Real-Time 3D Ultrasound Tracking of Microswimmers in Challenging Biomedical Spaces

By Dr Silpaja Chandrasekar, PhDReviewed by Susha Cheriyedath, M.Sc.Jan 25 2024

In a paper published in the journal Scientific Reports, researchers introduced a robust three-dimensional (3D) ultrasound tracking system designed in their lab for real-time monitoring of acoustically actuated microswimmers navigating challenging spaces within the human body. The presented tracking system employs two ultrasound probes, a step motor, and a host controller to track the microdrone's 3D arbitrary motion.

Benchtop experiments evaluating the tracking performance revealed high reliability, with an average error of less than 0.3 mm across six trials compared to synchronized camera recordings. These promising results underscore the potential of the developed 3D ultrasound tracking system for biomedical applications, offering an efficient strategy for navigating and tracking microswimmers in real-time. This technology could represent a significant step towards the clinical translation of microswimmer technology in medicine.

Related Work

Previous research has explored the potential of microswimmers for biomedical applications, particularly in navigating challenging environments within the human body. Acoustic actuation, with its noninvasiveness and deep tissue penetration, has been a focus, leading to the development of 3D swimming microdrones. Despite the advancements in microswimmer technology, real-time tracking of these devices for in vivo applications continues to pose a significant challenge. Most past tracking strategies, primarily two-dimensional (2D) focused, need more practicality in vivo.

3D Microdrone and Tracking

The 3D swimming microdrone possesses multi-directional movement capabilities through strategically placed microtubes. These microtubes, namely "lateral one," "lateral two," and "vertical," enable navigation in three dimensions. Acoustically activated air bubbles within these hydrophobic microtubes generate propulsion forces, facilitating movement in various directions. The resonant frequencies of these microtubes, crucial for optimal propulsion, are designed based on their lengths. Researchers included a dummy cavity in the design to maintain the microdrone's upright posture.

The experimental setup involves the microdrone operating in a water-glycerin solution for neutral buoyancy. An acoustic piezo actuator drives the microswimmer in three dimensions, and the trajectory may vary due to non-uniform media friction. The 3D ultrasound tracking system, consisting of two ultrasound probes, a step motor, and a host controller, facilitates real-time tracking. The ultrasound probes, operating at 5 MHz, provide synchronized imaging for tracking the microdrone's arbitrary motion.

The ultrasound imaging data is processed in real-time to determine the microdrone's elevational motion direction. The motor control algorithm ensures the microdrone remains within the ultrasound field of view, leveraging intensity ratios between the brightest points in the ultrasound images. These scenarios, depicted for step motor control, utilize features within the elevational beam to track the motion of the microdrone.

Researchers utilize a custom-written particle tracking algorithm in matrix laboratory (MATLAB) to actively assess tracking accuracy, employing it to reconstruct trajectories based on data acquired from ultrasound and camera sources. The algorithm preprocesses ultrasound images, estimating the microdrone's position by averaging results from both probes. The tracking error is the position discrepancy between ultrasound and camera trajectories at each frame.

Normalized error, calculated as the error divided by the total moving distance, is also determined to evaluate error accumulation. The methodology integrates the design of the 3D microdrone, the setup of the ultrasound tracking system, and a detailed analysis of tracking results, providing a thorough understanding of the developed technology's capabilities and accuracy in real-time tracking.

Ultrasound Tracking System Evaluation

The time-lapse images showcase the microswimmer's 3D motion, captured by the camera and two ultrasound probes during a representative trial. The microswimmer's lateral-axial motion is effectively imaged in the y–z plane, demonstrating the ultrasound tracking system's ability to follow its trajectory. The reconstructed trajectories from ultrasound and camera tracking indicate good agreement, affirming the system's accuracy. The step-motor displacement corresponds to elevational motion, while tracking error analysis and the reconstructed 3D trajectory provide comprehensive insights into the tracking system's performance.

Researchers conducted six trials to assess the performance of ultrasound tracking. The error analysis incorporates absolute error and normalized error from all six trials. The average tracking errors remained under 0.6 mm at each time point, showcasing the system's precision. The normalized error decreased with the microswimmer distance traveled, indicating no accumulation of systemic errors. The tracking error analysis in axial and lateral directions reflects the spatial resolution limitations of the ultrasound probes.

While errors were observed, particularly in lateral resolution, they fell within reasonable ranges, aligning with the probe's spatial resolution. Importantly, no systemic errors were evident, emphasizing the reliability of the tracking system. This advancement from 2D to 3D ultrasound tracking in real-time marks progress in microswimmer technology. The system's demonstrated reliability lays a foundation for real-time feedback control, which is crucial for future biomedical applications.

Despite the promising results, room for technical enhancement exists. Hardware constraints, primarily probe center frequency, and footprint limitations contribute to occasional high error-to-microdrone-length ratios. Future iterations could benefit from higher-frequency probes and finer footprints for improved accuracy. Incorporating state estimators and the Kalman filter offers the potential for refining ultrasound tracking outcomes. The current frame rate, determined by computer speed, can be enhanced for tracking faster motions.

Conclusion

To sum up, the lab-designed 3D ultrasound tracking system has demonstrated its capability to reliably track the undetermined motion of the acoustically actuated 3D swimming microdrone in real-time. The average error of less than 0.3 mm, determined through comparisons with camera tracking across six different benchtop trials, provided evidence of the system's reliability.

Seamless integration with feedback-controlled propulsion positions the tracking system for enhanced capabilities. In biomedical applications, the designed tracking system offers a promising approach for three-dimensional motion tracking of microswimmers by leveraging the practical advantages of ultrasound. It holds the potential for advancing the clinical translation of microswimmer technology.