MBFormer: Microbubble Transformer for 3D Time-Series Data Processing to Improve Bound Bubble Detection in Nonde-structive Ultrasound Molecular Imaging
Abstract
Development of nondestructive ultrasound molecular imaging (UMI) is essential for early cancer detection through real-time screening using clinical ultrasound systems. Current techniques face challenges in accurately detecting targeted microbubbles (MBs) bound to specific biomarkers, primarily due to false-positive detections of unbound free-floating MBs. We propose a transformer model for time-series video processing to improve the differentiation of bound MBs. We propose a hierarchical transformer, termed MBFormer (microbubble transformer), featuring a positional-embedding-free encoder and a lightweight decoder. Leveraging attention within 3D spatio-temporal data to effectively capture stationary signals from bound MBs while suppressing nonstationary signals from unbound MBs. Since MBs appears as relatively small textures compared with conventional segmentation targets in medical imaging, such as organs and tumors, we optimized the model with two hierarchical layers, each with an attention block, to process ultrasound video data. The network outputs the molecular signal amplitude to visualize fine MB textures. Performance was evaluated using an in vivo breast cancer model, compared against a prior CNN-based UMI method and SegFormer3D baseline, a representative 3D transformer. MBFormer (AUC = 0.943) outperformed both CNN (AUC = 0.897) and SegFormer3D (AUC = 0.766) in detecting bound MBs. The CNN showed residual molecular signal from free MBs in the cardiac chambers, whereas SegFormer3D failed to detect fine MB textures. Overall, MBFormer demonstrated enhanced detection of bound MBs while suppressing free MBs and achieved a frame rate of 16.7 to 18.1 FPS, demonstrating its potential for real-time application. We anticipate that this transformer-based UMI model can facilitate real-time, free-hand nondestructive UMI in clinical systems.
Related
- Nonlinear pluggable optics: Digital signal processing-free Intensity Modulated Direct Detection links using analog photonic Next Generation Reservoir Computing
- A Fast and Scalable Transformer Pipeline for Binary Black Hole Detection
- A dynamic pore-network model with discrete microbubble transport
- In vitro targeted microbubble adhesion exhibits a maximum at intermediate acoustic pressures
- Continuous vision transformer-based neural operator for scalable beam propagation modeling in optical fibers
BibTeX
@misc{baek2026mbformer,
title = {{MBFormer: Microbubble Transformer for 3D Time-Series Data Processing to Improve Bound Bubble Detection in Nonde-structive Ultrasound Molecular Imaging}},
author = {Jihye Baek and Jeong Hoon Lee and Hoda Hashemi and Arutselvan Natarajan and Farbod Tabesh and Ramasamy Paulmurugan and Jeremy J. Dahl},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.30618},
eprint = {2609.30618},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.30618},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - MBFormer: Microbubble Transformer for 3D Time-Series Data Processing to Improve Bound Bubble Detection in Nonde-structive Ultrasound Molecular Imaging
AU - Jihye Baek
AU - Jeong Hoon Lee
AU - Hoda Hashemi
AU - Arutselvan Natarajan
AU - Farbod Tabesh
AU - Ramasamy Paulmurugan
AU - Jeremy J. Dahl
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2609.30618
UR - https://arxiv.org/abs/2609.30618
AB - Development of nondestructive ultrasound molecular imaging (UMI) is essential for early cancer detection through real-time screening using clinical ultrasound systems. Current techniques face challenges in accurately detecting targeted microbubbles (MBs) bound to specific biomarkers, primarily due to false-positive detections of unbound free-floating MBs. We propose a transformer model for time-series video processing to improve the differentiation of bound MBs. We propose a hierarchical transformer, termed MBFormer (microbubble transformer), featuring a positional-embedding-free encoder and a lightweight decoder. Leveraging attention within 3D spatio-temporal data to effectively capture stationary signals from bound MBs while suppressing nonstationary signals from unbound MBs. Since MBs appears as relatively small textures compared with conventional segmentation targets in medical imaging, such as organs and tumors, we optimized the model with two hierarchical layers, each with an attention block, to process ultrasound video data. The network outputs the molecular signal amplitude to visualize fine MB textures. Performance was evaluated using an in vivo breast cancer model, compared against a prior CNN-based UMI method and SegFormer3D baseline, a representative 3D transformer. MBFormer (AUC = 0.943) outperformed both CNN (AUC = 0.897) and SegFormer3D (AUC = 0.766) in detecting bound MBs. The CNN showed residual molecular signal from free MBs in the cardiac chambers, whereas SegFormer3D failed to detect fine MB textures. Overall, MBFormer demonstrated enhanced detection of bound MBs while suppressing free MBs and achieved a frame rate of 16.7 to 18.1 FPS, demonstrating its potential for real-time application. We anticipate that this transformer-based UMI model can facilitate real-time, free-hand nondestructive UMI in clinical systems.
ER -