OBJECTIVE MEASUREMENTS OF IMAGE QUALITYINTRODUCTIONMETHODSImage quality testsEquipmentTissue-mimicking phantomImage capture and computerised analysisAxial and lateral resolution, and slice thicknessContrast resolutionAnechoic target detection abilityLow contrast penetration depthRESULTSAxial resolutionTHI modeSonoCT modeHSonoCT modeLateral resolutionTHI modeSonoCT modeHSonoCT modeSlice thicknessTHI modeSonoCT modeHSonoCT modeContrast resolutionTHI modeSonoCT modeHSonoCT modeAnechoic target detectionTHI modeSonoCT modeHSonoCT modeLow-contrast penetration depthTHI modeSonoCT modeHSonoCT modeDISCUSSIONCONCLUSIONSdoi:10.1016/j.ultrasmedbio.2003.10.002● Original ContributionOBJECTIVE MEASUREMENTS OF IMAGE QUALITYJACINTA E. BROWNE,* AMANDA J. WATSON,* NICHOLAS M. GIBSON,†NICHOLAS J. DUDLEY†and ALEX T. ELLIOTT‡*Ultrasound Equipment Evaluation Project, and†Department of Clinical Physics, Western Infirmary, Glasgow, UK;and‡Department of Medical Physics, Nottingham City Hospital, Nottingham, UK(Received 23 April 2003; in final form 2 October 2003)Abstract—Tissue harmonic imaging (THI) and compound imaging have been reported clinically to improvecontrast resolution, tissue differentiation and overall image quality. However, there have been limited studies todate to quantify objectively the improvements in image quality achieved with these new imaging techniques. Theaim of this study was to quantify differences in image quality that exist between conventional B-mode imaging,harmonic imaging, compound imaging and harmonic compound imaging. An ATL HDI 5000 scanner with threeprobes (C5-2, L7-4 and L12-5) was tested with two different types of test object, the Gammex-RMI model 404GS LE and the Gammex-RMI 403 GS LE. The measurement limitations associated with subjective analysismethods were not present in this study because an automated image analysis program was used to determine theimage quality parameters. Therefore, subtle differences between the four imaging modes could be detected.Significant improvements in lateral resolution and slice thickness as a function of depth were found with THI.Contrast resolution and anechoic target detection improved with compound imaging, and harmonic compoundimaging improved lateral resolution, slice thickness as a function of depth and contrast resolution. (E-mail:[email protected]) © 2004 World Federation for Ultrasound in Medicine & Biology.Key Words: Computerised image analysis program, Tissue harmonic imaging, Compound imaging, Harmoniccompound imaging, Quality control.INTRODUCTIONIn recent years, diagnostic ultrasound (US) image qualityhas greatly improved, due to advances in technology andintroduction of new techniques, such as compound im-aging and tissue harmonic imaging (Claudon et al. 2002).Developments in transducer design have resulted intransducers with greater band width and sensitivity(Averkiou et al. 1997). Broadband transducers allowlarge band width pulses to be generated, which is vitalfor pulse-inversion tissue harmonic imaging and alsoleads to improvements in axial resolution. Tissue har-monic imaging (THI) is a real-time US imaging tech-nique that uses echoes at twice the transmitted frequencyto form the image, unlike conventional B-mode imagingthat uses echoes at the transmitted frequency to form theimage (Desser and Jeffrey 2001). The higher frequencyharmonic signal is not present in the transmitted signal,but is generated as it propagates through tissue due to thephenomenon of nonlinear sound propagation. Imagingwith the harmonic signal means that the harmonic beamonly passes once at echo reception through the body walland any fat that is present (Desser and Jeffrey 2001). THIhas been reported to improve image quality, particularlyin clinical applications such as cardiology and abdominalimaging (Desser and Jeffrey 2001; Li and Zagzebski2000; Tranquart et al. 1999; Averkiou et al. 1997; Chris-topher 1997; Ward et al. 1997).Real-time compound imaging has been made possi-ble, only recently, due to the substantial computationalpower of modern all-digital US systems, despite its prin-ciples and benefits being known since the 1950s (Howry1955). There are two types of compound imaging, spatialcompound imaging and frequency compound imaging.Spatial compound imaging involves acquiring a numberof coplanar images of the same object from differentangles, and then combining them into a single compoundimage (Entrekin et al. 2000). Frequency compoundinginvolves acquiring images of the same object at differentfrequencies and then combining them into a single com-pound image (Gatenby et al. 1989). Spatial compoundAddress correspondence to: Dr. Jacinta Browne, UltrasoundEquipment Evaluation Project, Dept. of Clinical Physics, GroundFloor, 38 Church Street, Glasgow, G11 6NT UK. E-mail:[email protected] in Med. & Biol., Vol. 30, No. 2, pp. 229–237, 2004Copyright © 2004 World Federation for Ultrasound in Medicine & BiologyPrinted in the USA. All rights reserved0301-5629/04/$–see front matter229imaging has been reported to improve contrast resolu-tion, tissue differentiation and overall image quality inbreast imaging, vascular imaging and musculoskeletalimaging (Entrekin et al. 2000, 2001; Jespersen et al.2000). Despite the above clinical studies being con-ducted to evaluate the improvements in image quality,there have been limited technical evaluations to providequantitative information about improvements in imagequality with THI, compound imaging and the newertechnique of harmonic compound imaging that combinesboth methods and was introduced by ATL-Philips. Theaim of this study was to quantify differences in imagequality that exist between conventional B-mode imaging,THI, compound imaging and harmonic compound imag-ing.METHODSImage quality testsThe image quality parameters that were measured inthe study were: 1. axial resolution; 2. lateral resolution;3. slice thickness; 4. contrast resolution; 5. anechoictarget detection; and 6. low-contrast penetration depth.These image quality test parameters were chosen fortesting the US scanner to obtain a full evaluation of itsimaging capabilities, as recommended by a number ofprofessional bodies (Price 1995; Goodsitt et al. 1998).EquipmentThe US scanner used in this study was the HDI5000 (Philips/ATL, Bothell, WA), a high-end system,dedicated for general US examinations with a 2- to5-MHz (C5-2) curvilinear transducer, a 4- to 7-MHz(L7-4)