Echo Image Optimization

What looks wrong? Tap it

2D
Colour
Doppler
Planned

Archives usually store at 30 Hz: in younger children, quantification may be better done on the machine at the acquisition frame rate (ASE 2024 pediatric).

All problems and fixes in one table
Try the steps in order. Adult marks a value from an adult guideline. Artifact fixes: ASE 2026 (adult-oriented, physics applies at any age).
Looks likeTry, in orderWatchSource
Too dark or too bright
  1. Overall gain: just a few echoes in the blood, crisp blood–endocardial borders
  2. TGC: even brightness from top to bottom; uneven pods cause bands
  3. Dynamic range (compression), see the next problem
  4. Brighten with receiver gain before output power; raise power only when the signal is weak against noise (educational site)
In 3D, dropout from low gain or uneven TGC cannot be recovered later ASE 2019; EAE/ASE 2012; echocardiographer.org
Haze or noise in a cavity
  1. Lower the gain
  2. Harmonic imaging (fewer side-lobe echoes)
  3. Focus at the depth of interest
  4. Change the window or angle
Near-field clutter can mimic an apical LV thrombus: it ignores wall motion and crosses the wall; check with colour at a low scale, other planes, contrast, CT or CMR ASE 2026 §3.2, 3.3, 9
Blurry, little detail
  1. Highest transmit frequency that still reaches the structure
  2. Focus at the structure
  3. Less depth; structure centred
  4. Acquisition (preprocessing) zoom, not zoom on the stored image
  5. Window closest to the structure; measure along the beam
Higher frequency penetrates less. Lateral resolution falls with depth; axial is better than lateral ASE 2019; ASE 2024 pediatric
Flat gray or harsh black and white
  1. Dynamic range (compression): lower for a poor window (more contrast), higher to see subtle tissue (more grays)
  2. The lab’s gray map, kept the same for serial studies
Too few grays can hide a thin wall, thrombus or vegetation; too many look washed out. In coronaries, extra compression may hide a luminal thrombus ASE 2019; ASE 2024 pediatric
Jerky, low frame rate
  1. Less depth
  2. Narrower sector
  3. One focal zone
  4. Acquisition zoom
  5. Fewer scan lines, where the system allows
  6. Persistence (frame averaging) low for the heart
Fewer lines or a narrower sector cost lateral resolution or field of view. Children’s faster heart rates need higher frame rates than adults’ ASE 2019; ASE 2024 pediatric; Nestaas 2018; Zander 2020 (background)
Ghost or double structure
  1. Change the window or angle
  2. Lower the gain
  3. Harmonic imaging (side lobe)
  4. Focus at the depth of interest; higher frequency (beam width)
Reverberation ignores anatomic boundaries; an artifact moves with the structure that makes it, not on its own. If unresolved, use 3D, contrast, CT or CMR ASE 2026 §2–3
Shadow or missing wall
  1. Image from a window where the structure is perpendicular to the beam
  2. Change the plane or angle around a shadowing reflector
  3. Colour Doppler to tell a true defect from dropout
A septum parallel to the beam can drop out falsely; use more than one view ASE 2024 pediatric; ASE 2026 §2.5, 6.2, 9, Table 2
Colour bleeds past the flow
  1. Colour gain: raise until speckle appears outside the flow, then lower until it just disappears
  2. Raise the wall filter (blooming)
The same jet looks larger at a lower Nyquist and higher colour gain; record blood pressure with jet size ASE 2019; ASE 2026 §5.7
Unwanted colour aliasing
  1. Raise the scale (Nyquist)
  2. Shift the baseline
  3. Less depth
  4. Lower transmit frequency
Aliasing is used on purpose for PISA (baseline shifted toward the jet) ASE 2026 §5.6, 10
No colour in slow flow
  1. Lower the Nyquist to about 30 cm/s for atria and pulmonary veins Adult
  2. Low wall filter
  3. Colour gain up to just below noise
Routine jets use 50–70 cm/s Adult. No pediatric guideline gives a number ASE 2019; ASE 2002
Colour frame rate low
  1. 2D sector shallow and narrow first
  2. Smallest colour box that covers the flow
  3. Side-by-side colour compare only when needed
In children keep the colour frame rate ≥20 Hz ASE 2024 pediatric Table 7; ASE 2019
Jet in an odd place
  1. Another window or angle
  2. Lower colour gain
  3. Adjust the focus and sweep for an out-of-plane jet
Suspect an artifact when a jet has no flow convergence or vena contracta ASE 2026 §5.1–5.5
Spectral aliasing
  1. Raise the scale
  2. Shift the baseline away from the flow
  3. Shallower sample (higher PRF)
  4. Lower transmit frequency
  5. High PRF or CW
CW and high PRF cannot say where along the beam the velocity is; some systems switch to high PRF when the scale is raised ASE 2026 §4.3; ASE 2019
Weak or incomplete envelope
  1. Align with flow, colour first
  2. A little more spectral gain
  3. Nonimaging CW probe
  4. Other windows
Measure only a well-defined envelope; not faint signal outside it ASE 2002; ASE 2019
Velocity looks too high
  1. Lower the spectral gain
  2. Measure the modal (densest) envelope, outer edge
  3. At a CW peak, ignore the fine linear “beard”
  4. If two jets overlap, use a smaller sample, avoid far field, other window
Overgain overestimates velocity; so does measuring contrast blooming ASE 2026 §4.1, 4.5, 4.6, 7.3; ASE 2002
Velocity looks too low
  1. Align with flow from several windows (nonimaging probe)
  2. Report the highest-velocity window; don’t average windows
  3. No angle correction
Underestimation 6% at 20°, 50% at 60°; the gradient error is larger (4V²) ASE 2026 §4.4; ASE 2019; ASE 2002
Clutter or gap at the baseline
  1. Wall filter: show the start and end of the flow
  2. Low for slow flow (for example pulmonary veins); higher for fast jets
A high filter removes the slow parts of the signal ASE 2019; ASE 2026 §4.2
Tissue Doppler looks wrong
  1. Tissue Doppler preset
  2. Low gain
  3. Small sample at the annulus; neonates 2 mm
  4. Beam parallel to the wall
  5. Neonates: scale ±0.16 m/s; freeze the 2D while recording
High gain overestimates pulsed TDI. Colour TDI reads about 20% lower than pulsed; not interchangeable ASE 2024 neonatal; Nestaas 2018
Strain or 3D to follow
  1. Frame rate: targets on Strain; narrow sector, less depth
  2. Same vendor and software for follow-up
  3. 3D: optimise 2D first; gain and compression mid-range, slightly higher TGC Adult; breath-hold; fewest beats
Strain from copies stored at 30 frames/s differed from acquisition-rate images abstract ASE 2024 pediatric; ASE/EACVI 2025; EAE/ASE 2012; Koopman 2011

Children and neonates

Settings by mode

Presets and auto-optimise buttons are starting points. Preprocessing = set while acquiring; postprocessing = changed after storage (ASE 2019).

2D (grayscale)
ASE 2019 Adult, unless another source is named
ControlSetting
Transmit frequencyStart high; lower only for penetration; highest possible throughout. Adult range 2.0–5.0 MHz
Harmonic imagingUsually the second harmonic (twice the transmit frequency): borders enhanced, low-level noise removed. Use with caution in coronaries (ASE 2024 pediatric)
Depth and sectorFill the display with the structure; extra depth costs frame rate or line density
FocusAt the depth of the structure; more focal zones lower frame rate
Overall gainJust a few echoes in the blood; crisp blood–endocardial borders
TGCEven brightness with depth; misplaced pods make bands
Dynamic range (compression)Lab default; lower for difficult windows, higher for more grays
Gray map, colourisationGray map chosen with the vendor’s specialist and kept constant; B-mode colourisation (tint) changes perception, not information
ZoomAcquisition zoom rescans a smaller region, so resolution improves and frame rate can rise; zoom on the stored image only magnifies
M-mode
  • Sweep: 100–200 mm/s for timing; slower to see several beats Adult (ASE 2019).
  • Steerable (anatomic) M-mode: built from the 2D image, so its temporal and range resolution are no better than 2D (ASE 2019).
  • Neonatal TAPSE: tissue Doppler–enhanced M-mode with a narrow sector, >200 frames/s (ASE 2024 neonatal, Table 3).
Spectral Doppler (PW, CW)
ASE 2002; ASE 2019 Adult
ControlSetting
AngleParallel to flow, colour-guided; no angle correction. <20° off gives <10% velocity error but more gradient error (ASE 2002). For neonatal LVO, as close to 0° as possible, definitely <20° (ASE 2024 neonatal)
Scale and baselineSignal as large as possible without aliasing; don’t cut off flow in the other direction
Sweep speed100 mm/s default, 2–3 beats per sweep; all velocity and time measurements at ≥100 mm/s; 25 mm/s with a respirometer for respiratory change
Sample volumeSmall enough to limit spectral broadening; for flow (VTI), 5–7 mm long at the site of the diameter measurement
Wall filterShows the start and end of the signal; low for slow flow, higher for fast jets
GainFull spectrum without noise; a smooth envelope with a clear modal velocity
CWNo Nyquist limit, no depth information; nonimaging probe for the highest velocity (called a Pedoff probe on an educational site)
High PRFTwo gates double the Nyquist; velocity could come from either gate
Colour Doppler
ASE 2019 Adult; ASE 2024 pediatric
ControlSetting
OrderColour before spectral, to place the sample and align the beam (pediatric)
Sector and box2D sector as shallow and narrow as possible, then the smallest box over the flow; frame rate ≥20 Hz in children
Colour gainUp until speckle appears outside the flow, down until it just disappears; readjust often
Nyquist50–70 cm/s for routine jets Adult; about 30 cm/s for atria and pulmonary veins Adult; jets look larger at a lower scale
Colour mapOne map for the whole lab
Colour compareIt lowers frame rate, so use it sparingly; helps small low-velocity structures (pulmonary veins, coronaries) (pediatric)
Tissue Doppler
  • Adult Preset: larger sample than blood PW, scale below 25 cm/s, measured at 100 mm/s (ASE 2019). Sample <5 mm, low gain and filter (ASE 2002).
  • Neonates: gate 2 mm just below the annulus, beam parallel to the wall; a very narrow sector raises frame rate for time intervals (ASE 2024 neonatal). Angle <20°, scale ±0.16 m/s (avoids aliasing in most neonates), low gain, 2D frozen, sample 2–3 mm (Nestaas 2018).
  • Colour TDI: sector width changes frame rate more than depth; fewer beams raise frame rate but cost lateral resolution; heavy time averaging hides isovolumic phases (Nestaas 2018).
  • Normal values: Systolic, Diastolic.
Strain (acquisition only)
  • Children: faster heart rates need higher frame rates; reproducible on one platform, poor across platforms (ASE 2024 pediatric). Neonatal frame-rate targets: Strain.
  • Adult Frame rate: generally >40 frames/s (ASE 2019); RV-focused view 60–90; LA 50–90 (ASE/EACVI 2025).
  • Follow-up: same vendor and software release where possible (ASE/EACVI 2025).
3D
  • Before acquiring: optimise 2D, put the region of interest in the axial plane, then choose: volumes need temporal resolution, septal defects and outflow tracts spatial, valves a balance (ASE 2024 pediatric).
  • Gated multibeat: best of both, but needs stillness and breath-hold (hard in young children) and fails with ectopy; 3D colour usually has low temporal resolution in children (ASE 2024 pediatric).
  • Adult Gain and compression: mid-range (50 units) with slightly higher TGC, so the volume can be adjusted later; smallest volume that covers the structure (EAE/ASE 2012; ASE 2019).
  • Stitching: breath-hold, fewest beats, single-beat or 2-beat acquisition (ASE 2026).

After the scan

What can and cannot change after storage
  • Our note: frame rate, depth, sector, focus, frequency, harmonics, acquisition zoom, Doppler angle, sample position and size are preprocessing settings (ASE 2019) and are not in the vendor’s after-storage list, so get them right while scanning. In 3D, dropout from low gain or uneven TGC cannot be recovered (EAE/ASE 2012). Jet size depends on the Nyquist and gain used (ASE 2019).
  • Adjustable on the vendor’s own stored data: for example, on one system (2021 specification): 2D gain, dynamic range, gray map, Chroma map and zoom; colour baseline, map, write priority, blend and variance; spectral gain, compress, reject, baseline and sweep speed (vendor terms).
  • Archived loops: usually stored at 30 Hz, so quantification in younger children may be better at acquisition (ASE 2024 pediatric). Strain was somewhat higher from GE images at the acquisition frame rate than from the same images stored at 30 frames/s (49 children; Koopman 2011 abstract).
How to measure: Doppler and strain conventions
  • PW: outer edge of the dense (modal) envelope; trace VTI on the modal velocity, ignoring dispersion near the peak (ASE 2002).
  • CW: outer border of a well-defined envelope only; at an aortic stenosis peak, ignore the fine linear signals (“beard”) and measure the outline (“chin”) (ASE 2026 §4.5).
  • Windows: report the highest-velocity, best-quality signal; don’t average across windows (ASE 2019).
  • Beats: sinus rhythm 3–5 averaged; irregular rhythm such as atrial fibrillation 5–10. Constriction or tamponade: respiratory trace, inspiration and expiration averaged separately (ASE 2002).
  • Sweep: velocity and time measurements at ≥100 mm/s Adult (ASE 2019).
  • Speckle tracking: check the tracking against the wall motion; smoothing can hide a failed segment (ASE/EACVI 2025). More on Strain.
  • Linear and area measurements: edge and timing conventions differ by dataset; see Chamber Size and Echo Z-Scores.

Artifacts

Artifact, clue and fix (ASE 2026)
ASE 2026 Table 1 and Sections 2–8. Artifacts break the machine’s assumptions: a straight path, one round trip, 1,540 m/s, echoes only from the main beam, uniform attenuation.
ArtifactClueFix
Reverberation, comet tailRepeats deeper along the beam; ignores anatomic boundaries; prosthetic valves, wires, aortic rootChange angle and probe position; lower gain; may not go away: 3D, contrast, CT or CMR
Mirror imageA copy deeper than a strong reflector (pericardium, pleura)Find the reflector; optimise gain; change plane or angle
Shadowing, enhancementDark beyond calcium or a prosthesis; bright beyond fluidChange plane or angle; TEE or another modality
Speed displacementStructure displaced behind fat (lipomatous septum)Hard to avoid; expect misplacement and measurement error
RefractionSide-by-side double (double aortic valve)Other window or plane; adjust the angle
Beam width, slice thicknessGhost echoes in a cavity; a lead in the wrong chamber; size overestimatedFocus at the depth; other windows; lower gain; higher frequency
Side lobeCurved or straight line symmetric about a bright reflector (false flap)Lower gain; harmonic imaging; other angle or window
Near-field clutterApical haze that ignores wall motionColour at a low scale, other planes, contrast; CT or CMR
Spectral mirror, double envelope, tiger stripesCopy across the baseline; two envelopes; bands (an oscillating structure)Lower gain and power; align; adjust the plane; look for the oscillating leaflet
3D stitching, dropoutSteps between sub-volumes; ill-defined “holes” where a structure runs along the beamBreath-hold, fewest beats or single-beat; slightly more gain, optimise compression, image perpendicular, colour
EquipmentLines or noise in every view (cautery, a second ultrasound system, a faulty probe)Swap the probe; no simultaneous cautery or second system; shield electrical sources

Useful artifacts: aliasing makes the PISA shell; colour splay can point to significant eccentric MR (ASE 2026 Table 3). Contrast agents have their own artifacts (Section 7).

Resolution

Axial, lateral, elevational, temporal, contrast
ASE 2024 pediatric, ASE 2019, ASE 2026; physics background from Zander 2020 (general ultrasound review) and echocardiographer.org (educational site)
ResolutionImprove byCost
Axial (along the beam)Higher frequency, shorter pulse; measure along the beam, where resolution is bestLess penetration
Lateral (across the beam)Focus at the structure; nearest window (the beam spreads with depth); narrower sector, more lines; harmonics (fewer side lobes)More lines or a wider sector lower frame rate
Elevational (slice thickness)Focus at the structure; higher frequency; thinnest at the focus, thicker with depthOut-of-plane ghosts and overestimated size when thick
Temporal (frame rate)Less depth, narrower sector, one focal zone, acquisition zoom, fewer lines, smaller colour box, colour compare off; low persistenceLateral resolution or field of view
ContrastDynamic range, gray map, gain, frequencyToo few grays hide subtle tissue

Processing controls and vendor terms

Generic controls: persistence, speckle reduction, compounding, reject
Background from Zander 2020, a general (mostly abdominal) ultrasound review; no echo guideline read defines these
ControlWhat it doesCaution
Persistence (frame averaging)Blends earlier frames into the current one; smoother imageCan hide abnormalities; keep low for the heart
Speckle reductionSmooths the grainy textureEdges less sharp; details of 1–2 mm may be lost
Compounding (spatial, frequency)Combines images from several angles or frequencies; less speckle and noiseCan remove useful shadowing and enhancement
RejectVendor term in post-processing listsNo guideline read defines it; see the system’s manual
Vendor-named controls
Philips Affiniti CVx specifications (4522 991 71451, September 2021), the manufacturer’s own description; model- and release-specific
NameVendor’s descriptionGeneric function
XRES, Variable XRESSpeckle noise reduction; the user picks levels of noise reduction, edge enhancement and textural smoothing, from crisp to smoothAdaptive speckle reduction with edge enhancement
SonoCTBeam-steered real-time compound imaging, up to nine lines of sight; listed for linear, curved, tightly curved and volume arrays (no cardiac sector probe listed)Spatial compounding
iSCAN, AutoSCANOne-button TGC and gain; frame-by-frame automatic TGCAutomatic gain and TGC
LGCLateral gain compensation on cardiac sector probesGain by sector band
ChromaColourised luminance mapsB-mode colourisation (tint)
Image BoostSuppresses clutter and boosts myocardial signal (named adult cardiac probes)Clutter suppression

Adjustable after storage on that system: 2D gain, dynamic range, inversion, zoom, gray map, Chroma map; colour baseline, invert, map, hide, write priority, blend, variance, zoom; spectral gain, compress, invert, baseline, angle correct, display format, sweep speed, reject, Chroma map. Other vendors’ terms (for example GE DDP, UD clarity, Tint) are not shown: no manufacturer-hosted document for them was read. A vendor’s description states intent, not measured effect.

Which knob to turn when the image looks wrong, settings by mode, children and neonates, artifacts and post-processing. No figures: see the figures in ASE 2026 and ASE 2019.

Key points

Sources and how this page was checked

Guidelines (full text read)

Other sources

How it was checked