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.

Lentigo maligna (LM), a form of melanoma in situ that predominantly affects sun-exposed areas such as the face, has an ill-defined clinical border and has a high rate of recurrence. Atypical Intraepidermal Melanocytic Proliferation (AIMP) is a term used to describe the melanocytic proliferation of an uncertain malignant potential. Clinically and histologically, AIMP can be difficult to distinguish from LM, and indeed AIMP may in some cases progress to LM. Reflectance Confocal Microscopy (RCM) is often used to investigate these lesions non-invasively, however, RCM is often not readily available nor is the associated expertise for RCM image interpretation. Here, we demonstrate machine learning architectures that can correctly classify lesions between LM and AIMP on stacks of RCM images. Overall, our methods showcase the potential for computer-aided diagnosis in dermatology, which in conjunction with the remote acquisition, can expand the range of diagnostic tools in the community.