# Multimodality Biomedical Image Registration using Free Point Transformer Networks

arXiv: https://arxiv.org/abs/2008.01885 · HTML: https://mlsanity.com/paper/2008.01885/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Aug 2020.

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

- [Real-time multimodal image registration with partial intraoperative point-set data](https://mlsanity.com/paper/2109.05023/) (2021)
- [Non-rigid Medical Image Registration using Physics-informed Neural Networks](https://mlsanity.com/paper/2302.10343/) (2023)
- [Meta-Learning Initializations for Interactive Medical Image Registration](https://mlsanity.com/paper/2210.15371/) (2022)
- [A Narrative Review of Image Processing Techniques Related to Prostate Ultrasound](https://mlsanity.com/paper/2407.00678/) (2024)
- [Morphological Change Forecasting for Prostate Glands using Feature-based Registration and Kernel Density Extrapolation](https://mlsanity.com/paper/2101.06425/) (2021)
- [Learning Generalized Non-Rigid Multimodal Biomedical Image Registration from Generic Point Set Data](https://mlsanity.com/paper/2207.10994/) (2022)
- [A Two-Step Deep Learning Method for 3DCT-2DUS Kidney Registration During Breathing](https://mlsanity.com/paper/2305.13855/) (2023)
- [Salient Region Matching for Fully Automated MR-TRUS Registration](https://mlsanity.com/paper/2501.03510/) (2025)
- [NimbleReg: A light-weight deep-learning framework for diffeomorphic image registration](https://mlsanity.com/paper/2503.07768/) (2025)

## Abstract (arXiv)

We describe a point-set registration algorithm based on a novel free point transformer (FPT) network, designed for points extracted from multimodal biomedical images for registration tasks, such as those frequently encountered in ultrasound-guided interventional procedures. FPT is constructed with a global feature extractor which accepts unordered source and target point-sets of variable size. The extracted features are conditioned by a shared multilayer perceptron point transformer module to predict a displacement vector for each source point, transforming it into the target space. The point transformer module assumes no vicinity or smoothness in predicting spatial transformation and, together with the global feature extractor, is trained in a data-driven fashion with an unsupervised loss function. In a multimodal registration task using prostate MR and sparsely acquired ultrasound images, FPT yields comparable or improved results over other rigid and non-rigid registration methods. This demonstrates the versatility of FPT to learn registration directly from real, clinical training data and to generalize to a challenging task, such as the interventional application presented.