PT  - JOURNAL ARTICLE
AU  - Zhang, Jikai
AU  - LaBella, Dominic
AU  - Zhang, Dylan
AU  - Houk, Jessica L.
AU  - Rudie, Jeffrey D.
AU  - Zou, Haotian
AU  - Warman, Pranav
AU  - Mazurowski, Maciej A.
AU  - Calabrese, Evan
TI  - Development and Evaluation of Automated Artificial Intelligence–Based Brain Tumor Response Assessment in Patients with Glioblastoma
AID  - 10.3174/ajnr.A8580
DP  - 2025 Apr 24
TA  - American Journal of Neuroradiology
4099  - http://www.ajnr.org/content/early/2025/04/24/ajnr.A8580.short
4100  - http://www.ajnr.org/content/early/2025/04/24/ajnr.A8580.full
AB  - This project aimed to develop and evaluate an automated, AI-based, volumetric brain tumor MRI response assessment algorithm on a large cohort of patients treated at a high-volume brain tumor center. We retrospectively analyzed data from 634 patients treated for glioblastoma at a single brain tumor center over a 5-year period (2017–2021). The mean age was 56 ± 13 years. 372/634 (59%) patients were male, and 262/634 (41%) patients were female. Study data consisted of 3,403 brain MRI exams and corresponding standardized, radiologist-based brain tumor response assessments (BT-RADS). An artificial intelligence (AI)-based brain tumor response assessment (AI-VTRA) algorithm was developed using automated, volumetric tumor segmentation. AI-VTRA results were evaluated for agreement with radiologist-based response assessments and ability to stratify patients by overall survival. Metrics were computed to assess the agreement using BT-RADS as the ground-truth, fixed-time point survival analysis was conducted to evaluate the survival stratification, and associated P-values were calculated. For all BT-RADS categories, AI-VTRA showed moderate agreement with radiologist response assessments (F1 = 0.587–0.755). Kaplan-Meier survival analysis revealed statistically worse overall fixed time point survival for patients assessed as image worsening equivalent to RANO progression by human alone compared to by AI alone (log-rank P = .007). Cox proportional hazard model analysis showed a disadvantage to AI-based assessments for overall survival prediction (P = .012). In summary, our proposed AI-VTRA, following BT-RADS criteria, yielded moderate agreement for replicating human response assessments and slightly worse stratification by overall survival.2D2-dimensionalAIartificial intelligenceAI-VTRAartificial intelligence volumetric tumor response assessmentBT-RADSBrain Tumor Reporting and Data SystemC-indexconcordance indexFeTSFederated Tumor SegmentationGBMglioblastomaIDHisocitrate dehydrogenaseNLPnatural language processingOSoverall survivalRANOResponse Assessment in Neuro-OncologyRECISTResponse Evaluation Criteria in Solid TumorsSDstandard deviationVDETvolumetric differences for enhancing tumorVDFLAIRvolumetric differences for FLAIR