# Long-term Dependency for 3D Reconstruction of Freehand Ultrasound Without External Tracker

arXiv: https://arxiv.org/abs/2310.10248 · HTML: https://mlsanity.com/paper/2310.10248/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 42 references that are in the map.
- First posted on arXiv 16 Oct 2023.

## Builds on (most cited first, 15 of 15)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [On the Continuity of Rotation Representations in Neural Networks](https://mlsanity.com/paper/1812.07035/) (2018)
- [Efficiently Identifying Task Groupings for Multi-Task Learning](https://mlsanity.com/paper/2109.04617/) (2021)
- [NeRF--: Neural Radiance Fields Without Known Camera Parameters](https://mlsanity.com/paper/2102.07064/) (2021)
- [Memory-Efficient Backpropagation Through Time](https://mlsanity.com/paper/1606.03401/) (2016)
- [Reviving and Improving Recurrent Back-Propagation](https://mlsanity.com/paper/1803.06396/) (2018)
- [Freehand Ultrasound Image Simulation with Spatially-Conditioned Generative Adversarial Networks](https://mlsanity.com/paper/1707.05392/) (2017)
- [Self Context and Shape Prior for Sensorless Freehand 3D Ultrasound Reconstruction](https://mlsanity.com/paper/2108.00274/) (2021)
- [Deep Motion Network for Freehand 3D Ultrasound Reconstruction](https://mlsanity.com/paper/2207.00177/) (2022)
- [Sensorless Freehand 3D Ultrasound Reconstruction via Deep Contextual Learning](https://mlsanity.com/paper/2006.07694/) (2020)
- [Computing CNN Loss and Gradients for Pose Estimation with Riemannian Geometry](https://mlsanity.com/paper/1805.01026/) (2018)
- [Image quality assessment for machine learning tasks using meta-reinforcement learning](https://mlsanity.com/paper/2203.14258/) (2022)
- [Trackerless freehand ultrasound with sequence modelling and auxiliary transformation over past and future frames](https://mlsanity.com/paper/2211.04867/) (2022)
- [Multi-IMU with Online Self-Consistency for Freehand 3D Ultrasound Reconstruction](https://mlsanity.com/paper/2306.16197/) (2023)

## Built on by (most cited first, 7 of 7)

- [TUS-REC2024: A Challenge to Reconstruct 3D Freehand Ultrasound Without External Tracker](https://mlsanity.com/paper/2506.21765/) (2025)
- [Nonrigid Reconstruction of Freehand Ultrasound without a Tracker](https://mlsanity.com/paper/2407.05767/) (2024)
- [MoNetV2: Enhanced Motion Network for Freehand 3D Ultrasound Reconstruction](https://mlsanity.com/paper/2506.15835/) (2025)
- [Enhancing Free-hand 3D Photoacoustic and Ultrasound Reconstruction using Deep Learning](https://mlsanity.com/paper/2502.03505/) (2025)
- [UltrasODM: A Dual Stream Optical Flow Mamba Network for 3D Freehand Ultrasound Reconstruction](https://mlsanity.com/paper/2512.07756/) (2025)
- [MLRecon: Robust Markerless Freehand 3D Ultrasound Reconstruction via Coarse-to-Fine Pose Estimation](https://mlsanity.com/paper/2603.00990/) (2026)
- [Seeing Globally, Refining Locally: Global Visual Guidance and Local Ultrasound Cues for Robust Freehand 3-D Ultrasound Reconstruction](https://mlsanity.com/paper/2607.12398/) (2026)

## Abstract (arXiv)

Objective: Reconstructing freehand ultrasound in 3D without any external tracker has been a long-standing challenge in ultrasound-assisted procedures. We aim to define new ways of parameterising long-term dependencies, and evaluate the performance. Methods: First, long-term dependency is encoded by transformation positions within a frame sequence. This is achieved by combining a sequence model with a multi-transformation prediction. Second, two dependency factors are proposed, anatomical image content and scanning protocol, for contributing towards accurate reconstruction. Each factor is quantified experimentally by reducing respective training variances. Results: 1) The added long-term dependency up to 400 frames at 20 frames per second (fps) indeed improved reconstruction, with an up to 82.4% lowered accumulated error, compared with the baseline performance. The improvement was found to be dependent on sequence length, transformation interval and scanning protocol and, unexpectedly, not on the use of recurrent networks with long-short term modules; 2) Decreasing either anatomical or protocol variance in training led to poorer reconstruction accuracy. Interestingly, greater performance was gained from representative protocol patterns, than from representative anatomical features. Conclusion: The proposed algorithm uses hyperparameter tuning to effectively utilise long-term dependency. The proposed dependency factors are of practical significance in collecting diverse training data, regulating scanning protocols and developing efficient networks. Significance: The proposed new methodology with publicly available volunteer data and code for parametersing the long-term dependency, experimentally shown to be valid sources of performance improvement, which could potentially lead to better model development and practical optimisation of the reconstruction application.