Fetus head circumference (HC) estimation from ultrasound images via segmentation-based and segmentation-free approaches.
⭐ The highlight of this work:
- The HC is estimated via segmentation-based and segmentation-free approaches respectively;
- Methodology on evaluating the segmentation-based and segmentation-free methods;
- The evaluation of two approaches is performed under a fair experimental environment.
👉 We evaluated both two types of approaches in several aspects:
- The architectures (Memory, Parameters);
- The prediction accuracy;
- Agreement analysis;
- Saliency maps;
- Actual inference time and memory cost;
- Comparison with state-of-the-art.
💻 About the code:
The code is implemented with Python 3.* and the Deep learning library Tensorflow (Keras 2.*).
The work is finished together with Caroline Petitjean and Samia Ainouz in LITIS lab.
Please consider citing this paper when you find it useful:
@inproceedings{zhang2020direct,
title={Direct estimation of fetal head circumference from ultrasound images based on regression CNN},
author={Zhang, Jing and Petitjean, Caroline and Lopez, Pierre and Ainouz, Samia},
booktitle={Medical Imaging with Deep Learning},
pages={914--922},
year={2020},
organization={PMLR}
}
@article{zhang2022segmentation,
title={Segmentation-based vs. regression-based biomarker estimation: a case study of fetus head circumference assessment from ultrasound images},
author={Zhang, Jing and Petitjean, Caroline and Ainouz, Samia},
journal={Journal of Imaging},
volume={8},
number={2},
pages={23},
year={2022},
publisher={MDPI}
}