RCLM-net: A rich-club properties based lightweight deep neural network for multimodal sequence spine MR images segmentation

• Published in: Magnetic resonance imaging Vol. 123; p. 110505
• Main Authors: He, Siyuan, Zhang, Mengchao, Yan, Xia, Wang, Kun, Chang, Xinyi
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Inc 01.11.2025
• Subjects: Algorithms; Deep Learning; Humans; Image Interpretation, Computer-Assisted - methods; Image Processing, Computer-Assisted - methods; Lightweight deep neural network; Magnetic Resonance Imaging - methods; MR image; Multimodal image segmentation; Multimodal Imaging - methods; Neural Networks, Computer; Rich-Club properties; Spinal Diseases - diagnostic imaging; Spine - diagnostic imaging; Rich-Club properties; Lightweight deep neural network; Multimodal image segmentation; MR image
• ISSN: 0730-725X; 1873-5894; 1873-5894
• DOI: 10.1016/j.mri.2025.110505

The diagnosis of spinal diseases relies on the comprehensive assessment of multimodal sequence MR images. As a prior foundation for intelligent aid diagnosis, it is important to design an efficient MR image segmentation model. However, previous models for multimodal images tend to have a large number of parameters, which is not suitable for low-computing-power application scenarios, especially for primary medical sites. To address the above problem, this study constructs a lightweight neural network model for multimodal sequence spine MR images segmentation, named RCLW-Net (Rich-Club Lightweight Net), which is inspired by the “Rich-Club” properties of brain functional networks. Specifically, our network effectively reduced the number of model parameters by constructing a compact network structure that satisfies the Rich-Club properties and combining the Recurrent residual Convolution as the base convolution module. Verified by the five-fold cross validation of the three-modal sequence MR spine images containing 200 patients' data, proposed RCLW-Net has only 11 % of the parameter scale of the classical Unet++, but the segmentation performance is better. Moreover, proposed RCLW-Net can be successfully deployed on a lightweight embedded device Jetson Nano B01. The low number of parameters and the high performance of the RCLW-Net show the potential of the application in low-computing power scenarios. •RCLM-Net model based on Rich-Club properties with much smaller number of parameters has encouraging segmentation accuracy.•RCLM-Net is suitable for spine multimodal sequence MR images segmentation, which is meaningful for aid diagnosis algorithms.•This model is suitable for deployment on low-computing power devices and is easier to implement in remote areas.