REAL‐TIME AI‐BASED VOLUMETRIC ANALYSIS OF THE HIPPOCAMPUS IN PATIENTS WITH ALZHEIMER DISEASE
Annotatsiya
Abstract Background Patients suffering from a neurodegenerative condition such as Alzheimer's disease are more likely to experience hippocampal atrophy. Unfortunately, manually segmenting the hippocampus on a brain MRI scan is a time‐consuming procedure. Delineating the structure on several slices can consume a significant amount of radiologists' valuable time. We aim to use the real‐time AI method for hippocampal segmentation in patients with Alzheimer disease Method The study involved 65 participants in total, with a mean age of 72.65 5.35 years. The hippocampus volumes and asymmetry index (AI) were calculated using a 3D T1 isotropic sequence on United Imaging Healthcare 1.5 T scanner. Volumes were determined automatically using uAI‐Discovery Brain software. We used the T1‐3D sequence (voxel size [mm3]: 0.8 × 0.8 × 1; time [ms]: TE = 2.6; TR = 2000; TI = 900; duration [min]: 9:19) for manual and automatic segmentation of the hippocampus. Result The study revealed a high degree of concordance (ICC = 0.95) for both hippocampi. As a result, no major disparities existed between them. The subjects' mean right (2985.65344.25 mm3) and left (2898.69 225.2 mm3) volume was determined. These volumes and the asymmetry index were significantly different from those found in a sample of healthy participants (p = 0.00001 and p = 0.003, respectively), but were comparable to those obtained in a sample of subjects with AD Conclusion To emphasize, precise segmentation of the hippocampus via AI‐method can provide critical diagnostic information and open a new path in clinical settings