Low-cost 3D vision-based triangulation system for ultrasonic probe positioning

Introduction: In ultrasonic imaging, such as echocardiography, accurately positioning the probe in relation to the patient's body or an external coordinate system is typically done manually. However, when developing speckle-tracking methods for echocardiology, ensuring consistency in probe positioning is essential for reliable data interpretation. To address this challenge, we present a vision-based system and method for probe positioning in this study. Materials and Methods: Our system comprises two cameras, a calibration frame with eight markers of known coordinates in the frames' local coordinate system, and a probe holder with four markers. The calibration process involves image segmentation via region growing and extraction of the camera projection matrices. Subsequently, our positioning method also utilises marker segmentation, followed by estimating the markers' positions using triangulation. Results: To evaluate the system's performance, we conducted tests using a validation plate with five coplanar circular markers. The distances between each pair of points were calculated, and their errors compared to the true distances were found to be within a maximum of 0.7 mm. This level of accuracy is comparable to ultrasonic imaging resolution and thus deemed sufficient for the intended purpose. Conclusion: For those interested in replicating or modifying our methods, the supplementary material includes the complete design of the calibration frame and the Matlab code.

---

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki (2025).