From Anatomy to Genomics Using a Multi-Task Deep Learning Approach for Comprehensive Glioma Profiling

Akmalbek Abdusalomov; Sabina Umirzakova; Obidjon Bekmirzaev; Adilbek Dauletov; Abror Buriboev; Alpamis Kutlimuratov; Akhram Nishanov; Rashid Nasimov; Ryumduck Oh

doi:10.3390/bioengineering12090979

From Anatomy to Genomics Using a Multi-Task Deep Learning Approach for Comprehensive Glioma Profiling

Bioengineering (Basel). 2025 Sep 15;12(9):979. doi: 10.3390/bioengineering12090979.

Authors

Affiliations

¹ Department of Computer Engineering, Gachon University Sujeong-Gu, Seongnam-Si 13120, Republic of Korea.
² Department of Information Processing and Management Systems, Tashkent State Technical University, Tashkent 100095, Uzbekistan.
³ Department of Artificial Intelligence, Tashkent State University of Economics, Tashkent 100066, Uzbekistan.
⁴ Department of Digital Technologies, Alfraganus University, Tashkent 100190, Uzbekistan.
⁵ Department of AI-Software, Gachon University, Sujeong-Gu, Seongnam-Si 13120, Republic of Korea.
⁶ Department of Applied Informatics, Kimyo International University, Tashkent 100170, Uzbekistan.
⁷ Faculty of Software Engineering, Tashkent University of Information Technologies Named After Muhammad Al-Khwarizmi, Tashkent 100200, Uzbekistan.
⁸ Department of Computer Systems, Tashkent University of Information Technologies Named After Muhammad Al-Khwarizmi, Tashkent 100200, Uzbekistan.
⁹ Department of Computer Software, Korea National University of Transportation, Chungju-si 27909, Republic of Korea.

Abstract

Background: Gliomas are among the most complex and lethal primary brain tumors, necessitating precise evaluation of both anatomical subregions and molecular alterations for effective clinical management.

Methods: To find a solution to the disconnected nature of current bioimage analysis pipelines, where anatomical segmentation based on MRI and molecular biomarker prediction are done as separate tasks, we use here Molecular-Genomic and Multi-Task (MGMT-Net), a one deep learning scheme that carries out the task of the multi-modal MRI data without any conversion. MGMT-Net incorporates a novel Cross-Modality Attention Fusion (CMAF) module that dynamically integrates diverse imaging sequences and pairs them with a hybrid Transformer-Convolutional Neural Network (CNN) encoder to capture both global context and local anatomical detail. This architecture supports dual-task decoders, enabling concurrent voxel-wise tumor delineation and subject-level classification of key genomic markers, including the IDH gene mutation, the 1p/19q co-deletion, and the TERT gene promoter mutation.

Results: Extensive validation on the Brain Tumor Segmentation (BraTS 2024) dataset and the combined Cancer Genome Atlas/Erasmus Glioma Database (TCGA/EGD) datasets demonstrated high segmentation accuracy and robust biomarker classification performance, with strong generalizability across external institutional cohorts. Ablation studies further confirmed the importance of each architectural component in achieving overall robustness.

Conclusions: MGMT-Net presents a scalable and clinically relevant solution that bridges radiological imaging and genomic insights, potentially reducing diagnostic latency and enhancing precision in neuro-oncology decision-making. By integrating spatial and genetic analysis within a single model, this work represents a significant step toward comprehensive, AI-driven glioma assessment.

Keywords: end-to-end tumor characterization; integrated glioma diagnosis; multi-task medical imaging; non-invasive molecular profiling; volumetric brain MRI analysis.