The prognostic role of radiomics in gastroenteropancreatic neuroendocrine tumors

#3834

Introduction: Gastroenteropancreatic neuroendocrine neoplasia (GEP NENs) are rare and heterogenous diseases.

Aim(s): To evaluate if radiomics should be able to predict, at initial diagnosis, the clinical outcome of patients affected by metastatic GEP NENs.

Materials and methods: Patients with histological diagnosis of GEP NENs and liver metastases were enrolled in the study. The population was divided into progressive or no progressive if no diseases progression was observed during follow-up. All CT-images were acquired before starting any medical treatment. A total of 107 radiomic features were extracted with a dedicated software (3DSlicer v4.10.2) and compared between the two groups using T-test or Mann-Withney test whereas each performance was evaluated by ROC curve. Survival analyses were made by Kaplan-Meyer curves. Logistic regression analysis was used to construct three models to predict disease progression. A p value < 0.05 was considered statistically significant.

Conference:

Presenting Author:

Authors: Magi L, Polici M, Rinzivillo M, Caruso D, Zerunian M,

Keywords: radiomics, neuroendocrine tumor, prognosis, disease progression,