2186 Artificial Intelligence-Based Volumetric Measurement for Longitudinal Clinical Assessment of High-Grade Glioma Treatment Response: A Multi-Dataset Validation

View at publisher Summarise with VIRAT
Authors: Zerubabbel K. Asfaw, Gianina C. Hernandez‐Marquez, Tirone Young, Cole S. Brown, Lewis E. Tomalin, Isabelle M. Germano
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

INTRODUCTION: High-grade gliomas (HGGs) require ongoing imaging to guide treatment, traditionally relying on labor-intensive and variable manual MRI segmentation. While FDA-approved artificial intelligence (AI) tools offer automated tumor volume segmentation, their clinical utility in decision-making remains understudied. METHODS: We applied the FDA-approved, AI-based tool Neosoma HGG to quantify tumor volumes in 214 subjects from public datasets and 49 from an institutional cohort. AI-derived volumes were compared to expert manual and other AI-based measurements. Therapeutic response assessments using RANO criteria were evaluated against MDTB diagnoses. Segmentation times were analyzed using mixed-model regression. RESULTS: We analyzed 1,648 MRI sequences of 95 HGG patients across three datasets. Contrast-enhancing (CE) tumor volumes were consistent across AI platforms, and Neosoma HGG significantly reduced segmentation time (p < 0.0001). AI-informed disease state assessments showed moderate-to-substantial agreement with MDTB diagnoses for progressive disease (kappa = 0.45–0.64). Key discrepancies arose from limitations of Neosoma HGG in distinguishing pseudo-progression from tumor progression. FLAIR-derived edema volumes varied significantly between AI platforms (p < 0.001), with discordances largely due to over-segmentation beyond the tumor region. CONCLUSIONS: AI-based volumetric segmentation improves efficiency and standardization in monitoring HGG, especially for CE tumor burden. However, moderate concordance with MDTB assessments and difficulties with FLAIR imaging underscore its current limitations. AI may serve as a clinical decision support tool, with further refinement needed to improve specificity and integrate multimodal imaging data.

Fields of Study & Subjects

Brain Tumor Detection and Classification Glioma Diagnosis and Treatment Artificial Intelligence in Healthcare and Education

Publication details

Journal / source
Neurosurgery
Volume
72
Issue
Supplement_1
Pages
297–297
Publication date
26 March 2026
Language
EN