UZ | RU | EN
Journal Management System
To develop train and evaluate a deep learning–based model for fully automated morphometric assessment of cerebral hemisphere structures using high-resolution MRI data; to quantify its performance against expert manual segmentation using established statistical metrics; and to assess the clinical utility of the automated morphometric parameters derived from the model for characterizing structural brain differences in healthy individuals and patients with neurological conditions.

DEEP LEARNING–BASED MORPHOMETRIC ASSESSMENT OF CEREBRAL HEMISPHERE STRUCTURES USING MRI

3 authors
2 мая 2026

Authors

G.E. Raxmonova

M.D. Rahmanova

I.I. Izbasarov

Publication Information

Volume: 92
Issue: 2
Pages: 477-483
Journal: ОБРАЗОВАНИЕ НАУКА И ИННОВАЦИОННЫЕ ИДЕИ В МИРЕ

Metrics

24 views
PDF available
Download PDF

Abstract

Magnetic resonance imaging (MRI) remains the gold standard non-invasive technique for detailed morphological visualization of central nervous system structures in vivo. Its superior soft-tissue contrast, multiplanar capability, and absence of ionizing radiation make MRI indispensable for neurological assessment in both clinical and research settings. Morphometric analysis of cerebral hemispheres — encompassing quantitative evaluation of cortical thickness, gray matter volume, white matter integrity, and hemispheric symmetry indices — plays a pivotal role in the early detection, differential diagnosis, and longitudinal monitoring of a wide spectrum of neurological and psychiatric disorders, including Alzheimer's disease, schizophrenia, multiple sclerosis, focal epilepsy, and traumatic brain injury.

Keywords

To develop train and evaluate a deep learning–based model for fully automated morphometric assessment of cerebral hemisphere structures using high-resolution MRI data; to quantify its performance against expert manual segmentation using established statistical metrics; and to assess the clinical utility of the automated morphometric parameters derived from the model for characterizing structural brain differences in healthy individuals and patients with neurological conditions.

Full Text

Full text unavailable

Please download the PDF file to view the full text

References

[1]

Future work will focus on expanding the training dataset to include pediatric

[2]

populations, rare neurological conditions, and multi-site multi-vendor MRI data to

[3]

further improve generalizability. Prospective clinical validation studies are planned to

[4]

assess the impact of AI-assisted morphometric reporting on diagnostic accuracy and

[5]

clinical decision-making. Integration of the framework into a clinical PACS

[6]

compatible workflow with automated report generation is currently under development

[7]

at Tashkent State Medical University. These efforts aim to translate the demonstrated

[8]

technical capabilities of deep learning–based neuroimaging analysis into tangible

[9]

improvements in patient care within the Central Asian healthcare context.

Related Articles