Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Compact Implicit Neural Representations for Plane Wave Images

Mathilde Monvoisin, Yuxin Zhang, Diana Mateus

TL;DR

This work proposes a novel approach using Implicit Neural Representations (INRs) to compactly encode multi-planar sequences while preserving crucial orientation-dependent information inUltrafast Plane-Wave imaging, the first application of INRs for PW angular interpolation.

Abstract

Ultrafast Plane-Wave (PW) imaging often produces artifacts and shadows that vary with insonification angles. We propose a novel approach using Implicit Neural Representations (INRs) to compactly encode multi-planar sequences while preserving crucial orientation-dependent information. To our knowledge, this is the first application of INRs for PW angular interpolation. Our method employs a Multi-Layer Perceptron (MLP)-based model with a concise physics-enhanced rendering technique. Quantitative evaluations using SSIM, PSNR, and standard ultrasound metrics, along with qualitative visual assessments, confirm the effectiveness of our approach. Additionally, our method demonstrates significant storage efficiency, with model weights requiring 530 KB compared to 8 MB for directly storing the 75 PW images, achieving a notable compression ratio of approximately 15:1.

Compact Implicit Neural Representations for Plane Wave Images

TL;DR

This work proposes a novel approach using Implicit Neural Representations (INRs) to compactly encode multi-planar sequences while preserving crucial orientation-dependent information inUltrafast Plane-Wave imaging, the first application of INRs for PW angular interpolation.

Abstract

Ultrafast Plane-Wave (PW) imaging often produces artifacts and shadows that vary with insonification angles. We propose a novel approach using Implicit Neural Representations (INRs) to compactly encode multi-planar sequences while preserving crucial orientation-dependent information. To our knowledge, this is the first application of INRs for PW angular interpolation. Our method employs a Multi-Layer Perceptron (MLP)-based model with a concise physics-enhanced rendering technique. Quantitative evaluations using SSIM, PSNR, and standard ultrasound metrics, along with qualitative visual assessments, confirm the effectiveness of our approach. Additionally, our method demonstrates significant storage efficiency, with model weights requiring 530 KB compared to 8 MB for directly storing the 75 PW images, achieving a notable compression ratio of approximately 15:1.
Paper Structure (7 sections, 5 equations, 4 figures, 1 table)

This paper contains 7 sections, 5 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. Experiments
  4. Experimental setup
  5. Representation Ability Across Different Training Sizes
  6. Impact of the Rendering
  7. Conclusion

Figures (4)

  • Figure 1: PW angular dependent Implicit Neural Representation.
  • Figure 2: Quantitative comparison of image quality conservation on the SR dataset. 30000 iterations, calculated between GT and ${\mathbf{o}}\xspace'$
  • Figure 3: Quantitative comparison of image quality conservation on the SC, ER, EC datasets. 10000 iterations, calculated between GT and ${\mathbf{o}}\xspace'$
  • Figure 4: Qualitative results. The model was trained with CS1 (38 views). Regions of interest are outlined in color.