Deep Learning Approaches for Glioblastoma Prognosis in Resource-Limited Settings: A Study using Basic Patient Demographic, Clinical, and Surgical Inputs.

Researchers

Journal

Modalities

Models

Abstract

Glioblastoma (GBM) is the most common brain tumor in the United States, with an annual incidence rate of 3.21 per 100,000. It is the most aggressive type of diffuse glioma and has a median survival time of months following treatment. This study aims to assess the accuracy of different novel Deep Learning (DL) models trained on a set of simple clinical, demographic, and surgical variables to assist in clinical practice, even in areas with constrained healthcare infrastructure.Our study included 37,095 GBM patients from the SEER database. All predictors were based on demographic, clinicopathologic, and treatment information of the cases. Our outcomes of interest were months of survival and vital status. Concordance index (C-index) and integrated Brier scores (IBS) were used to evaluate the models’ performances.The patient characteristics and the statistical analyses were consistent with epidemiological literature. The models C-index and IBS ranged from 0.6743 to 0.6918 and from 0.0934 to 0.1034 respectively. PMF (0.6918), MTLR (0.6916), and Logistic Hazard (0.6916) had the highest C-index scores. The models with the lowest IBS were the PMF (0.0934), MTLR (0.0935), and Logistic Hazard (0.0936). These models had an accuracy (1-IBS) of 90.66%; 90.65%, and 90.64% respectively. The deep learning algorithms were deployed on an interactive web-based tool for practical use available via: https://glioblastoma-survanalysis.herokuapp.com/ CONCLUSION: Novel DL algorithms can better predict glioblastoma prognosis than baseline methods and can potentially lead to more personalized patient care regardless of extensive electronic health record availability.Copyright © 2023 Elsevier Inc. All rights reserved.