Fast Medical Shape Reconstruction via Meta-learned Implicit Neural Representations

TL;DR

This paper introduces a meta-learning approach to implicit neural representations for rapid and accurate medical shape reconstruction, significantly reducing inference time and enhancing generalization across diverse anatomical datasets.

Contribution

The authors propose a meta-learning framework that improves initialization of neural implicit models, enabling faster inference and better generalization in medical shape reconstruction tasks.

Findings

Reduces inference time by an order of magnitude.

Handles various input configurations and modalities.

Generalizes well to unseen shape domains.

Abstract

Efficient and fast reconstruction of anatomical structures plays a crucial role in clinical practice. Minimizing retrieval and processing times not only potentially enhances swift response and decision-making in critical scenarios but also supports interactive surgical planning and navigation. Recent methods attempt to solve the medical shape reconstruction problem by utilizing implicit neural functions. However, their performance suffers in terms of generalization and computation time, a critical metric for real-time applications. To address these challenges, we propose to leverage meta-learning to improve the network parameters initialization, reducing inference time by an order of magnitude while maintaining high accuracy. We evaluate our approach on three public datasets covering different anatomical shapes and modalities, namely CT and MRI. Our experimental results show that our model can handle various input configurations, such as sparse slices with different orientations and spacings. Additionally, we demonstrate that our method exhibits strong transferable capabilities in generalizing to shape domains unobserved at training time.