Automatic Detection of Acoustic Window During Routine Echocardiographic Imaging

Martin Andersen1, Johannes Struijk2, Samuel Emil Schmidt2
1AAU, 2Aalborg University


Abstract

Ultrasound is a powerful tool in cardiac diagnostics, providing both anatomical and physiological information, which may be used for describing cardiac health. However, the modality is well known for the low image quality, especially compared to other modalities such as Computed Tomography. However, ultrasound provides real-time images, and can provide temporal resolution comparable to the electrocardiogram, at the cost of image quality. A problem with generating echocardiographic images is locating the optimal acoustic window for generating echocardiographic images, which is affected by several anatomical structures, such as the narrow spacing between ribs. This manuscript describes an algorithm to detect the ribs automatically, capable of providing clinicians real-time information regarding the acoustic window between the ribs. The algorithm was tested in an in-house tissue mimicking phantom, with specular objects the size of ribs, with high acoustic impedance, were used to mimic the acoustic shadows of ribs in the phantom. The algorithm was capable of automatically detecting and visualising the rib-sized objects, and the accompanying shadow artefacts. An algorithm such as the one presented in this manuscript may provide improvements to current ultrasound systems by providing clinicians information regarding capturing an optimal acoustic window. Furthermore, the algorithm may be implemented for dynamic transmit-receive operations to reduce shadow and reverberation artefacts.