Abstract 1660: Reflectance confocal microscopy modeling of lentigo maligna

Abstract

Abstract Early melanoma detection in sun-damaged skin can be a challenge. Clinically, some solar lentigines or flat seborrheic keratoses may be relatively indistinguishable from Lentigo Maligna. Reflectance Confocal Microscopy can now be used in the clinic to evaluate these lesions in vivo and increase biopsy accuracy. Spatial relationships of cells and architecture of associated skin structures play a critical role in appropriate diagnoses. As we move toward artificial intelligence using computer diagnostic platforms, it will be important to base algorithms on data with high diagnostic yield. We have undertaken efforts to create 3D reconstructions from sequential depths of confocal images of skin lesion in vivo in order to allow for more precise definition spatial relationships. These efforts have facilitated clearer evaluation of architectural changes at the dermal epidermal junction, location of cells around adnexal structures, and extent of melanocytic dendricity and diversity within the lesion. The use of color maps for efficient identification and accurate localization of cellular and architectural features of interest is another approach to advanced evaluation of confocal image data. We anticipate that these efforts will further improve accuracy for both human and computer diagnoses from reflectance confocal images. We appreciate generous support of the Theodora B Betz Foundation. Citation Format: Katharine L. Hanlon, James M. Grichnik. Reflectance confocal microscopy modeling of lentigo maligna [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1660.