Abstract:Ultrasound Localization Microscopy (ULM) relies on the injection of microbubbles (MBs) to obtain highly resolved density maps of blood circulation in vivo, with a resolution that can reach 10 {\mu}m ~ {\lambda}/10 in the rodent brain. Static mean velocity maps can be extracted but are intrinsically biased by potential significant changes in the number of MBs detected during the cardiac cycle. Dynamic ULM (DULM) is a technique developed for non-invasive pulsatility measurements in the brain of rodents, leading to temporally resolved velocity and density cine-loops. It was previously based on external triggers such as the electrocardiogram (ECG), limiting its use to datasets acquired specifically for DULM applications while also increasing the required acquisition time. This study presents a new motion matching method using tissue Doppler that eliminates the need for ECG-gating in DULM experiments. DULM can now be performed on any ULM datasets, recovering pertinent temporal information, and improving the robustness of the mean velocity estimates.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to improve the ability of dynamic ultrasound localization microscopy (DULM) technology to perform non - invasive imaging of cerebral hemodynamics in mice, rats and cats without using electrocardiogram (ECG) triggering. Specifically, the researchers developed a motion - matching algorithm based on tissue Doppler to eliminate the need for ECG triggering, so that dynamic circulation information, such as cine - loops of blood flow velocity and density, can be extracted from any ultrasound localization microscopy (ULM) dataset, and the robustness of average velocity estimation can be improved. ### Main Objectives: 1. **Eliminate Dependence on External Triggers**: Traditional DULM technology requires external trigger signals (such as ECG) to synchronize acquisition time, which limits its application range and increases experimental complexity and the time required. The method proposed in this paper eliminates this requirement, making DULM applicable to any ULM dataset. 2. **Improve the Accuracy of Average Velocity Estimation**: Due to the variation in the number of microbubbles detected during the cardiac cycle, traditional ULM technology has biases in estimating the average blood flow velocity. The new method corrects these biases through the motion - matching algorithm, improving the accuracy of average velocity estimation. 3. **Achieve High - Resolution Imaging of Whole - Brain Pulsatility Index (PI)**: Through the improved DULM technology, the researchers were able to generate high - resolution PI maps throughout the brain, which is very important for evaluating the pulsatile behavior of blood vessels. ### Technical Details: - **Motion - Matching Algorithm**: Utilize the periodic characteristics of tissue Doppler signals to align different frame groups with reference frame groups, thereby achieving time registration. This enables the recovery of blood flow dynamic information synchronized with the cardiac cycle even without ECG triggering. - **Data Processing Flow**: It includes steps such as beamforming of IQ data, spatio - temporal filtering, microbubble detection and tracking, and phase distortion correction, and finally generates high - resolution maps of vascular density and average velocity. - **Verification Method**: By introducing random delays in the rat dataset with ECG triggering to simulate the situation without ECG triggering, the effectiveness of the motion - matching algorithm is verified. In addition, this algorithm is directly applied in the mouse and cat datasets to demonstrate its effectiveness in practical applications. ### Experimental Results: - **Rat Experiment**: Successfully recovered the blood flow dynamic information during the cardiac cycle, verifying the effectiveness of the motion - matching algorithm. - **Mouse Experiment**: Successfully generated cine - loops of blood flow velocity and density without ECG triggering, and calculated the pulsatility index (PI) of multiple blood vessels. - **Cat Experiment**: Applied this method in a three - dimensional dataset and also successfully extracted blood flow dynamic information and PI. In conclusion, this research significantly improves the applicability and accuracy of DULM technology by developing a new motion - matching algorithm, providing a new tool for non - invasive cerebral hemodynamics imaging.

Dynamic Imaging using any Ultrasound Localization Microscopy Dataset

Dynamic Ultrasound Localization Microscopy Without ECG-Gating

A Tracking prior to Localization workflow for Ultrasound Localization Microscopy

3D transcranial Dynamic Ultrasound Localization Microscopy in the mouse brain using a Row-Column Array

Quantitative pulsatility measurements using 3D dynamic ultrasound localization microscopy

A Novel Method for Ultrasound Localization Microscopy Based on Temporal Correlation Through Deep Learning

Deep Learning for Ultrasound Localization Microscopy

Non-invasive transcranial volumetric ultrasound localization microscopy of the rat brain with continuous, high volume-rate acquisition

In vivo ultrasound localization microscopy for high-density microbubbles

Deep Learning in Ultrasound Localization Microscopy: Applications and Perspectives

Longitudinal Awake Imaging of Deep Mouse Brain Microvasculature with Super-resolution Ultrasound Localization Microscopy

Deep learning for fast super-resolution ultrasound microvessel imaging

Fast super-resolution ultrasound microvessel imaging using spatiotemporal data with deep fully convolutional neural network

A Deep Learning Framework for Spatiotemporal Ultrasound Localization Microscopy

Improved Ultrasound Localization Microscopy Using Deep Learning Enhanced Beamformer

Microbubble Identification Based on Decision Theory for Ultrasound Localization Microscopy

Three-Dimensional Ultrasound Localization Microscopy with Bipartite Graph-Based Microbubble Pairing and Kalman-Filtering-Based Tracking on a 256-Channel Verasonics Ultrasound System with a 32 × 32 Matrix Array

Sparse channel sampling for ultrasound localization microscopy (SPARSE-ULM)

Improved Ultrasound Localization Microscopy Based on Microbubble Uncoupling via Transmit Excitation

Qulm-Dl: Quantitative Ultrasound Localization Microscopy Via Deep Learning