I$^3$Net: Inter-Intra-slice Interpolation Network for Medical Slice Synthesis

TL;DR

This paper introduces I$^3$Net, a novel neural network for medical slice synthesis that leverages inter- and intra-slice information, significantly improving the quality of anisotropic medical images from limited data.

Contribution

The paper proposes I$^3$Net, a new network architecture that exploits high in-plane resolution and cross-view information for superior slice interpolation in medical imaging.

Findings

Outperforms state-of-the-art methods in PSNR by at least 1.14dB.

Achieves 43.90dB PSNR with faster inference on the MSD dataset.

Effectively utilizes frequency domain features and cross-view information for improved synthesis.

Abstract

Medical imaging is limited by acquisition time and scanning equipment. CT and MR volumes, reconstructed with thicker slices, are anisotropic with high in-plane resolution and low through-plane resolution. We reveal an intriguing phenomenon that due to the mentioned nature of data, performing slice-wise interpolation from the axial view can yield greater benefits than performing super-resolution from other views. Based on this observation, we propose an Inter-Intra-slice Interpolation Network (I$^3$Net), which fully explores information from high in-plane resolution and compensates for low through-plane resolution. The through-plane branch supplements the limited information contained in low through-plane resolution from high in-plane resolution and enables continual and diverse feature learning. In-plane branch transforms features to the frequency domain and enforces an equal learning opportunity for all frequency bands in a global context learning paradigm. We further propose a cross-view block to take advantage of the information from all three views online. Extensive experiments on two public datasets demonstrate the effectiveness of I$^3$Net, and noticeably outperforms state-of-the-art super-resolution, video frame interpolation and slice interpolation methods by a large margin. We achieve 43.90dB in PSNR, with at least 1.14dB improvement under the upscale factor of $\times$2 on MSD dataset with faster inference. Code is available at https://github.com/DeepMed-Lab-ECNU/Medical-Image-Reconstruction.