End-to-End Learning of Multi-Organ Implicit Surfaces from 3D Medical Imaging Data

2025-09-15

AI Summary

Addressing the challenges of balancing resolution and computational efficiency, as well as poor transferability of generic implicit representations in multi-organ surface reconstruction from 3D medical images, this paper proposes ImplMORe—the first end-to-end implicit neural surface reconstruction framework tailored for multi-organ medical imaging. ImplMORe employs a 3D CNN encoder to extract local anatomical features and jointly models high-fidelity, topologically consistent organ surfaces in implicit space via an occupancy function coupled with multi-scale continuous interpolation, thereby overcoming voxel-resolution limitations. Evaluated on the TotalSegmentator dataset, ImplMORe significantly outperforms conventional explicit methods: it reconstructs richer surface details and achieves higher shape fidelity, while supporting both single- and multi-organ reconstruction in a unified framework. Moreover, it demonstrates strong generalization capability across diverse anatomical structures and imaging conditions.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

The fine-grained surface reconstruction of different organs from 3D medical imaging can provide advanced diagnostic support and improved surgical planning. However, the representation of the organs is often limited by the resolution, with a detailed higher resolution requiring more memory and computing footprint. Implicit representations of objects have been proposed to alleviate this problem in general computer vision by providing compact and differentiable functions to represent the 3D object shapes. However, architectural and data-related differences prevent the direct application of these methods to medical images. This work introduces ImplMORe, an end-to-end deep learning method using implicit surface representations for multi-organ reconstruction from 3D medical images. ImplMORe incorporates local features using a 3D CNN encoder and performs multi-scale interpolation to learn the features in the continuous domain using occupancy functions. We apply our method for single and multiple organ reconstructions using the totalsegmentator dataset. By leveraging the continuous nature of occupancy functions, our approach outperforms the discrete explicit representation based surface reconstruction approaches, providing fine-grained surface details of the organ at a resolution higher than the given input image. The source code will be made publicly available at: https://github.com/CAMMA-public/ImplMORe

Problem

• Reconstructing multi-organ surfaces from 3D medical imaging
• Overcoming resolution limitations in organ representation
• Adapting implicit representations for medical image applications

Innovation

• End-to-end deep learning method
• Implicit surface representations for reconstruction
• Multi-scale interpolation in continuous domain