Diagnosing growth in low-grade gliomas with and without artificial intelligence-measured longitudinal volume measurements: A retrospective observational study

Background: Low-grade or grade 2 diffuse gliomas (LGG) infiltrate the brains leading to significant neurological morbidity. This retrospective observational study evaluates the ability of AI-assisted volumetric analysis to correctly detect tumor growth in longitudinal studies of LGG as compared to the standard clinical method. Methods: = 22). All gliomas were from radiation-naïve patients; only 2 patients had completed treatment with temozolomide. The dates of tumor growth were gathered from clinical notes. Longitudinal tumor volumes were calculated by the MRIMath FLAIR AI. Golden truths were obtained by physician reviews using the MRIMath Smart contouring system. Growth by significant shifts in tumor volumes was detected by using the statistical method of online change-of-point method. Results: In the clinical progression group, automatic AI segmentation followed by human review detected tumor growth at a median of 21 months earlier than visual inspection. In the clinically stable group, AI with human review identified growth in 13/22 cases at a median of 23 months earlier than the last magnetic resonance imaging. AI without human review generated similar results but with a 25% false positive and an 8.33% false negative rate. The median time spent by physicians in reviewing, revising, and approving the AI segmentations is 2 minutes. Conclusions: These findings highlight the clinical potential of AI-assisted volumetric analysis followed by physician oversight for the timely detection of tumor progression in LGG patients.