Implementation and validation of three automated methods for measuring ultrasound maximum depth of penetration: Application to ultrasound quality control

Abstract

Systematic measurements of maximum depth of penetration (DOP) of ultrasound (US) scanners are essential for quality control (QC). Conventionally, DOP measurements are performed visually and as such they could be affected by various external factors, scanner control settings, and operator related errors. Automated methods should be free of the issues associated with interoperator dependence and are an attractive alternative to the visual DOP measurements. We implement and test three automated methods for measuring DOP. The methods base their measurements on signal to noise (SNR) analysis of uniform US phantom images. Two of the methods use pairs of phantom images. The third one uses a single phantom image and an "in-air" image. The validation tests included precision, sensitivity, repeatability, and usability in routine QC application. Methods based on pairs of phantom images measure the DOP with precision +/-0.2 cm or better. Precision of the single phantom image method is +/-0.05 cm, and that method is also the most sensitive of the three. All three methods are demonstrated to be repeatable among different users. Since the images for the DOP computation are collected free-hand the sensitivity to hand-transducer motion during image acquisition was also tested. Unlike the single-phantom-image based method, the methods using image pairs were found to be very sensitive to transducer motion and therefore less convenient for clinical QC applications. In conclusion, the single-phantom-image method is best suited for routine QC in a real-life clinical practice.

Four-Year Experience with a Clinical Ultrasound Quality Control Program

“…Because of the subjective nature of this approach, it can be relatively time consuming. We now make routine QC measurements of DOP automatically using custom software to analyze pairs of images, one ''in-air'' with no coupling gel on the transducer and the other of a gel phantom (Gorny et al 2005;Stekel et al 2007). The program determines noise as a function of depth by computing the average value of all pixels at each distance from the transducer face in the in-air image.…”

A novel approach to a phantom based method for maximum depth of penetration measurement in diagnostic ultrasound: a preliminary study

“…In scientific literature the maximum depth of ultrasound signal visualization (or penetration depth) is often related to the maximum depth at which echo signals from scattering within a tissue-mimicking phantom can be distinguished from electronic noise [10][11][12][13]. In order to evaluate the parameter above, some theoretical considerations can be done on ultrasound image and characteristic quantities that contribute to propagation of the ultrasound wave in tissues.…”

“…Among technical requirements for an ultrasound diagnostic system a very important role is played by the maximum depth range at which echoes from weakly reflecting scatterers in tissues can be detected above noise, therefore the evaluation of the parameter above is very critical to determine both the diagnostic capabilities and the quality status of the medical system [9]. Nevertheless most of the objective methods for maximum depth of signal visualization measurements are based on noise level estimation [10][11][12][13], that is often not univocally defined for ultrasound imaging. In this work some theoretical considerations on the parameter are illustrated and two methods have been implemented and compared for three diagnostic systems: one of them is developed by the Authors and does not require the noise level estimation.…”

Evaluations of UltraiQ software for objective ultrasound image quality assessment using images from a commercial scanner

“…Thijssen et al reported and implemented a test protocol including a series of quantitative measurements of overall dynamic range, sensitivity, contrast resolution, and spatial resolution 3. Other methods for quantitatively assessing metrics such as resolution integral, depth of penetration (DOP), entropy, correlation, clutter, and spherical target detection have also been proposed 4, 5, 6, 7, 8, 9…”