August 12, 2025Open Access

Implicit neural representation for scalable 3D reconstruction from sparse ultrasound images

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Proposed method improves 3D reconstruction accuracy in mice while reducing data needs, suggesting new approaches for ultrasound imaging.

Achieving a remarkable 93% accuracy in tumor volume measurements using only 36 B-mode images demonstrates significant advancement.
The method achieved 7.9 times the performance compared to traditional algorithms, indicating improved efficiency and effectiveness.
This approach utilizes implicit neural representations alongside a 1D array system for continuous 3D reconstruction from sparse data.
Findings suggest potential for on-the-fly interactive analysis using less data than conventional methods, highlighting its significance.

Grutman et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark