Search

Background: Lentigo maligna (LM), a form of melanoma in situ, has no risk of causing metastasis unless dermal invasive melanoma (LMM) supervenes. Furthermore, the detection of invasion impacts prognosis and management., Objective: To assess the accuracy of RCM for the detection of invasion component on LM/LMM lesions., Methods: In the initial case-control study, the performance of one expert in detecting LMM at the time of initial RCM assessment of LM/LMM lesions was recorded prospectively (n = 229). The cases were assessed on RCM-histopathology correlation sessions and a panel with nine RCM features was proposed to identify LMM, which was subsequently tested in a subset of initial cohort (n = 93) in the matched case-control study by two blinded observers. Univariable and multivariable logistic regression models were performed to evaluate RCM features predictive of LMM. Reproducibility of assessment of the nine RCM features was also evaluated., Results: A total of 229 LM/LMM cases evaluated by histopathology were assessed blindly and prospectively by an expert confocalist. On histopathology, 210 were LM and 19 were LMM cases. Correct identification of an invasive component was achieved for 17 of 19 LMM cases (89%) and the absence of a dermal component was correctly diagnosed in 190 of 210 LM cases (90%). In the matched case-control (LMM n = 35, LM n = 58), epidermal and junctional disarray, large size of melanocytes and nests of melanocytes were independent predictors of LMM on multivariate analysis. The interobserver analysis demonstrated that these three features had a fair reproducibility between the two investigators (K = 0.4). The multivariable model including those three features showed a high predictive performance AUC = 74% (CI 95% 64-85%), with sensitivity of 63% (95% CI 52-78%) and specificity of 79% (CI 95% 74-88%), and likelihood ratio of 18 (p-value 0.0026)., Conclusion: Three RCM features were predictive for identifying invasive melanoma in the background of LM., (© 2023 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology.)

Simple Summary: Melanoma is a serious public health concern that causes significant illness and death, especially among young adults in Australia and New Zealand. Reflectance confocal microscopy is a non-invasive imaging technique commonly used to differentiate between different types of melanomas, but it requires specialized expertise and equipment. In this study, we used machine learning to develop classifiers for classifying patient image stacks between two types of melanoma. Our approach achieved high accuracy, demonstrating the utility of computer-aided diagnosis to improve expertise and access to reflectance confocal imaging among the dermatology community. Lentigo maligna (LM) is an early form of pre-invasive melanoma that predominantly affects sun-exposed areas such as the face. LM is highly treatable when identified early but has an ill-defined clinical border and a high rate of recurrence. Atypical intraepidermal melanocytic proliferation (AIMP), also known as atypical melanocytic hyperplasia (AMH), is a histological description that indicates melanocytic proliferation with uncertain malignant potential. Clinically and histologically, AIMP can be difficult to distinguish from LM, and indeed AIMP may, in some cases, progress to LM. The early diagnosis and distinction of LM from AIMP are important since LM requires a definitive treatment. Reflectance confocal microscopy (RCM) is an imaging technique often used to investigate these lesions non-invasively, without biopsy. However, RCM equipment is often not readily available, nor is the associated expertise for RCM image interpretation easy to find. Here, we implemented a machine learning classifier using popular convolutional neural network (CNN) architectures and demonstrated that it could correctly classify lesions between LM and AIMP on biopsy-confirmed RCM image stacks. We identified local z-projection (LZP) as a recent fast approach for projecting a 3D image into 2D while preserving information and achieved high-accuracy machine classification with minimal computational requirements. [ABSTRACT FROM AUTHOR]