Deep learning-driven brain tumor classification and segmentation using non-contrast MRI

• Published in: Scientific reports Vol. 15; no. 1; pp. 27831 - 24
• Main Authors: Lu, Nan-Han, Huang, Yung-Hui, Liu, Kuo-Ying, Chen, Tai-Been
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 30.07.2025; Nature Publishing Group; Nature Portfolio
• Subjects: 692/4028; 692/700/1421/65; Accuracy; Adult; Aged; Artificial intelligence; Automation; Brain cancer; Brain MRI; Brain Neoplasms - classification; Brain Neoplasms - diagnostic imaging; Brain Neoplasms - pathology; Brain research; Brain tumors; Classification; Convolutional neural networks (CNNs); Datasets; Deep Learning; Efficiency; Female; Fully convolutional networks (FCNs); Glioma; Humanities and Social Sciences; Humans; Image processing; Image Processing, Computer-Assisted - methods; Innovations; Machine learning; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Medical imaging; Middle Aged; multidisciplinary; Neural networks; Neural Networks, Computer; Privacy; Science; Science (multidisciplinary); Segmentation; Tumor classification; Tumors; Deep learning; Fully convolutional networks (FCNs); Convolutional neural networks (CNNs); Brain MRI; Artificial intelligence; Tumor segmentation; Tumor classification
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-025-13591-2

This study aims to enhance the accuracy and efficiency of MRI-based brain tumor diagnosis by leveraging deep learning (DL) techniques applied to multichannel MRI inputs. MRI data were collected from 203 subjects, including 100 normal cases and 103 cases with 13 distinct brain tumor types. Non-contrast T1-weighted (T1w) and T2-weighted (T2w) images were combined with their average to form RGB three-channel inputs, enriching the representation for model training. Several convolutional neural network (CNN) architectures were evaluated for tumor classification, while fully convolutional networks (FCNs) were employed for tumor segmentation. Standard preprocessing, normalization, and training procedures were rigorously followed. The RGB fusion of T1w, T2w, and their average significantly enhanced model performance. The classification task achieved a top accuracy of 98.3% using the Darknet53 model, and segmentation attained a mean Dice score of 0.937 with ResNet50. These results demonstrate the effectiveness of multichannel input fusion and model selection in improving brain tumor analysis. While not yet integrated into clinical workflows, this approach holds promise for future development of DL-assisted decision-support tools in radiological practice.