Contrast response analysis for medical ultrasound imaging

Traditional measures of spatial resolution are helpful but incomplete for characterizing the imaging performance of medical ultrasound systems because the challenge is not so much resolving strong point reflectors as limiting the impact of stronger scattering regions on weaker regions. This paper discusses the use of an alternative measure which is based on the ability of a system to accurately image an anechoic region surrounded by a uniform scattering medium. This measure, called contrast response, includes effects of both mainlobe width and sidelobe level. Since the impulse response of a linear system to a target is the convolution of the system spatial impulse response (beam pattern) and the target scattering function, the system output can be described by a three dimensional integral. The contrast response is defined as the ratio of this system output when the anechoic target is not present to the output when it is present, evaluated over a range of target sizes. Contrast response analysis is useful for both system design and performance comparison. Examples are presented which compare the performance of systems with different apodisation functions and aperture sizes. This analysis approach suggests why the strong window functions popular in signal processing have not been used for apodization in medical ultrasound and why 256 channel systems have not demonstrated a dramatic performance improvement over 128 channel systems.