# Artificial Intelligence for Lentigo Maligna: Automated Margin Assessment via Sox-10-Based Melanocyte Density Mapping

## TL;DR

This study explores using AI to help pathologists assess the margins of lentigo maligna by mapping melanocyte density, improving efficiency and accuracy in diagnosis.

## Contribution

The novel contribution is an AI tool that automates melanocyte density mapping using Sox-10 staining for lentigo maligna margin assessment.

## Key findings

- The AI model achieved 87.84% sensitivity, 72.82% specificity, and 79.10% accuracy in melanocyte density assessment.
- The tool provides a three-color heat map for visualizing melanocyte density, improving workflow efficiency for pathologists.
- The model can be adapted for other nuclear stains like PRAME or MITF, expanding its potential applications.

## Abstract

Since lentigo maligna, a melanoma in situ, occurs primarily on sun-damaged skin, it is often difficult to distinguish it histologically from the latter, especially in peripheral areas, as both have similar histological characteristics. Artificial intelligence is already widely used in dermatopathology. This study, therefore, examines whether such a tool can be helpful in assessing the margins of lentigo maligna. Our tool is primarily based on the detection of epidermal melanocytes in digitised slides, as increased melanocyte density is a decisive criterion for malignancy in lentigo maligna. It reliably detects these and also provides colour coding of the melanocytes in a traffic light system, with areas of low, borderline, and increased density. This procedure should be improved in further studies, but it has the potential to make daily workflows more time-efficient and to provide support in cases of doubt.

Lentigo maligna (LM) is a melanoma in situ with high cumulative sun damage. Histological evaluation of resection margins is difficult and time-consuming. Melanocyte density (MD) is a suitable, quantifiable, and reproducible diagnostic criterion. In this retrospective single-centre study, we investigated whether an artificial intelligence (AI) tool can support the assessment of LM. Training and evaluation were based on MD in Sox-10-stained digitalised slides. In total, 86 whole slide images (WSIs) from LM patients were annotated and used as a training set. The test set consisted of 177 slides. The tool was trained to detect the epidermis, measure its length, and determine the MD. A cut-off of ≥30 melanocytes per 0.5 mm of epidermis length was defined as positive. Our AI model automatically recognises the epidermis and measures the MD. The model was trained on nuclear immunohistochemical signals and can also be applied to other nuclear stains, such as PRAME or MITF. The WSI is automatically visualised by a three-colour heat map with a subdivision into low, borderline, and high melanocyte density. The cut-offs can be adjusted individually. Compared to manually counted ground truth MD, the AI model achieved high sensitivity (87.84%), specificity (72.82%), and accuracy (79.10%), and an area under the curve (AUC) of 0.818 in the test set. This automated tool can assist (dermato) pathologists by providing a quick overview of the WSI at first glance and making the time-consuming assessment of resection margins more efficient and more reproducible. The AI model can provide significant benefits in the daily routine workflow.