Melanoma Lesions Research Articles

Dual-energy CT iodine concentration as a biomarker for immunotherapy treatment response in metastatic melanoma and renal cell carcinoma patients.

To investigate the predictive value of tumor iodine concentration obtained with dual-energy CT (DECT) for treatment response in patients treated with immune checkpoint inhibitors (ICI). Retrospective single-center study of consecutive metastatic melanoma and renal cell carcinoma (RCC) patients undergoing first-line ICI treatment. The iodine concentration measurement time points include prior to initiation of therapy (baseline [BL]), after initiation (follow-up [FU1]), and either time point nearest to 12 months or at time of progression (final follow-up [FFU]). Target lesion DECT-based whole-volume tumor normalized iodine concentration average (NICave) and size measurements were obtained. Reference standard was individual lesion FFU status categorized as responders or nonresponders per RECIST 1.1. Logistic regression model assessed NICave change and FU1 lesion response as predictors of FFU lesion outcome. Model's performance was summarized with AUC. Intraclass correlation coefficient (ICC) summarized inter-rater agreement of NICave. Forty-six patients were included (mean age 61 ± 11 years, 12 women; 16 melanoma). Sixty-four of 175 target lesions were confirmed nonresponders at FFU. In a multivariable model, lesion status at FU1 (odds ratio [OR]: 27.4, p < 0.001) and changes in NICave from BL to FU1 (OR: 2.42 per 1-SD increase, p = 0.019) were significant predictors of lesion status at FFU. The model's AUC was 0.86 (95% CI: 0.76-0.93). Inter-rater reliability of NICave was 0.98 (95% CI: 0.97-0.99). Changes in iodine concentration from baseline to first follow-up improve identification of delayed responding metastatic melanoma and RCC lesions treated with immune checkpoint inhibitor, initially classified as nonresponders by size change. Question How can pseudoprogression/delayed treatment response in metastatic renal cell carcinoma (RCC) and melanoma patients on first-line immune checkpoint inhibitors be accurately identified? Findings Combining iodine concentration change from Dual-energy CT (baseline to first follow-up) with RECIST-based lesion size change improved prediction of final lesion outcome. Clinical relevance DECT-based whole-volume tumor iodine concentration for target lesions is useful as a predictive imaging biomarker for distinguishing delayed response from true progression in patients with metastatic RCC and melanoma treated with first-line immune checkpoint inhibitors.

Minimal sourced and lightweight federated transfer learning models for skin cancer detection

One of the most fatal diseases that affect people is skin cancer. Because nevus and melanoma lesions are so similar and there is a high likelihood of false negative diagnoses challenges in hospitals. The aim of this paper is to propose and develop a technique to classify type of skin cancer with high accuracy using minimal resources and lightweight federated transfer learning models. Here minimal resource based pre-trained deep learning models including EfficientNetV2S, EfficientNetB3, ResNet50, and NasNetMobile have been used to apply transfer learning on data of shape. To compare with applied minimal resource transfer learning, same methodology has been applied using best identified model i.e. EfficientNetV2S for images of shape. The identified minimal and lightweight resource based EfficientNetV2S with images of shape have been applied for federated learning ecosystem. Both, identically and non-identically distributed datasets of shape have been applied and analyzed through federated learning implementations. The results have been analyzed to show the impact of low-pixel images with non-identical distributions over clients using parameters such as accuracy, precision, recall and categorical losses. The classification of skin cancer shows an accuracy of IID 89.83% and Non-IID 90.64%.

BRD4-targeted photodegradation nanoplatform for light activatable melanoma therapy.

The targeted protein degradation (TPD) strategy modulates tumor growth pathways by degrading proteins of interest (POIs) and has reshaped anti-tumor drug research and development. Recently, the emergence of photodegradation-targeting chimeras (PDTACs) and laser irradiation at specific sites enables precise spatiotemporal controllability of TPD. Capitalizing on the advances of PDTACs, herein, we report a nanoplatform for efficiently delivering PDTAC molecule for photodegradation of bromodomain-containing protein 4 (BRD4) proteins, the key activators of oncogenic transcription. The PDTAC molecule, named as PPa-JQ1, is synthesized through the covalent attachment of the BRD4-targeting ligand JQ1-acid, to the photosensitizer pyropheophorbide-a (PPa), utilizing a 1,6-hexanediamine linker. The PPa-JQ1 is further encapsulated by human serum albumin (HSA) to obtain the HSA@PPa-JQ1 nanoplatform, which facilitates targeted and efficacious delivery to melanoma lesions. Both in vitro and in vivo therapeutic outcomes demonstrate that HSA@PPa-JQ1 can efficiently generate reactive oxygen species (ROS) to degrade BRD4 upon light irradiation, which eventually induces tumor death. Our study represents the first case to validate the anti-tumor therapeutic efficacy of PDTACs by systemic administration, providing the foundation for further application of PDTACs.

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.

Effect of patient-contextual skin images in human- and artificial intelligence-based diagnosis of melanoma: Results from the 2020 SIIM-ISIC melanoma classification challenge.

While the high accuracy of reported AI tools for melanoma detection is promising, the lack of holistic consideration of the patient is often criticized. Along with medical history, a dermatologist would also consider intra-patient nevi patterns, such that nevi that are different from others on a given patient are treated with suspicion. To evaluate whether patient-contextual lesion-images improves diagnostic accuracy for melanoma in a dermoscopic image-based AI competition and a human reader study. An international online AI competition was held in 2020. The task was to classify dermoscopy images as melanoma or benign lesions. A multi-source dataset of dermoscopy images grouped by patient were provided, and additional use of public datasets was permitted. Competitors were judged on area under the receiver operating characteristic (AUROC) on a private leaderboard. Concurrently, a human reader study was hosted using a subset of the test data. Participants gave their initial diagnosis of an index case (melanoma vs. benign) and were then presented with seven additional lesion-images of that patient before giving a second prediction of the index case. Outcome measures were sensitivity and specificity. The top 50 of 3308 AI competition entries achieved AUROC scores ranging from 0.943 to 0.949. Few algorithms considered intra-patient lesion patterns and instead most evaluated images independently. The median sensitivity and specificity of human readers before receiving contextual images were 60.0% and 86.7%, and after were 60.0% and 85.7%. Human and AI algorithm performance varied by image source. This study provided an open-source state-of-the-art algorithm for melanoma detection that has been evaluated at multiple centres. Patient-contextual images did not positively impact performance of AI algorithms or human readers. Providing seven contextual images and no total body image may have been insufficient to test the applicability of the intra-patient lesion patterns.

Higher subfoveal choroidal thickness in choroidal melanomas than in choroidal nevi.

To compare subfoveal choroidal thickness (SFCT) between eyes with choroidal melanoma and choroidal nevi. Retrospective study of 126 consecutive patients in a tertiary ocular oncology center. Eyes with tumors located less than two disc-diameters from the fovea were excluded. In eyes with naevi, factors of potential transformation into melanoma were recorded (orange pigment, subretinal fluid, thickness >2 mm, diameter >5 mm, ultrasound hollowness). SFCT was assessed by 3 independent observers on horizontal spectral-domain OCT scans. Sixty-seven eyes with choroidal melanoma and 59 eyes with choroidal nevi were included. The melanoma and nevi groups did not differ in gender (P=0.14) nor age (P=0.34). There was a very good agreement between the three independent observers for SFCT measurements (intraclass correlation coefficient=0.89). Mean SFCT was higher in melanomas (294.3±89.9 µm) than naevi (260.3±76.7 µm) (P=0.013), and the difference remained significant between melanomas and 28 naevi with ≥2 growth risk factors (256.3±77.0 µm) (P=0.027). In a multivariate model, the significant contributors to SFCT were presence of melanoma (P=0.004), younger age (P<0.0001) and shorter lesion distance to the fovea (P=0.016). SFCT may reflect the interplay between melanocytic tumors and their choroidal microenvironment. Its clinical utility should be explored in future studies.

A noninvasive ultrasound vibro-elastography technique for assessing ocular lesions

ObjectiveThis research aims to develop a noninvasive ultrasound vibro-elastography technique for assessing ocular lesions including intraocular melanoma and nevus lesions. MethodWave speed (WS) was noninvasively measured in the lesions at three different frequencies (i.e., 100 Hz, 150 Hz, and 200 Hz). The nearby normal tissue of choroid and sclera was also analyzed as controls. Viscoelasticity of these tissues was analyzed using the wave speed dispersion curve and the Voigt model. ResultsIn this pilot study of 10 cases (5 melanomas vs. 5 nevus) with ages (mean ± SD) of (53.2 ± 6.82) vs. (72.2 ± 7.71) yo, the WS (m/s), elasticity (kPa) and viscosity (Pa.s) of lesion (melanoma vs. nevus) locations of the eye (i.e., WS @100 Hz: 3.63 vs. 3.09 (m/s), @150 Hz: 4.10 vs. 3.66 (m/s), @200 Hz: 4.78 vs. 4.07 (m/s); elasticity: 9.59 vs. 6.89 (kPa); viscosity: 12.46 vs. 9.26(Pa.s)) were analyzed. There were significant differences of WS ratio (WS close to the surface and WS inside the tumor) between the nevus and melanoma of all three frequencies (i.e., nevus vs. melanoma; @100 Hz: 1.59 vs. 2.95, p = 0.0285; @150 Hz: 1.58 vs. 3.53, p = 0.0054; @200 Hz: 1.70 vs. 3.31, p = 0.0124). The melanoma lesions are stiffer than the nevus lesions. It also shows that the lesion tissues are stiffer than the control tissues. However, there were no significant differences in WS, elasticity, viscosity among melanoma lesions, nevus lesions, and control tissues. ConclusionsWe demonstrate for the first time that noninvasive ultrasound vibro-elastography can be used for assessing ocular lesions. The results show that lesion tissues are stiffer than control tissues. They also show that melanoma lesions are stiffer than the nevus lesions. We plan to study more ocular lesion patients and improve the specificity and sensitivity of wave speed or viscoelasticity between lesions and controls. SignificanceUltrasound vibro-elastography is an innovative and noninvasive technique for assessing ocular lesions.

Advancing dermoscopy through a synthetic hair benchmark dataset and deep learning-based hair removal.

Early detection of melanoma is crucial for improving patient outcomes, and dermoscopy is a critical tool for this purpose. However, hair presence in dermoscopic images can obscure important features, complicating the diagnostic process. Enhancing image clarity by removing hair without compromising lesion integrity can significantly aid dermatologists in accurate melanoma detection. We aim to develop a novel synthetic hair dermoscopic image dataset and a deep learning model specifically designed for hair removal in melanoma dermoscopy images. To address the challenge of hair in dermoscopic images, we created a comprehensive synthetic hair dataset that simulates various hair types and dimensions over melanoma lesions. We then designed a convolutional neural network (CNN)-based model that focuses on effective hair removal while preserving the integrity of the melanoma lesions. The CNN-based model demonstrated significant improvements in the clarity and diagnostic utility of dermoscopic images. The enhanced images provided by our model offer a valuable tool for the dermatological community, aiding in more accurate and efficient melanoma detection. The introduction of our synthetic hair dermoscopic image dataset and CNN-based model represents a significant advancement in medical image analysis for melanoma detection. By effectively removing hair from dermoscopic images while preserving lesion details, our approach enhances diagnostic accuracy and supports early melanoma detection efforts.

DDDR-07. FUNCTIONAL GENOMICSIN VIVO USING CRISPR/CAS9 IDENTIFIES POTENTIAL METABOLIC VULNERABILITIES WITHIN MELANOMA BRAIN METASTASIS

Abstract BACKGROUND Inhibition of the oxidative phosphorylation (OxPhos) pathway in mice suppressed the growth of melanoma brain metastasis but not extracranial or primary skin melanoma. This indicates that melanoma brain metastasis may have targetable metabolic vulnerabilities. Human clinical trials, however showed toxicities following OxPhos pathway inhibition. This suggests that more selective metabolic targets are needed to minimize off target effects. METHODS We designed a focused CRISPR guide RNA (gRNA) lentiviral knockout library targeting 106 metabolism genes significantly enriched in human melanoma brain metastasis (n=88) relative to extracranial (n=50) or primary skin melanoma (n=52). We infected mouse B16F10 melanoma cell lines with the CRISPR library and implanted 1X106 cells subcutaneously or intracerebrally into male (n=3) and female (n=3) C57/Bl6 mice. DNA from cell lines and tumor samples were sent for next generation sequencing. Candidate metabolism genes that may drive growth of melanoma lesion within the brain were identified based on loss of gRNA copies in brain relative to skin. RESULTS The CRISPR/Cas9 in vivo screen identified metabolic genes whose knockout resulted in loss of gRNA clones within established B16F10 melanoma brain lesions relative to skin lesions. These putative metabolism targets included specific components in the oxidative phosphorylation pathway, choline metabolism, and ubiquitin cascade activation. CONCLUSION Using a focused CRISPR/Cas9 in vivo screen, we uncovered potential metabolic targets within established mouse melanoma brain lesions. vWe are currently performing the in vivo screen in melanoma mouse models with spontaneous brain metastasis. Top 3 hits across all models will be validated individually prior to a compound screen for inhibitors.

Categorical classification of skin cancer using a weighted ensemble of transfer learning with test time augmentation

Skin cancer is the abnormal development of cells on the surface of the skin and is one of the most fatal diseases in humans. It usually appears in locations that are exposed to the sun, but can also appear in areas that are not regularly exposed to the sun. Due to the striking similarities between benign and malignant lesions, skin cancer detection remains a problem, even for expert dermatologists. Considering the inability of dermatologists to diagnose skin cancer accurately, a convolutional neural network (CNN) approach was used for skin cancer diagnosis. However, the CNN model requires a significant number of image datasets for better performance; thus, image augmentation and transfer learning techniques have been used in this study to boost the number of images and the performance of the model, because there are a limited number of medical images. This study proposes an ensemble transfer-learning-based model that can efficiently classify skin lesions into one of seven categories to aid dermatologists in skin cancer detection: (i) actinic keratoses, (ii) basal cell carcinoma, (iii) benign keratosis, (iv) dermatofibroma, (v) melanocytic nevi, (vi) melanoma, and (vii) vascular skin lesions. Five transfer learning models were used as the basis of the ensemble: MobileNet, EfficientNetV2B2, Xception, ResNext101, and DenseNet201. In addition to the stratified 10-fold cross-validation, the results of each individual model were fused to achieve greater classification accuracy. An annealing learning rate scheduler and test-time augmentation (TTA) were also used to increase the performance of the model during the training and testing stages. A total of 10,015 publicly available dermoscopy images from the HAM10000 (Human Against Machine) dataset, which contained samples from the seven common skin lesion categories, were used to train and evaluate the models. The proposed technique attained 94.49% accuracy on the dataset. These results suggest that this strategy can be useful for improving the accuracy of skin cancer classification. However, the weighted average of f1-score, recall, and precision were obtained to be 94.68%, 94.49%, and 95.07%, respectively.

Explainable AI framework for skin cancer classification, and melanoma segmentation

Identification and classification of skin diseases are two critical challenges faced in diagnosing and treating patients suffering from them. Deep learning models have been created to best identify and classify skin problems to detect and identify them correctly and effectively. This paper proposes a comprehensive framework for accurate skin cancer prediction, classification, and melanoma surgical lesion extraction. Primarily, a comprehensive extraction method leveraging the unique approach of DenseNet201 and the Local Interpretable Model-Agnostic Explanation offers accurate insight into model decision making and prediction. Secondly, the model has a mid-extraction phase that utilizes advanced convolutional neural network levels to detect the boundaries of melanoma lesions correctly. The framework results in terms of IOU, accuracy, precision, recall, and other metrics compared to existing models like FPN, MAN, and U-Net. The framework presented in our model is smart, easy to use, and can provide functional and accurate information, which means it can be used in clinical practice.

Ensemble feature selection and tabular data augmentation with generative adversarial networks to enhance cutaneous melanoma identification and interpretability

BackgroundCutaneous melanoma is the most aggressive form of skin cancer, responsible for most skin cancer-related deaths. Recent advances in artificial intelligence, jointly with the availability of public dermoscopy image datasets, have allowed to assist dermatologists in melanoma identification. While image feature extraction holds potential for melanoma detection, it often leads to high-dimensional data. Furthermore, most image datasets present the class imbalance problem, where a few classes have numerous samples, whereas others are under-represented.MethodsIn this paper, we propose to combine ensemble feature selection (FS) methods and data augmentation with the conditional tabular generative adversarial networks (CTGAN) to enhance melanoma identification in imbalanced datasets. We employed dermoscopy images from two public datasets, PH2 and Derm7pt, which contain melanoma and not-melanoma lesions. To capture intrinsic information from skin lesions, we conduct two feature extraction (FE) approaches, including handcrafted and embedding features. For the former, color, geometric and first-, second-, and higher-order texture features were extracted, whereas for the latter, embeddings were obtained using ResNet-based models. To alleviate the high-dimensionality in the FE, ensemble FS with filter methods were used and evaluated. For data augmentation, we conducted a progressive analysis of the imbalance ratio (IR), related to the amount of synthetic samples created, and evaluated the impact on the predictive results. To gain interpretability on predictive models, we used SHAP, bootstrap resampling statistical tests and UMAP visualizations.ResultsThe combination of ensemble FS, CTGAN, and linear models achieved the best predictive results, achieving AUCROC values of 87% (with support vector machine and IR=0.9) and 76% (with LASSO and IR=1.0) for the PH2 and Derm7pt, respectively. We also identified that melanoma lesions were mainly characterized by features related to color, while not-melanoma lesions were characterized by texture features.ConclusionsOur results demonstrate the effectiveness of ensemble FS and synthetic data in the development of models that accurately identify melanoma. This research advances skin lesion analysis, contributing to both melanoma detection and the interpretation of main features for its identification.

A Multifunctional Nanoparticle Dual Loading with Chlorin e6 and STING Agonist for Combinatorial Therapy of Melanoma.

Photodynamic therapy (PDT) is a noninvasive therapeutic approach that is effective in killing primary tumors with minimal surgical trauma, but its usage in metastatic lesions of melanoma is restricted by spatial limitations. Recently, stimulator of interferon genes (STING) agoinst-mediated innate immunity can activate the STING pathway and further promote dendritic cell (DC) maturation, tumor-specific cytotoxic T lymphocyte, and natural killer cell infiltration and has emerged as a promising approach for cancer therapy. Herein, the authors intriduce facile nanoparticles named HTCS, which can co-deliver STING agonist (2'3'-cGAMP) and a mitochondrial targeting modified photosensitizer (TPP-PEI-Ce6). While HTCS were intravenously injected to mice, they were endocytosed into tumor cells through hyaluronic acid-mediated active targeting. Thereafter, TPP-PEI-Ce6 was delivered to mitochondria to generate a large variety of reactive oxygen species and killed tumor cells effectively. Then the tumor cell debris further gave rise to immunogenic cell death, which played a role in immunosuppression. Furthermore, 2'3'-cGAMP contained in cell debris activated the STING pathway to promote the release of inflammatory cytokines and the maturation of DCs. As a consequence, the HTCS could achieve photodynamic multiple immunotherapy for melanoma. This work demonstrates multifunctional nanoparticles that efficiently inhibit tumors by PDT and reversing their immunosuppression to realize a versatile therapeutic strategy.

Vital Characteristics Cellular Neural Network (VCeNN) for Melanoma Lesion Segmentation: A Biologically Inspired Deep Learning Approach.

Cutaneous melanoma is a highly lethal form of cancer. Developing a medical image segmentation model capable of accurately delineating melanoma lesions with high robustness and generalization presents a formidable challenge. This study draws inspiration from cellular functional characteristics and natural selection, proposing a novel medical segmentation model named the vital characteristics cellular neural network. This model incorporates vital characteristics observed in multicellular organisms, including memory, adaptation, apoptosis, and division. Memory module enables the network to rapidly adapt to input data during the early stages of training, accelerating model convergence. Adaptation module allows neurons to select the appropriate activation function based on varying environmental conditions. Apoptosis module reduces the risk of overfitting by pruning neurons with low activation values. Division module enhances the network's learning capacity by duplicating neurons with high activation values. Experimental evaluations demonstrate the efficacy of this model in enhancing the performance of neural networks for medical image segmentation. The proposed method achieves outstanding results across numerous publicly available datasets, indicating its potential to contribute significantly to the field of medical image analysis and facilitating accurate and efficient segmentation of medical imagery. The proposed method achieves outstanding results across numerous publicly available datasets, with an F1 score of 0.901, Intersection over Union of 0.841, and Dice coefficient of 0.913, indicating its potential to contribute significantly to the field of medical image analysis and facilitating accurate and efficient segmentation of medical imagery.

Deep learning approach for skin melanoma and benign classification using empirical wavelet decomposition.

Melanoma is a malignant skin cancer that causes high mortality. Early detection of melanoma can save patients' lives. The features of the skin lesion images can be extracted using computer techniques to differentiate early between melanoma and benign skin lesions. A new model of empirical wavelet decomposition (EWD) based on tan hyperbolic modulated filter banks (THMFBs) (EWD-THMFBs) was used to obtain the features of skin lesion images by MATLAB software. The EWD-THMFBs model was compared with the empirical short-time Fourier decomposition method based on THMFBs (ESTFD-THMFBs) and the empirical Fourier decomposition method based on THMFBs (EFD-THMFBs). The accuracy rates obtained for EWD-THMFBs, ESTFD-THMFBs, and EFD-THMFBs models were 100%, 98.89%, and 83.33%, respectively. The area under the curve (AUC) was 1, 0.97, and 0.91, respectively. The EWD-THMFBs model performed best in extracting features from skin lesion images. This model can be programmed on a mobile to detect skin lesions in rural areas by a nurse before consulting a dermatologist.

Primary Dermal Melanoma

Primary dermal melanoma (PDM) is a rare subtype of melanoma with an estimated incidence of less than 1%. PDM presents entirely in the dermis or subcutis and histopathologically mimics cutaneous metastases of melanoma due to its lack of connection to the overlying epidermis. A thorough history and examination, including imaging evaluation for metastatic disease, will reveal no prior history or concurrent lesion of primary cutaneous or metastatic melanoma. Despite its histopathologic similarities to melanoma metastasis, PDM is associated with an unexpectedly favorable prognosis. This review discusses the clinical and histopathologic features of PDM as a distinct subtype of melanoma, as well as the current approach to clinical staging and management.

SEDARU-net: a squeeze-excitation dilated based residual U-Net with attention mechanism for automatic melanoma lesion segmentation

SEDARU-net: a squeeze-excitation dilated based residual U-Net with attention mechanism for automatic melanoma lesion segmentation

Subungual melanoma: molecular analysis of 31 cases from early stage to invasive melanoma.

The distinction between the benign subungual melanocytic lesions and an early lesion of subungual melanoma (SUM) remains a diagnostic challenge. We evaluated the routine diagnostic utility of array Comparative Genomic Hybridization (aCGH) to detect whole-genome copy number variations (CNV) as well as targeted next-generation sequencing (NGS) in SUM. This retrospective study included 20 cases of in situ SUM and 11 cases of invasive SUM. Analysis by aCGH detected common oncogene amplifications in all but one case of invasive SUM (n = 10) and in all cases of in situ SUM with a melanocyte count (MC) >45/mm (n = 4 true positive) and the average number of CNV was 8.5. Thirteen remaining cases of in situ SUM gave false negative results (n = 13), owing to a lack of sufficient melanocytes to analyse (median MC of 35.35; range: 10.16-39.5). Molecular analysis failed in four cases (three in situ SUM and one invasive SUM) due to insufficient amounts of DNA. Across the whole cohort, the sensitivity of aCGH was 52%, but when adjusting the cutoff to MC >45/mm, the sensitivity was 93%. Targeted NGS was less informative than aCGH analyses in our series of SUM. To distinguish malignant from benign lesions, especially in situ SUM versus atypical lentiginous melanocytic proliferations, aCGH analysis should be performed when the MC is above 45 melanocytes per linear millimetre. This pangenomic method can detect oncogene amplifications, as well as a number of CNV >3, which strongly support the diagnosis of malignancy.

Automated Standardization of Melanin Bleaching Procedures of Heavily Pigmented Melanocytic Lesions With Low-Concentration Hydrogen Peroxide on an Automated Platform.

Melanin is a natural pigment in the human body that is primarily found in the skin and hair. It protects the skin from damage by ultraviolet radiation. Although this pigment plays a crucial role in protecting the human body, it represents a challenge for pathologists to evaluate highly pigmented tissue samples from melanoma or pigmented skin lesions. Abundant melanin may obscure tissue morphology, which makes it very difficult for pathologists to make a differential diagnosis. Melanin pigment is brown-to-black and granular, and its distribution is often uneven in tissues. The presence of these pigments can complicate the analysis of immunohistochemistry (IHC) for 2 reasons. First, they have a direct physical masking effect on antigen-antibody interactions. Second, 3,3-diaminobenzidine, the most commonly used chromogen, has a brown color that is difficult to distinguish from melanin pigment. Therefore, melanin bleaching has become a crucial step in handling pigmented melanocytic lesions. Bleaching techniques aid pathologists in histopathologic examination of melanin-rich tissue. In this study, we integrated melanin bleaching and IHC on an automated IHC platform to set up a rapid and fully automated procedure. Bleaching steps were performed before antigen retrieval. Samples were treated with 1% hydrogen peroxide solution in Tris-HCl buffer (pH 10) at 80°C for 8 minutes, achieving optimal conditions for melanin bleaching while preserving tissue morphology and antigenicity. This rapid, effective, fully automated, and standardized workflow can be applied to routine staining procedures in clinical laboratories, thereby improving the quality of pathological diagnosis.

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.