# Intraoperative Liver Surface Completion with Graph Convolutional VAE

arXiv: https://arxiv.org/abs/2009.03871 · HTML: https://mlsanity.com/paper/2009.03871/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Sep 2020.

## Built on by (most cited first, 9 of 9)

- [Improving anatomical plausibility in medical image segmentation via hybrid graph neural networks: applications to chest x-ray analysis](https://mlsanity.com/paper/2203.10977/) (2022)
- [An objective comparison of methods for augmented reality in laparoscopic liver resection by preoperative-to-intraoperative image fusion](https://mlsanity.com/paper/2401.15753/) (2024)
- [3D Shape Variational Autoencoder Latent Disentanglement via Mini-Batch Feature Swapping for Bodies and Faces](https://mlsanity.com/paper/2111.12448/) (2021)
- [Hybrid graph convolutional neural networks for landmark-based anatomical segmentation](https://mlsanity.com/paper/2106.09832/) (2021)
- [3D Generative Model Latent Disentanglement via Local Eigenprojection](https://mlsanity.com/paper/2302.12798/) (2023)
- [Evaluation of Intra-operative Patient-specific Methods for Point Cloud Completion for Minimally Invasive Liver Interventions](https://mlsanity.com/paper/2503.11969/) (2025)
- [Depth Augmented and FE Free 3D/2D Liver Registration for Laparoscopic Liver AR](https://mlsanity.com/paper/2602.17517/) (2026)
- [Latent Disentanglement in Mesh Variational Autoencoders Improves the Diagnosis of Craniofacial Syndromes and Aids Surgical Planning](https://mlsanity.com/paper/2309.10825/) (2023)
- [Toward Patient-specific Partial Point Cloud to Surface Completion for Pre- to Intra-operative Registration in Image-guided Liver Interventions](https://mlsanity.com/paper/2505.19518/) (2025)

## Abstract (arXiv)

In this work we propose a method based on geometric deep learning to predict the complete surface of the liver, given a partial point cloud of the organ obtained during the surgical laparoscopic procedure. We introduce a new data augmentation technique that randomly perturbs shapes in their frequency domain to compensate the limited size of our dataset. The core of our method is a variational autoencoder (VAE) that is trained to learn a latent space for complete shapes of the liver. At inference time, the generative part of the model is embedded in an optimisation procedure where the latent representation is iteratively updated to generate a model that matches the intraoperative partial point cloud. The effect of this optimisation is a progressive non-rigid deformation of the initially generated shape. Our method is qualitatively evaluated on real data and quantitatively evaluated on synthetic data. We compared with a state-of-the-art rigid registration algorithm, that our method outperformed in visible areas.