"DCSLK: Combined Large Kernel Shared Convolutional Model with Dynamic Channel Sampling".
Authors
Affiliations (4)
Affiliations (4)
- Doctoral Candidate, Xinjiang University, Medical Image Segmentation, Information Security. Electronic address: [email protected].
- Department of Gastroenterology, No. 940 Hospital, Joint Logistics Support Force of the Chinese People's Liberation Army, Multi-Modal Medical Image Segmentation. Electronic address: [email protected].
- Information Department, No. 940 Hospital, Joint Logistics Support Force of the Chinese People's Liberation Army, Bio-information Security. Electronic address: [email protected].
- School of Computer Science and Technology, Xinjiang University, Medical Image Segmentation. Electronic address: [email protected].
Abstract
This study centers around the competition between Convolutional Neural Networks (CNNs) with large convolutional kernels and Vision Transformers in the domain of computer vision, delving deeply into the issues pertaining to parameters and computational complexity that stem from the utilization of large convolutional kernels. Even though the size of the convolutional kernels has been extended up to 51×51, the enhancement of performance has hit a plateau, and moreover, striped convolution incurs a performance degradation. Enlightened by the hierarchical visual processing mechanism inherent in humans, this research innovatively incorporates a shared parameter mechanism for large convolutional kernels. It synergizes the expansion of the receptive field enabled by large convolutional kernels with the extraction of fine-grained features facilitated by small convolutional kernels. To address the surging number of parameters, a meticulously designed parameter sharing mechanism is employed, featuring fine-grained processing in the central region of the convolutional kernel and wide-ranging parameter sharing in the periphery. This not only curtails the parameter count and mitigates the model complexity but also sustains the model's capacity to capture extensive spatial relationships. Additionally, in light of the problems of spatial feature information loss and augmented memory access during the 1×1 convolutional channel compression phase, this study further puts forward a dynamic channel sampling approach, which markedly elevates the accuracy of tumor subregion segmentation. To authenticate the efficacy of the proposed methodology, a comprehensive evaluation has been conducted on three brain tumor segmentation datasets, namely BraTs2020, BraTs2024, and Medical Segmentation Decathlon Brain 2018. The experimental results evince that the proposed model surpasses the current mainstream ConvNet and Transformer architectures across all performance metrics, proffering novel research perspectives and technical stratagems for the realm of medical image segmentation.