Prediction of melanoma evolution in melanocytic nevi via artificial intelligence: A call for prospective data

DermSynth3D: Synthesis of in-the-wild annotated dermatology images

In recent years, deep learning (DL) has shown great potential in the field of dermatological image analysis. However, existing datasets in this domain have significant limitations, including a small number of image samples, limited disease conditions, insufficient annotations, and non-standardized image acquisitions. To address these shortcomings, we propose a novel framework called DermSynth3D. DermSynth3D blends skin disease patterns onto 3D textured meshes of human subjects using a differentiable renderer and generates 2D images from various camera viewpoints under chosen lighting conditions in diverse background scenes. Our method adheres to top-down rules that constrain the blending and rendering process to create 2D images with skin conditions that mimic in-the-wild acquisitions, ensuring more meaningful results. The framework generates photo-realistic 2D dermatological images and the corresponding dense annotations for semantic segmentation of the skin, skin conditions, body parts, bounding boxes around lesions, depth maps, and other 3D scene parameters, such as camera position and lighting conditions. DermSynth3D allows for the creation of custom datasets for various dermatology tasks. We demonstrate the effectiveness of data generated using DermSynth3D by training DL models on synthetic data and evaluating them on various dermatology tasks using real 2D dermatological images. We make our code publicly available at https://github.com/sfu-mial/DermSynth3D.

Advancing Dermatological Diagnostics: Interpretable AI for Enhanced Skin Lesion Classification

A crucial challenge in critical settings like medical diagnosis is making deep learning models used in decision-making systems interpretable. Efforts in Explainable Artificial Intelligence (XAI) are underway to address this challenge. Yet, many XAI methods are evaluated on broad classifiers and fail to address complex, real-world issues, such as medical diagnosis. In our study, we focus on enhancing user trust and confidence in automated AI decision-making systems, particularly for diagnosing skin lesions, by tailoring an XAI method to explain an AI model's ability to identify various skin lesion types. We generate explanations using synthetic images of skin lesions as examples and counterexamples, offering a method for practitioners to pinpoint the critical features influencing the classification outcome. A validation survey involving domain experts, novices, and laypersons has demonstrated that explanations increase trust and confidence in the automated decision system. Furthermore, our exploration of the model's latent space reveals clear separations among the most common skin lesion classes, a distinction that likely arises from the unique characteristics of each class and could assist in correcting frequent misdiagnoses by human professionals.

Melanin Stacking Differences in Pigmented and Non-Pigmented Melanomas: Quantitative Differentiation between Pigmented and Non-Pigmented Melanomas Based on Light-Scattering Properties

Cutaneous melanoma is a cancer with metastatic potential characterized by varying amounts of pigment-producing melanocytes, and it is one of the most aggressive and fatal forms of skin malignancy, with several hundreds of thousands of cases each year. Early detection and therapy can lead to decreased morbidity and decreased cost of therapy. In the clinic, this often translates to annual skin screenings, especially for high-risk patients, and generous use of the ABCDE (asymmetry, border irregularity, color, diameter, evolving) criteria. We have used a new technique termed vibrational optical coherence tomography (VOCT) to non-invasively differentiate between pigmented and non-pigmented melanomas in a pilot study. The VOCT results reported in this study indicate that both pigmented and non-pigmented melanomas have similar characteristics, including new 80, 130, and 250 Hz peaks. Pigmented melanomas have larger 80 Hz peaks and smaller 250 Hz peaks than non-pigmented cancers. The 80 and 250 Hz peaks can be used to quantitative characterize differences between different melanomas. In addition, infrared light penetration depths indicated that melanin in pigmented melanomas has higher packing densities than in non-pigmented lesions. Using machine learning techniques, the sensitivity and specificity of differentiating skin cancers from normal skin are shown to range from about 78% to over 90% in this pilot study. It is proposed that using AI on both lesion histopathology and mechanovibrational peak heights may provide even higher specificity and sensitivity for differentiating the metastatic potential of different melanocytic lesions.

The Many Roles of Dermoscopy in Melanoma Detection

Dermoscopy is a non-invasive method of examination that aids the clinician in many ways, especially in early skin cancer detection. Melanoma is one of the most aggressive forms of skin cancer that can affect individuals of any age, having an increasing incidence worldwide. The gold standard for melanoma diagnosis is histopathological examination, but dermoscopy is also very important for its detection. To highlight the many roles of dermoscopy, we analyzed 200 melanocytic lesions. The main objective of this study was to detect through dermoscopy hints of melanomagenesis in the studied lot. The most suspicious were 10 lesions which proved to be melanomas confirmed through histopathology. The second objective of this study was to establish if dermoscopy can aid in estimating the Breslow index (tumoral thickness) of the melanomas and to compare the results to the histopathological examination. We found that the tumoral thickness may be estimated through dermoscopy, but the histopathological examination is superior. To conclude, the aim of this study was to showcase the versatility and many roles of dermoscopy, besides being one of the most important tools for early melanoma diagnosis.

Review: The Key Factors to Melanomagenesis

Melanoma is the most dangerous form of skin cancer that develops from the malignant transformation of the melanocytes located in the basal layer of the epidermis (cutaneous melanoma). Melanocytes may also be found in the meninges, eyes, ears, gastrointestinal tract, genito-urinary system, or other mucosal surfaces (mucosal melanoma). Melanoma is caused by an uncontrolled proliferation of melanocytes, that at first may form a benign lesion (nevogenesis), but in time, it may transition to melanoma, determining what it is named, melanomagenesis. Some tumors may appear spontaneously (de novo melanoma) or on preexisting lesions (nevus-associated melanoma). The exact cause of melanoma may not be fully understood yet, but there are some factors that initiate and promote this malignant process. This study aims to provide a summary of the latest articles regarding the key factors that may lead to melanomagenesis. The secondary objectives are to reveal the relationship between nevi and melanoma, to understand the cause of "de novo" and "nevus-associated melanoma" and highlight the differences between these subtypes.

DenseNet-II: an improved deep convolutional neural network for melanoma cancer detection

Research in the field of medicine and relevant studies evince that melanoma is one of the deadliest cancers. It defines precisely that the condition develops due to uncontrolled growth of melanocytic cells. The current trends in any disease detection revolve around the usage of two main categories of models; these are general machine learning models and deep learning models. Further, the experimental analysis of melanoma has an additional requirement of visual records like dermatological scans or normal camera lens images. This further accentuates the need for a more accurate model for melanoma detection. In this work, we aim to achieve the same, primarily by the extensive usage of neural networks. Our objective is to propose a deep learning CNN framework-based model to improve the accuracy of melanoma detection by customizing the number of layers in the network architecture, activation functions applied, and the dimension of the input array. Models like Resnet, DenseNet, Inception, and VGG have proved to yield appreciable accuracy in melanoma detection. However, in most cases, the dataset was classified into malignant or benign classes only. The dataset used in our research provides seven lesions; these are melanocytic nevi, melanoma, benign keratosis, basal cell carcinoma, actinic keratoses, vascular lesions, and dermatofibroma. Thus, through the HAM10000 dataset and various deep learning models, we diversified the precision factors as well as input qualities. The obtained results are highly propitious and establish its credibility.

Assessing the Economic Value of Clinical Artificial Intelligence: Challenges and Opportunities

OBJECTIVES

Clinical artificial intelligence (AI) is a novel technology, and few economic evaluations have focused on it to date. Before its wider implementation, it is important to highlight the aspects of AI that challenge traditional health technology assessment methods.

METHODS

We used an existing broad value framework to assess potential ways AI can provide good value for money. We also developed a rubric of how economic evaluations of AI should vary depending on the case of its use.

RESULTS

We found that the measurement of core elements of value-health outcomes and cost-are complicated by AI because its generalizability across different populations is often unclear and because its use may necessitate reconfigured clinical processes. Clinicians' productivity may improve when AI is used. If poorly implemented though, AI may also cause clinicians' workload to increase. Some AI has been found to exacerbate health disparities. Nevertheless, AI may promote equity by expanding access to medical care and, when properly trained, providing unbiased diagnoses and prognoses. The approach to assessment of AI should vary based on its use case: AI that creates new clinical possibilities can improve outcomes, but regulation and evidence collection may be difficult; AI that extends clinical expertise can reduce disparities and lower costs but may result in overuse; and AI that automates clinicians' work can improve productivity but may reduce skills.

CONCLUSIONS

The potential uses of clinical AI create challenges for health technology assessment methods originally developed for pharmaceuticals and medical devices. Health economists should be prepared to examine data collection and methods used to train AI, as these may impact its future value.

Digital dermoscopy monitoring of melanocytic lesions: Two novel calculators combining static and dynamic features to identify melanoma

BACKGROUND

Early diagnosis is the most effective intervention to improve the prognosis of cutaneous melanoma. Even though the introduction of dermoscopy has improved the diagnostic accuracy, it can still be difficult to distinguish some melanomas from benign melanocytic lesions. Digital dermoscopy monitoring can identify dynamic changes of melanocytic lesions: To date, some algorithms were proposed, but a universally accepted one is still lacking.

OBJECTIVES

To identify independent predictive variables associated with the diagnosis of cutaneous melanoma and develop a multivariable dermoscopic prediction model able to discriminate benign from malignant melanocytic lesions undergoing digital dermoscopy monitoring.

METHODS

We collected dermoscopic images of melanocytic lesions excised after dermoscopy monitoring and carried out static and dynamic evaluations of dermoscopic features. We built two multivariable predictive models based on logistic regression and random forest.

RESULTS

We evaluated 173 lesions (65 cutaneous melanomas and 108 nevi). Forty-two melanomas were in situ, and the median thickness of invasive melanomas was 0.35 mm. The median follow-up time was 9.8 months for melanomas and 9.1 for nevi. The logistic regression and random forest models performed with AUC values of 0.87 and 0.89, respectively, were substantially higher than those of the static evaluation models (ABCD TDS score, 0.57; 7-point checklist, 0.59). Finally, we built two risk calculators, which translate the proposed models into user-friendly applications, to assist clinicians in the decision-making process.

CONCLUSIONS

The present study demonstrates that the integration of dynamic and static evaluations of melanocytic lesions is a safe approach that can significantly boost the diagnostic accuracy for cutaneous melanoma. We propose two diagnostic tools that significantly increase the accuracy in discriminating melanoma from nevi during digital dermoscopy monitoring.

Etiologies of Melanoma Development and Prevention Measures: A Review of the Current Evidence

Simple Summary

Melanoma constitutes a major public health risk, with the rates of diagnosis increasing on a yearly basis. Monitoring for risk factors and preventing dangerous behaviors that increase melanoma risk, such as tanning, are important measures for melanoma prevention. Additionally, assessing the effectiveness of various methods to prevent sun exposure and sunburns—which can lead to melanoma—is important to help identify ways to reduce the development of melanoma. We summarize the recent evidence regarding the heritable and behavioral risks underlying melanoma, as well as the current methods used to reduce the risk of developing melanoma and to improve the diagnosis of this disease.

Abstract

(1) Melanoma is the most aggressive dermatologic malignancy, with an estimated 106,110 new cases to be diagnosed in 2021. The annual incidence rates continue to climb, which underscores the critical importance of improving the methods to prevent this disease. The interventions to assist with melanoma prevention vary and typically include measures such as UV avoidance and the use of protective clothing, sunscreen, and other chemopreventive agents. However, the evidence is mixed surrounding the use of these and other interventions. This review discusses the heritable etiologies underlying melanoma development before delving into the data surrounding the preventive methods highlighted above. (2) A comprehensive literature review was performed to identify the clinical trials, observational studies, and meta-analyses pertinent to melanoma prevention and incidence. Online resources were queried to identify epidemiologic and clinical trial information. (3) Evidence exists to support population-wide screening programs, the proper use of sunscreen, and community-targeted measures in the prevention of melanoma. Clinical evidence for the majority of the proposed preventive chemotherapeutics is presently minimal but continues to evolve. (4) Further study of these chemotherapeutics, as well as improvement of techniques in artificial intelligence and imaging techniques for melanoma screening, is warranted for continued improvement of melanoma prevention.

A Clinicopathological Analysis of Melanocytic Nevi: A Retrospective Series

Purpose: Melanocytic nevi are common cutaneous lesions. This study aimed to demonstrate the concordance and discordance between clinical and histopathological diagnoses of melanocytic nevi and the importance of histological evaluation in differentiating malignant lesions from diseases with similar clinical manifestations.

Patients and Methods: We studied 4,561 consecutive patients with a clinical diagnosis of melanocytic nevi from 2014 to 2019. We compared the clinical diagnosis with the histopathological diagnosis to establish a histopathological concordance rate and then investigated the effects of clinical characteristics and the reasons for removal on misclassification.

Results: Among 4,561 patients who were clinically diagnosed with melanocytic nevi, the overall histopathological concordance rate was 82.11% (3,745 of 4,561 patients), while the histopathological discordance rate was 17.89% (816 of 4,561 patients). The histopathological concordance included 90.25% common acquired melanocytic nevi (3,380 of 3,745 patients) and 9.75% other benign melanocytic neoplasms (365 of 3,745 patients). The most common diagnostic change was to seborrheic keratosis (n = 470, 10.30%), followed by basal cell carcinoma (n = 64, 1.40%), vascular tumor (n = 53, 1.16%), fibroma (n = 43, 0.94%), epidermoid cyst (n = 34, 0.75%), wart (n = 30, 0.66%), melanoma (n = 24, 0.53%), Bowen's disease (n = 16, 0.35%), squamous cell carcinoma (n = 4, 0.09%), keratoacanthoma (n = 2, 0.04%), and other neoplasms (n = 76, 1.67%). Male sex, old age, location of the lesion, and the reasons for removal have a potential effect on misclassification. The percentages of misclassified lesions on the trunk and limbs and the perineum and buttocks were higher than those in lesions without a change in diagnosis. Importantly, locations of lesions on the head and neck were significantly related to a change in diagnosis to non-melanoma skin cancer, while locations on the hands and feet were significantly related to a change in diagnosis to melanoma. In addition to a typical clinical features, removal due to lesion changes or repeated stimulation was significantly associated with a change in diagnosis to melanoma.

Conclusions: Our study emphasizes the clinical differential diagnosis of melanocytic nevi, especially the possibility of malignant tumors. The occurrence of clinical features associated with clinicopathological discordance should raise the clinical suspect and be carefully differentiated from malignant tumors.

Toward automated assessment of mole similarity on dermoscopic images

Purpose: Current skin cancer detection relies on dermatologists' visual assessments of moles directly or dermoscopically. Our goal is to show that our similarity assessment algorithm on dermoscopic images can perform as well as a dermatologist's assessment. Approach: Given one target mole and two other moles from the same patient, our model determines which mole is more similar to the target mole. Similarity was quantified as the Euclidean distance in a feature space designed to capture mole properties such as size, shape, and color. We tested our model on 18 patients, each of whom had at least five moles, and compared the model assessments of mole similarity with that of three dermatologists. Fleiss' Kappa agreement coefficients and iteration tests were used to evaluate the agreement in similarity assessment among dermatologists and our model. Results: With the selected features of size, entropy (color variation), and cluster prominence (asymmetry), our algorithm's similarity assessments agreed moderately with the similarity assessments of dermatologists. The mean Kappa of 1000 iteration tests was 0.49 ( confidence interval   ( CI ) = [ 0.23 , 0.74 ] ) when comparing three dermatologists and our model, which is comparable to the agreement in similarity assessment among the dermatologists themselves (the mean Kappa of 1000 iteration tests for three dermatologists was 0.48, CI = [ 0.19 , 0.77 ] .) By contrast, the mean Kappa was 0.22 ( CI = [ - 0.00 , 0.43 ] ) when comparing the similarity assessments of three dermatologists and random guesses. Conclusions: Our study showed that our image feature-engineering-based algorithm can effectively assess the similarity of moles as dermatologists do. Such a similarity assessment could serve as the foundation for computer-assisted intra-patient evaluation of moles.

Robustness of convolutional neural networks in recognition of pigmented skin lesions

BACKGROUND

A basic requirement for artificial intelligence (AI)-based image analysis systems, which are to be integrated into clinical practice, is a high robustness. Minor changes in how those images are acquired, for example, during routine skin cancer screening, should not change the diagnosis of such assistance systems.

OBJECTIVE

To quantify to what extent minor image perturbations affect the convolutional neural network (CNN)-mediated skin lesion classification and to evaluate three possible solutions for this problem (additional data augmentation, test-time augmentation, anti-aliasing).

METHODS

We trained three commonly used CNN architectures to differentiate between dermoscopic melanoma and nevus images. Subsequently, their performance and susceptibility to minor changes ('brittleness') was tested on two distinct test sets with multiple images per lesion. For the first set, image changes, such as rotations or zooms, were generated artificially. The second set contained natural changes that stemmed from multiple photographs taken of the same lesions.

RESULTS

All architectures exhibited brittleness on the artificial and natural test set. The three reviewed methods were able to decrease brittleness to varying degrees while still maintaining performance. The observed improvement was greater for the artificial than for the natural test set, where enhancements were minor.

CONCLUSIONS

Minor image changes, relatively inconspicuous for humans, can have an effect on the robustness of CNNs differentiating skin lesions. By the methods tested here, this effect can be reduced, but not fully eliminated. Thus, further research to sustain the performance of AI classifiers is needed to facilitate the translation of such systems into the clinic.

Efficacy of smartphone-compatible optical instrument for assessing melanocytic nevi for malignancy

Early detection of melanocytic nevus progression to malignant melanoma is a pressing concern. Traditionally, patients with multiple melanocytic nevi (MMN) are monitored for extended periods of time and excisional biopsies are performed on individual suspicious melanocytic nevi (MN). This approach is costly and tremendously time-consuming for both doctors and patients. The aim of this study was to evaluate the efficacy of a smartphone-compatible optical instrument in the assessment of MN for malignancy. Seven patients aged 43 to 65 years with MMN on the trunk and upper/lower extremities were followed-up for 4 years. Dermoscopy images of MN were taken and analyzed using a Handyscope smartphone-compatible optical system operated at 20x magnification and a Handyscope3 application. A total of 74 MN were surgically removed during the follow-up period. None of the patients had melanoma. The results of dermoscopy image analysis generated by the convolutional neural network coincided with histopathology findings in all cases. The optical Handyscope system demonstrated its efficacy in assessing MN for malignancy. AI can be used for primary screening of MMN dermoscopy images. However, histopathological verification of the diagnosis is still needed.

Overdiagnosis of melanoma - causes, consequences and solutions

Malignant melanoma is the skin tumor that causes most deaths in Germany. At an early stage, melanoma is well treatable, so early detection is essential. However, the skin cancer screening program in Germany has been criticized because although melanomas have been diagnosed more frequently since introduction of the program, the mortality from malignant melanoma has not decreased. This indicates that the observed increase in melanoma diagnoses be due to overdiagnosis, i.e. to the detection of lesions that would never have created serious health problems for the patients. One of the reasons is the challenging distinction between some benign and malignant lesions. In addition, there may be lesions that are biologically equivocal, and other lesions that are classified as malignant according to current criteria, but that grow so slowly that they would never have posed a threat to patient's life. So far, these "indolent" melanomas cannot be identified reliably due to a lack of biomarkers. Moreover, the likelihood that an in-situ melanoma will progress to an invasive tumor still cannot be determined with any certainty. When benign lesions are diagnosed as melanoma, the consequences are unnecessary psychological and physical stress for the affected patients and incurred therapy costs. Vice versa, underdiagnoses in the sense of overlooked melanomas can adversely affect patients' prognoses and may necessitate more intense therapies. Novel diagnostic options could reduce the number of over- and underdiagnoses and contribute to more objective diagnoses in borderline cases. One strategy that has yielded promising results in pilot studies is the use of artificial intelligence-based diagnostic tools. However, these applications still await translation into clinical and pathological routine.