Realistic medical image super-resolution with pyramidal feature multi-distillation networks for intelligent healthcare systems

Abstract

There are two key requirements for medical lesion image super-resolution reconstruction in intelligent healthcare systems: clarity and reality. Because only clear and real super-resolution medical images can effectively help doctors observe the lesions of the disease. The existing super-resolution methods based on pixel space optimization often lack high-frequency details which result in blurred detail features and unclear visual perception. Also, the super-resolution methods based on feature space optimization usually have artifacts or structural deformation in the generated image. This paper proposes a novel pyramidal feature multi-distillation network for super-resolution reconstruction of medical images in intelligent healthcare systems. Firstly, we design a multi-distillation block that combines pyramidal convolution and shallow residual block. Secondly, we construct a two-branch super-resolution network to optimize the visual perception quality of the super-resolution branch by fusing the information of the gradient map branch. Finally, we combine contextual loss and L1 loss in the gradient map branch to optimize the quality of visual perception and design the information entropy contrast-aware channel attention to give different weights to the feature map. Besides, we use an arbitrary scale upsampler to achieve super-resolution reconstruction at any scale factor. The experimental results show that the proposed super-resolution reconstruction method achieves superior performance compared to other methods in this work.

Enhancing the Super-Resolution of Medical Images: Introducing the Deep Residual Feature Distillation Channel Attention Network for Optimized Performance and Efficiency

“…This superior performance accentuates its adaptability and robustness across varying datasets and magnifications. RDST [32], RMISR-BL [36], and LCRCA [35] demonstrate variable performances in different datasets, but none consistently outperform the proposed model. While pioneering models like SRCNN [17] have established foundational benchmarks, they appear to be surpassed by more advanced and refined successors that adeptly balance quality and efficiency.…”

“…Existing super-resolution methods often struggle with preserving high-frequency details, leading to blurred features or introducing artifacts and structural deformations. RMISR-BL [36] addresses these issues through a pyramidal feature multidistillation network. Key components include a multidistilliation block that combines pyramidal convolution and shallow residual blocks, a two-branch super-resolution network for optimizing visual quality, and the use of contextual loss and L1 loss in a gradient map branch to enhance visual perception.…”