# Metastatic Melanoma Prognosis Prediction Using a TC Radiomic-Based Machine Learning Model: A Preliminary Study

## TL;DR

This study uses CT imaging and machine learning to predict outcomes for metastatic melanoma patients, offering a potential tool for personalized treatment.

## Contribution

A radiomic-based machine learning model is proposed for early prognosis prediction in metastatic melanoma.

## Key findings

- The model achieved an 82% ROC-AUC in internal testing for predicting lesion outcomes.
- Radiomic features combined with AI showed better predictive ability than traditional methods.
- Results suggest potential for future multicenter validation and clinical decision support.

## Abstract

Although much progress has been made, melanoma still remains a disease with an often poor prognosis. So, identifying new imaging markers that are able to give prognosis predictions would help in the management of patients and would avoid unnecessary (and often harmful) treatments. With the aim of providing early and accurate prediction of clinical outcome, in this preliminary study, we developed a machine learning model based on radiomic features extracted from CT images of patients with metastatic melanoma. Through the use of radiomics, we have the ability to reveal aspects of the tumor not visible to the human eye. Integrated with artificial intelligence, it improves predictive ability and promotes personalized treatment choices. Although this is a pilot study, the results offer a promising basis for future multicenter validations.

Background/Objective: The approach to the clinical management of metastatic melanoma patients is undergoing a significant transformation. The availability of a large amount of data from medical images has made Artificial Intelligence (AI) applications an innovative and cutting-edge solution that could revolutionize the surveillance and management of these patients. In this study, we develop and validate a machine-learning model based on radiomic data extracted from a computed tomography (CT) analysis of patients with metastatic melanoma (MM). This approach was designed to accurately predict prognosis and identify the potential key factors associated with prognosis. Methods: To achieve this goal, we used radiomic pipelines to extract the quantitative features related to lesion texture, morphology, and intensity from high-quality CT images. We retrospectively collected a cohort of 58 patients with metastatic melanoma, from which a total of 60 CT series were used for model training, and 70 independent CT series were employed for external testing. Model performance was evaluated using metrics such as sensitivity, specificity, and AUC (area under the curve), demonstrating particularly favorable results compared to traditional methods. Results: The model used in this study presented a ROC-AUC curve of 82% in the internal test and, in combination with AI, presented a good predictive ability regarding lesion outcome. Conclusions: Although the cohort size was limited and the data were collected retrospectively from a single institution, the findings provide a promising basis for further validation in larger and more diverse patient populations. This approach could directly support clinical decision-making by providing accurate and personalized prognostic information.