ARTIFICIAL INTELLIGENCE–BASED DETERMINATION OF CEREBRAL HEMISPHERE MORPHOMETRIC PARAMETERS USING MAGNETIC RESONANCE IMAGING (MRI)

Dilorom Adilbekova; Islomjon Izbasarov; Gulbahor Raxmonova; Muhayyo Rahmanova; Umida Utkiraliyeva; Madina Muhammadiyeva

Authors

Dilorom Adilbekova Professor, Department of Anatomy and Clinical Anatomy, Tashkent State Medical University, Tashkent, Uzbekistan
Islomjon Izbasarov Student of Group 202, Faculty of General Medicine No. 2, Tashkent State Medical University, Tashkent, Uzbekistan
Gulbahor Raxmonova Professor of the Department of Medical Radiology No. 1, Tashkent State Medical University, Tashkent, Uzbekistan
Muhayyo Rahmanova Tashkent State Medical University, 1st Doctoral Student in Medical Radiology, Tashkent, Uzbekistan
Umida Utkiraliyeva Student of Group 202, Faculty of General Medicine No. 2, Tashkent State Medical University, Tashkent, Uzbekistan
Madina Muhammadiyeva Student of Group 202, Faculty of General Medicine No. 2, Tashkent State Medical University, Tashkent, Uzbekistan

Keywords:

Artificial Intelligence; Magnetic Resonance Imaging (MRI); Cerebral Hemisphere Morphometry; Deep Learning Segmentation; Cortical Thickness Analysis; Brain Volumetry; Digital Health Integration.

Abstract

Quantitative morphometric analysis of cerebral hemispheres using magnetic resonance imaging (MRI) has become an essential component of modern neuroradiology. Structural parameters such as hemisphere volume, cortical thickness, and surface area serve as sensitive biomarkers for neurodegenerative, neurovascular, and neurodevelopmental disorders. However, traditional manual and semi-automated segmentation approaches are time-consuming and prone to inter-observer variability. Recent advances in artificial intelligence (AI), especially deep learning-based segmentation models, have demonstrated high accuracy in structural brain analysis. This study provides a comprehensive AI-based framework for automated cerebral hemisphere morphometry and describes its implementation via the NeuroMorph AI web platform. The proposed system combines deep learning segmentation, volumetric computing, cortical surface reconstruction, and structural digital reporting by analyzing MRI images, providing a scalable and clinically applicable solution that is compatible with the modern healthcare digital transformation.