Enhancing Passive Cavitation Imaging Using Pth Root Compression Delay, Sum, and Integrate Beamforming: In Vitro and In Vivo Studies

Electrical Engineering Electronic Information Engineering Information Science Medical Imaging Life Sciences Bioengineering Medicine
Passive Cavitation Imaging Pth Root Compression Beamforming In Vitro Studies In Vivo Studies Ultrasound Therapy Cavitation Monitoring

Application of pth Root Compression Delay, Sum and Integrate Beamforming in Passive Cavitation Imaging

Academic Background

Passive Cavitation Imaging (PCI) is a technique used to monitor bubble activity during ultrasound therapy, widely applied in treatment scenarios such as drug delivery and tissue ablation (e.g., Histotripsy). However, existing PCI technologies suffer from low axial resolution and significant side lobe artifacts, especially when using the Delay, Sum and Integrate (DSI) beamforming algorithm. To improve the performance of PCI, researchers have been exploring new algorithms that can enhance image quality without significantly increasing computational complexity.

This study aims to evaluate the effectiveness of a pth root compression delay, sum and integrate (PRDSI) beamforming algorithm in PCI. By applying nonlinear compression and integration operations, this algorithm can effectively suppress side lobe artifacts and improve axial resolution, providing more precise monitoring for bubble-mediated ultrasound therapies.

Source of the Paper

This paper was co-authored by Abhinav Kumar Singh and Pankaj Warbal from the Indian Institute of Technology (IIT) Gandhinagar, and Katia Flores Basterrechea and Kenneth B. Bader from The University of Chicago. The paper was published in the IEEE Transactions on Biomedical Engineering journal in 2025.

Research Workflow

Experimental Design and Procedure

This study is divided into three main experimental stages, conducted in different models, including an in vitro blood flow model, a red blood cell-doped agar model, and an in vivo pig thrombus model. Below are the specific procedures for each stage:

1. In Vitro Blood Flow Model Experiment

  • Research Subject: A flow model perfused with ultrasound contrast agent (Sonovue).
  • Experimental Equipment: Focused ultrasound source (center frequency 2 MHz) and linear array probe (L11-5V).
  • Experimental Process: Sonovue was injected into the flow model and exposed to focused ultrasound, recording acoustic emission signals generated by cavitation.
  • Data Processing: Signals were processed using DSI, Robust Capon Beamforming (RCB), and PRDSI algorithms to generate PCI images.
  • Performance Evaluation: Image performance was assessed using metrics such as axial width, Signal-to-Interference Ratio (SIR), and binary statistical analysis.

2. Red Blood Cell-Doped Agar Model Experiment

  • Research Subject: An agar model doped with red blood cells.
  • Experimental Equipment: 1 MHz histotripsy ultrasound source and C5-2V array probe.
  • Experimental Process: Bubble clouds were generated in the agar model, and acoustic emission signals produced by bubble activity were recorded.
  • Data Processing: Signals were processed using DSI, RCB, and PRDSI algorithms to generate PCI images.
  • Performance Evaluation: Image performance was assessed using metrics such as axial width, SIR, and binary statistical analysis.

3. In Vivo Pig Thrombus Model Experiment

  • Research Subject: Thrombus in the femoral vein of a pig.
  • Experimental Equipment: 1.5 MHz histotripsy ultrasound source and L11-5V array probe.
  • Experimental Process: Histotripsy treatment was applied to the thrombus, and acoustic emission signals generated during the treatment were recorded.
  • Data Processing: Signals were processed using DSI, RCB, and PRDSI algorithms to generate PCI images.
  • Performance Evaluation: Image performance was assessed using metrics such as axial width, SIR, and binary statistical analysis.

Algorithm Design

The PRDSI algorithm effectively suppresses noise and artifacts by applying pth root compression and integration to the received radiofrequency signals. The specific steps are as follows: 1. Apply pth root compression to the received radiofrequency signals. 2. Integrate the compressed signals. 3. Eliminate DC offset to generate PCI images.

Data Analysis

This study employed binary statistical analysis and Receiver Operating Characteristic (ROC) curves to quantitatively evaluate PCI images. The performance of the PRDSI algorithm was assessed by comparing axial width, SIR, and computation time across different algorithms.

Key Results

1. In Vitro Blood Flow Model Experiment Results

  • Axial Width: When p=4, the axial width improved by 64.0±5.2% compared to the DSI algorithm.
  • Signal-to-Noise Ratio: When p=4, the signal-to-noise ratio improved by 11.6±5.5 dB compared to the DSI algorithm.
  • Computation Time: The computation time of the PRDSI algorithm increased by 33.5% compared to the DSI algorithm.

2. Red Blood Cell-Doped Agar Model Experiment Results

  • Axial Width: When p=4, the axial width improved by 56.7±13.0% compared to the DSI algorithm.
  • Signal-to-Noise Ratio: When p=4, the signal-to-noise ratio improved by 10.6±4.2 dB compared to the DSI algorithm.

3. In Vivo Pig Thrombus Model Experiment Results

  • Axial Width: When p=4, the axial width improved by 63.2±13.8% compared to the DSI algorithm.
  • Signal-to-Noise Ratio: When p=4, the signal-to-noise ratio improved by 3.7±3.4 dB compared to the DSI algorithm.

Conclusion

This study demonstrates the effectiveness of the PRDSI algorithm in passive cavitation imaging, significantly improving axial resolution and suppressing side lobe artifacts. Compared to existing DSI and RCB algorithms, the PRDSI algorithm shows clear advantages in imaging performance with only a slight increase in computational complexity. This algorithm provides a more precise monitoring tool for bubble-mediated ultrasound therapies, offering significant clinical application value.

Research Highlights

  1. Novel Algorithm Design: The PRDSI algorithm effectively suppresses noise and artifacts through pth root compression and integration operations.
  2. Comprehensive Experimental Validation: The study validates the effectiveness of the PRDSI algorithm in various in vitro and in vivo models.
  3. Clinical Application Potential: This algorithm offers new possibilities for real-time monitoring in ultrasound therapy.

Other Valuable Information

A limitation of this study is that the optimal parameter p for the PRDSI algorithm may vary depending on the application scenario. Future research could further optimize the algorithm and explore its application in a broader range of clinical settings.

Through this study, the PRDSI algorithm brings new breakthroughs to the field of passive cavitation imaging and is expected to play an important role in future ultrasound therapies.