases, respectively. The McNemar test showed no significant difference (P=.56). In malignancy discrimination, Claude 3 Opus outperformed ChatGPT with 47.06% sensitivity, 81.63% specificity, and 64% accuracy, compared to 45.1%, 42.86%, and 44%, respectively. The McNemar test showed a significant difference (P<.001). Claude 3 Opus had an odds ratio of 3.951 (95% CI 1.685-9.263) in discriminating malignancy, while ChatGPT-4 had an odds ratio of 0.616 (95% CI 0.297-1.278). Our study highlights the potential of LLMs in assisting dermatologists but also reveals their limitations. Both models made errors in diagnosing melanoma and benign lesions. These findings underscore the need for developing robust, transparent, and clinically validated AI models through collaborative efforts between AI researchers, dermatologists, and other health care professionals. While AI can provide valuable insights, it cannot yet replace the expertise of trained clinicians.

A machine learning approach for skin lesion classification on iOS: implementing and optimizing a convolutional transfer learning model with Create ML

The integration of machine learning (ML) into mobile applications presents unique challenges, particularly in resource-constrained environments such as iOS devices. Skin lesion classification is a critical task in dermatology, where accurate and efficient diagnostic tools can significantly aid in early detection of malignant lesions. This study aims to implement a machine learning-based iOS mobile application and develop a binary classification model for skin lesion images to determine whether a lesion is malignant. The research utilized Create ML to develop a convolutional neural network optimized for iOS, employing a transfer learning approach. A logistic regression model was cascaded with the convolutional neural network N to enhance classification accuracy. The model's performance was assessed through various validation metrics, ensuring its robustness and efficiency within the constraints of mobile hardware. A curated dataset from the International Skin Imaging Collaboration archive was used for training and testing. The model achieved an accuracy of 92%, a precision of 90%, a recall of 93%, and an F1-score of 91% in classifying skin lesions. These metrics validate the model's efficacy in identifying malignant lesions. Data curation involved collecting, labeling, and preparing a dataset from publicly available sources, ensuring the inclusion of diagnostically relevant features. The final model was integrated into an iOS application using Core ML and Vision frameworks. The developed application demonstrates reliable performance in classifying skin lesions with high accuracy on iOS devices. The inclusion of comprehensive performance metrics justifies the efficacy of the proposed approach. Future work will explore enhancements in model architecture and object detection capabilities to further improve diagnostic precision and application usability.

Prediction of melanoma Breslow thickness using deep transfer learning algorithms.

The distinction between in situ melanoma (MIS) and invasive melanoma is challenging even for expert dermatologists. The use of pretrained convolutional neural networks (CNNs) as ancillary decision systems needs further research. To develop, validate and compare three deep transfer learning (DTL) algorithms to predict MIS vs. invasive melanoma and melanoma with a Breslow thickness (BT) of < 0.8 mm vs. ≥ 0.8 mm. A dataset of 1315 dermoscopic images of histopathologically confirmed melanomas was created from Virgen del Rocio University Hospital and open repositories of the International Skin Imaging Collaboration archive and PolesieS et al.(Dermatol Pract Concept 2021; 11:e2021079). The images were labelled as MIS or invasive melanoma and < 0.8 mmor ≥ 0.8 mm of BT. We conducted three trainings, and overall means for receiver operating characteristic (ROC) curves, sensitivity, specificity, positive and negative predictive value, and balanced diagnostic accuracy outcomes were evaluated on the test set with ResNetV2, EfficientNetB6 and InceptionV3. The results of 10 dermatologists were compared with the algorithms. Grad-CAM gradient maps were generated, highlighting relevant areas considered by the CNNs within the images. EfficientNetB6 achieved the highest diagnostic accuracy for the comparison between MIS vs. invasive melanoma (61%) and BT < 0.8 mm vs. ≥ 0.8 mm (75%). For the BT comparison, ResNetV2 with an area under the ROC curve of 0.76 and InceptionV3 with an area under the ROC curve of 0.75, outperformed the results obtained by the dermatologist group with an area under the ROC curve of 0.70. EfficientNetB6 recorded the best prediction results, outperforming the dermatologists for the comparison of 0.8 mm of BT. DTL could be an ancillary aid to support dermatologists' decisions in the near future.

Classification of Skin Cancer Images Using Convolutional Neural Networks

Skin cancer is the most common human malignancy according to American Cancer Society. It is primarily diagnosed visually, starting with an initial clinical screening and followed potentially by dermoscopic (related to skin) analysis, a biopsy and histopathological examination. Skin cancer occurs when errors (mutations) occur in the DNA of skin cells. The mutations cause cells to grow out of control and form a mass of cancer cells. The aim of this study was to try to classify images of skin lesions with the help of Convolutional Neural Networks. Deep neural networks show humongous potential for image classification while taking into account the large variability exhibited by the environment. Here, we trained images on the basis of pixel values and classified them on the basis of disease labels. The dataset was acquired from an Open Source Kaggle Repository (Kaggle Dataset) which itself was acquired from ISIC (International Skin Imaging Collaboration) archive. The training was performed on multiple models accompanied with Transfer Learning. The highest model accuracy achieved was over 86.65%. The dataset used is publicly available to ensure credibility and reproducibility of the aforementioned result.

Classification of Skin Cancer Images Using Convolutional Neural Networks

Skin cancer is the most common human malignancy(American Cancer Society) which is primarily diagnosed visually, starting with an initial clinical screening and followed potentially by dermoscopic(related to skin) analysis, a biopsy and histopathological examination. Skin cancer occurs when errors (mutations) occur in the DNA of skin cells. The mutations cause the cells to grow out of control and form a mass of cancer cells. The aim of this study was to try to classify images of skin lesions with the help of convolutional neural networks. The deep neural networks show humongous potential for image classification while taking into account the large variability exhibited by the environment. Here we trained images based on the pixel values and classified them on the basis of disease labels. The dataset was acquired from an Open Source Kaggle Repository(Kaggle Dataset)which itself was acquired from ISIC(International Skin Imaging Collaboration) Archive. The training was performed on multiple models accompanied with Transfer Learning. The highest model accuracy achieved was over 86.65%. The dataset used is publicly available to ensure credibility and reproducibility of the aforementioned result.

Characterizing Malignant Melanoma Clinically Resembling Seborrheic Keratosis Using Deep Knowledge Transfer.

Simple SummaryMalignant melanomas (MMs) with aypical clinical presentation constitute a diagnostic pitfall, and false negatives carry the risk of a diagnostic delay and improper disease management. Among the most common, challenging presentation forms of MMs are those that clinically resemble seborrheic keratosis (SK). On the other hand, SK may mimic melanoma, producing ‘false positive overdiagnosis’ and leading to needless excisions. The evolving efficiency of deep learning algorithms in image recognition and the availability of large image databases have accelerated the development of advanced computer-aided systems for melanoma detection. In the present study, we used image data from the International Skin Image Collaboration archive to explore the capacity of deep knowledge transfer in the challenging diagnostic task of the atypical skin tumors of MM and SK.Malignant melanomas resembling seborrheic keratosis (SK-like MMs) are atypical, challenging to diagnose melanoma cases that carry the risk of delayed diagnosis and inadequate treatment. On the other hand, SK may mimic melanoma, producing a ‘false positive’ with unnecessary lesion excisions. The present study proposes a computer-based approach using dermoscopy images for the characterization of SΚ-like MMs. Dermoscopic images were retrieved from the International Skin Imaging Collaboration archive. Exploiting image embeddings from pretrained convolutional network VGG16, we trained a support vector machine (SVM) classification model on a data set of 667 images. SVM optimal hyperparameter selection was carried out using the Bayesian optimization method. The classifier was tested on an independent data set of 311 images with atypical appearance: MMs had an absence of pigmented network and had an existence of milia-like cysts. SK lacked milia-like cysts and had a pigmented network. Atypical MMs were characterized with a sensitivity and specificity of 78.6% and 84.5%, respectively. The advent of deep learning in image recognition has attracted the interest of computer science towards improved skin lesion diagnosis. Open-source, public access archives of skin images empower further the implementation and validation of computer-based systems that might contribute significantly to complex clinical diagnostic problems such as the characterization of SK-like MMs.

Novel strategy for applying hierarchical density-based spatial clustering of applications with noise towards spectroscopic analysis and detection of melanocytic lesions.

Advancements in dermoscopy techniques have elucidated identifiable characteristics of melanoma which revolve around the asymmetrical constitution of melanocytic lesions consequent of unfettered proliferative growth as a malignant lesion. This study explores the applications of hierarchical density-based spatial clustering of applications with noise (HDBSCAN) in terms of the direct diagnostic implications of applying agglomerative clustering in the spectroscopic analysis of malignant melanocytic lesions and benign dermatologic spots. 100 images of benign (n = 50) and malignant moles (n = 50) were sampled from the International Skin Imaging Collaboration Archive and processed through two separate Python algorithms. The first of which deconvolutes the three-digit tupled integer identifiers of pixel color in image composition into three separate matrices corresponding to the red, green and blue color channel. Statistical characterization of integer variance was utilized to determine the optimal channel for comparative analysis between malignant and benign image groups. The second applies HDBSCAN to the matrices, identifying agglomerative clustering in the dataset. The results indicate the potential diagnostic applications of HDBSCAN analysis in fast-processing dermoscopy, as optimization of clustering parameters according to a binary search strategy produced an accuracy of 85% in the classification of malignant and benign melanocytic lesions.

Multi skin lesions classification using fine-tuning and data-augmentation applying NASNet.

Skin lesions are one of the typical symptoms of many diseases in humans and indicative of many types of cancer worldwide. Increased risks caused by the effects of climate change and a high cost of treatment, highlight the importance of skin cancer prevention efforts like this. The methods used to detect these diseases vary from a visual inspection performed by dermatologists to computational methods, and the latter has widely used automatic image classification applying Convolutional Neural Networks (CNNs) in medical image analysis in the last few years. This article presents an approach that uses CNNs with a NASNet architecture to recognize in a more accurate way, without segmentation, eight skin diseases. The model was trained end-to-end on Keras with augmented skin diseases images from the International Skin Imaging Collaboration (ISIC). The CNN architectures were initialized with weight from ImageNet, fine-tuned in order to discriminate well among the different types of skin lesions, and then 10-fold cross-validation was applied. Finally, some evaluation metrics are calculated as accuracy, sensitivity, and specificity and compare with other CNN trained architectures. This comparison shows that the proposed system offers higher accuracy results, with a significant reduction on the training paraments. To the best of our knowledge and based in the state-of-art recompiling in this work, the application of the NASNet architecture training with skin image lesion from ISIC archive for multi-class classification and evaluated by cross-validation, represents a novel skin disease classification system.

The Role in Teledermoscopy of an Inexpensive and Easy-to-Use Smartphone Device for the Classification of Three Types of Skin Lesions Using Convolutional Neural Networks.

Background. The use of teledermatology has spread over the last years, especially during the recent SARS-Cov-2 pandemic. Teledermoscopy, an extension of teledermatology, consists of consulting dermoscopic images, also transmitted through smartphones, to remotely diagnose skin tumors or other dermatological diseases. The purpose of this work was to verify the diagnostic validity of images acquired with an inexpensive smartphone microscope (NurugoTM), employing convolutional neural networks (CNN) to classify malignant melanoma (MM), melanocytic nevus (MN), and seborrheic keratosis (SK). Methods. The CNN, trained with 600 dermatoscopic images from the ISIC (International Skin Imaging Collaboration) archive, was tested on three test sets: ISIC images, images acquired with the NurugoTM, and images acquired with a conventional dermatoscope. Results. The results obtained, although with some limitations due to the smartphone device and small data set, were encouraging, showing comparable results to the clinical dermatoscope and up to 80% accuracy (out of 10 images, two were misclassified) using the NurugoTM demonstrating how an amateur device can be used with reasonable levels of diagnostic accuracy. Conclusion. Considering the low cost and the ease of use, the NurugoTM device could be a useful tool for general practitioners (GPs) to perform the first triage of skin lesions, aiding the selection of lesions that require a face-to-face consultation with dermatologists.

Association between the dermoscopic morphology of peripheral globules and melanocytic lesion diagnosis.

The presence of peripheral globules is associated with enlarging melanocytic lesions; however, there are numerous patterns of peripheral globules distribution and it remains unknown whether specific patterns can help differentiate enlarging naevi from melanoma. To investigate whether morphological differences exist between the peripheral globules seen in different subsets of naevi and in melanoma. A cross-sectional study of clinical notes that mentioned peripheral globules, in addition to all melanoma images with peripheral globules on the International Skin Imaging Collaboration archive. Dermoscopic images were reviewed and annotated. Associations between diagnosis and categorical features were measured with odds ratios. Non-parametric tests were used for continuous factors. 184 lesions with peripheral globules from our clinic were included in the analysis; only 6 of these proved to be melanoma. 109 melanomas with peripheral globules from the International Skin Imaging Collaboration archive were added to the analysis. Melanomas were more common on the extremities and among older individuals. Melanomas were more likely to display atypical, tiered and/or focal peripheral globules. Only 5% of melanomas lacked dermoscopic melanoma-specific structures compared to 48% of naevi. Melanocytic lesions with atypical or asymmetrically distributed peripheral globules, especially when located on the extremities, should raise suspicion for malignancy. Melanocytic lesions with typical and symmetrically distributed peripheral globules, and with no other concerning dermoscopic features, are unlikely to be malignant.

Design and Implement of Deep Learning Model to Detect the Melanoma

Detecting Skin lesions on the human body is a big task to the doctors in the initial stage because of the low contrast on the body. This skin cancer can be occur due to sun rays. If the disease cannot detect in early stage, there it may cause death to human lives. Here there are some algorithms to predict the melanoma using deep learning techniques. ISIC International Skin Imaging Collaboration Archive set where it provides various images of melanoma and non-melanoma. There are so many challenges to identify the image with melanoma and non-melanoma types of skin cancer. In this paper we applied hair removal algorithm and k-means clustering algorithm where to remove unwanted substances from the original images. To classify the melanoma and non-melanoma skin cancer, this paper proposed prediction process and sequential CNN architecture.