Artifacts and pitfalls of imaging

From Wikiecho
Jump to: navigation, search

This chapter aims to classify and elucidate common but otherwise undesired alterations of imaging in everyday practice of echocardiography that impact on apprehension of either the anatomy or function of cardiac structures. Some of these conditions are easily recognized (e.g. shadowing) while others could generate significant misinterpretation of the final display and potentially affect the ensuing diagnostic potential of a study. These phenomena are due to ultrasound physics or operator interference.


Contents

Artifacts

Shadowing

This very common artifact might be characterized as the “Achilles heel” of echocardiography and occurs when the transmitted ultrasound beam encounters a medium of significantly different acoustic impedance and as a result is reflected back to the transducer. According to ultrasound physics the degree of transmission and reflection is related to the degree of acoustic mismatch between two media in close proximity[1]. This artifact is anticipated in highly reflecting artificial or biologic structures such as: mechanical valves (video 1) or calcium accumulation (video 2), etc.

Video 1: acoustic shadow produced by the presence of a mechanical aortic valve with otherwise low echogenic profile Video 2: descending thoracic aorta with mural thrombus. The acoustic shadow is produced by areas of focal calcification inside the thrombotic material

Reverberation

Reverberation (or reverb) is the persistence of sound in an enclosed space after its generation. It is relatively common artifact that limits the interpretation of the final imaging. This phenomenon is due to continuing repeated echoes between anatomic interfaces which are strong reflectors. Progressively the signal of the sound decays until it is beyond the range of detection of the transducer[2]. As time passes, the reflected volume of echoes is reduced and the phenomenon ceases. Reverberations are always seen away from a mechanical valve (video 3) or the interface between the parietal pleura or pericardium and the lung parenchyma (video 4).

Video 3: example of reverberation artefact from mechanical mitral valve prosthesis Video 4: reverberation artefact showing a second “phantasm” descending aorta. The phenomenon is being produced by the strong reflector parietal pleura adjacent to the posterior aortic wall.

Side lobes

The generation and transmission of the ultrasound beam from the transducers shows a pattern of “lobes” at various directions. The central ultrasound beam (main lobe) has the greatest energy and contributes principally to the final imaging. The other lobes are called “side lobes” and propagate to the side of the central beam in undesired directions producing the phenomenon of “edge effect”. Though the power density of the side lobes is generally much less than that of the main beam, it is usually adequate to produce artifacts. These artifacts created by the emitted side lobes reflecting to laterally positioned targets, are erroneously displayed and interpreted as if they originated from the central ultrasound beam. Since a requirement for an emerging side lobe artifact is a strong reflecting target, common sources of this significant problem in interpretation of imaging are the atrioventricular groove and the fibrous skeleton of the heart (video 5). Interestingly the same phenomenon is being admitted in antenna engineering and may also cause unwanted interference with radar receiver’s signals.

Video 5: subcostal projection with incidental demonstration of two aortic annuli (arrows) due to side lobe artefacts.

Near field clutter

This type of artifact is produced by the high-amplitude-oscillations of the piezoelectric crystals of the transducers. It has been practically vanished with the use of new echo systems with harmonic imaging, but in case it appears may mimic a possible apical left ventricular thrombus (video 6).


Video 6

Stitching artifact

This particular type of artifact emerged recently and is associated with 3D imaging. Three dimensional imaging though revolutionized echocardiography mostly by means of computer and circuit processing power has a major drawback coming from the relatively low speed of sound in soft tissues. In particular, there is insufficient time for sound to travel back and forth in large volumes while maintaining an adequate frame rate and reasonable resolution in live scanning modes. One maneuver to overcome this entails stitching multiple gates together to create a “full-volume” mode[3]. These gated “subvolumes” represent a pyramidal 3D dataset as would be acquired in the live 3D mode. This technique can generate 90° scanning volumes at frame rates over 30 Hz. However, the final imaging commonly shows undesired demarcation lines between the “stitched” subvolumes in the 3D image that have been named as “stitching artifacts” (video 7).

Video 7: 3D TEE study. In the lower left video (short axis projection) the interventricular septum shows an unusual interruption which should not be interpreted as interventricular defect but as a typical “stitching artifact”.

Pitfalls

Erroneous settings of the echo machine

Additional sources of limited imaging are the suboptimal or occasionally erroneous settings of the echo machine. In this context, after the appropriate position of the transducer several prerequisite decisions in the adjustment of the numerous imaging settings should be undertaken that can contribute to an improved quality of the display. The following table depicts various common problems encountered in daily practice and their suggested solutions.

Difficulty or dilemma Solution Comments
Obese patient with suboptimal imaging. Lower the transmission frequency of the ultrasound beam The lower the frequency of the transmitted sound the easier the penetration through media
Potentially two small structures lying along the axis of the beam cannot be differentiated Increase the transmission frequency of the ultrasound beam The higher the frequency and the shorter the pulse length or duration the better the axial resolution
In the apical 4-chamber projection the interatrial septum cannot be seen clearly (video 8,9) and may mimic PFO or ASD 1.Use modified views
2.Increase the gain of the received signal
3.Inject agitated saline as a contrast with Valsalva maneuver
Drop out phenomenon due to low thickness of the interrogating beam (fig 1).
The moving cardiac structures in real time scanning are aesthetically of rather low quality Lower the sector width of the display and / or the depth of field The lower the sector width and depth the higher the frame rate of 2D imaging
Chronic mitral regurgitation that seems to be severe though the left ventricle is not dilated Adjust properly the color gain settings of the machine The brightness (amplification) of the primary color which encodes velocities can be increased by the operator to unacceptable levels (video 10).
Patient with myocardial infarction and “good” myocardial perfusion after injection of contrast Lower the Color Doppler amplification of the machine (Doppler gain button) This key action of the operator demands continuous infusion of contrast and adequate experience (video 11 and 12)
Patient with no history of myocardial infarction and abnormal perfusion of the apex in rest images The focus button should be adjusted to the level of the mitral valve The higher the density of the ultrasound beams on the structure of interest the higher the contrast destruction
Fig1.jpg
Fig 1: schematic representation of drop out phenomenon. The bullet (ultrasound beam) passes through the narrow horizontal lines (reflector surfaces) with no collision (left). With a modified “shot” the bullet (sound) impacts on the oblique lines (reflectors); a prerequisite of imaging (right).
Video 8: patient with pulmonary arterial hypertension and a possible stretched PFO most possibly due to drop out phenomenon. Video 9: after injection of agitated saline with synchronized Valsalva maneuver there was no evidence of PFO or ASD.
Video 10: different color gain settings in a patient with moderate MR due to mitral prolapse. The Doppler gain should be adjusted with caution since the regurgitation might be under- or overestimated. Video 11: bolus infusion of contrast with unadjusted (higher) color Doppler gain in a patient with recent anterior myocardial infarction. The final display gives the erroneous information that the infracted region is still viable
Video 12: continuous infusion of contrast in the same patient with adjusted color Doppler gain. The apical myocardium is underperfused, an indication of non viability.

Erroneous measurements

Quantification of cardiac chamber size and function along with valvulopathies’ assessment are the central and most frequently requested indications of echocardiography. However non-standardized measurements have been often contradictory and less successful compared to other imaging techniques and therefore echocardiographic measurements are sometimes perceived as inconsistent. These particular types of errors are the commonest encountered even in modern echo labs equipped with new advanced technology ultrasound devices. For this reason, accuracy and reproducibility are factors that should be taken into consideration. Lack of standardization usually heralds great variation regarding the reliability of measurements. Once guidelines have been established, good reproducibility has been achieved and used in different clinical settings [4]

Example 1: 38 year old female with mitral prolapse (video 13). Two contradictory measurements with either M-mode (fig 2) or 2D technique (fig 3) are shown. In the first attempt the M-mode vertical interrogation axis is clearly improperly positioned overestimating the LV dimension (video 14).

Fig2.jpg Fig3.jpg
Fig 2: The end diastolic diameter of the left ventricle assessed with M-mode was 63 mm (wrong measurement). Fig 3: In the same patient the end diastolic diameter of the left ventricle assessed with 2D was only 46 mm.
Video 13 Video 14

Example 2: 70 year old male with mitral stenosis (video 15). Two contradictory measurements of mitral valve anatomic orifice are shown with conventional 2D echocardiography (fig 4, 5). A third attempt with 3D echocardiography by using the oblique plane on the 3D acquired data set correctly assessed the smallest opening area of the leaflets and actual MV orifice area [5] (fig 6).

Fig4.jpg Fig5.jpg
Fig 4: First attempt of measuring the mitral valve orifice (1,3 cm2, moderate stenosis) Fig 5: Second attempt of measuring the mitral valve orifice (1,8 cm2, mild stenosis)
Fig6.jpg
Fig 6: 3D measurement in the same patient (1,7 cm2, mild stenosis) Video 15

References

  1. Anderson B. Echocardiography. 2nd edn. Australia: MGA Graphics; 2007
  2. Feigenbaum’s H. Echocardiography. 6th edn. Philadelphia: Lippincott Williams & Wilkins; 2005
  3. Vegas A, Meineri M. Three-Dimensional Transesophageal Echocardiography is a major advance for intraoperative clinical management of patients undergoing cardiac surgery: a core review. Anesth Analg 2010;110:1548 –73
  4. Lang R, Bierig M, Devereux R, et al. Recommendations for chamber quantification. Eur J Echocardiography (2006) 7, 79-108
  5. Zamorano J, Cordeiro P, Sugeng L, Perez de Isla L,Weinert L, Macaya C, et al. Real-time three-dimensional echocardiography for rheumatic mitral valve stenosis evaluation: an accurate and novel approach. J Am Coll Cardiol 2004;43(11):2091–6

External links

European Association of Echocardiography. Physics of echocardiographic imaging. http://assets.escardio.org/Assets/Webcasts/EE2009/echocardiography-course/02-Echocardiographic-imaging-physics/index.htm

Personal tools
Namespaces

Variants
Actions
Navigation
Toolbox