CT common mistakes
CT · Understand the traps and how to avoid them.
Memorizing pitch and mAs formulas as isolated facts and then applying them to the wrong scan mode, for example using the helical pitch formula on an axial acquisition or ignoring that pitch does not change patient dose in the same way it changes tube output.
Candidates drill formulas from flashcards without connecting them to the acquisition geometry, so the variable definitions blur under time pressure and the wrong formula gets selected.
What to do instead
Build a one-page sheet that pairs each formula with a labeled sketch of the geometry (axial vs helical, detector width, table feed) and practice three numeric problems per mode, writing the units at every step so mismatched variables are caught immediately.
Confusing CT dose descriptors: reporting CTDIvol as if it were the patient dose, or mixing up CTDIw, CTDIvol, and DLP, and forgetting the dose-length product is CTDIvol multiplied by scan length.
The terms sound similar and are often presented together in tables, so candidates memorize definitions without a mental model of what is measured where in the phantom.
What to do instead
Draw the phantom: CTDIw is a weighted average across the 100 mm pencil chamber, CTDIvol divides by pitch, DLP multiplies by scan length. Then work backward from a DLP value to scan length and from CTDIvol to pitch to confirm the relationships stick.
Treating window width and window level as interchangeable or inverting their effects, for example widening the window to increase contrast or raising the level to brighten the whole image.
Both controls change displayed grayscale, and the exam often asks about the resulting appearance rather than the definition, which exposes a shallow understanding of the linear mapping.
What to do instead
Use the formula: width spans the HU range displayed, level is its midpoint. Practice predicting the appearance change for a given tissue (for example a 40 HU lesion) at several width/level pairs, and state the window in HU before answering.
Failing to adjust contrast timing and injection protocol for patient-specific factors such as cardiac output, weight, and venous access site, then misreading a suboptimal bolus as a scanner or reconstruction problem.
Protocols are often memorized as fixed recipes, so candidates do not connect physiology and access route to the timing of the arterial or portal venous phase.
What to do instead
For each common study, write the trigger method, the expected phase timing, and two patient factors that would shift it. Practice deciding whether to use bolus tracking, a test bolus, or a fixed delay for a low-output or upper-extremity-access patient.
Missing safety screening details before contrast: not checking renal function when indicated, not documenting prior reactions or asthma, and not verifying metformin or pregnancy status when the protocol requires it.
Screening is treated as a checkbox rather than a clinical decision, and candidates under exam conditions skip the step that determines whether contrast or premedication is appropriate.
What to do instead
Memorize a short screening checklist in order (indication, allergy/reaction history, renal function criteria, pregnancy, metformin, access) and rehearse it against vignettes, stating explicitly which finding changes management and why.
Misapplying CTDIvol and DLP reference values or size-specific dose estimates when comparing a protocol to diagnostic reference levels, for example comparing a pediatric head study to adult reference values.
Reference levels are stratified by body region and patient size, and candidates often grab the first number they remember without matching the population.
What to do instead
Keep a small table of reference levels by region and patient size, and for each practice question write the population and region before selecting a value. Confirm whether the question asks about CTDIvol (per-slice) or DLP (whole exam).
Mixing up reconstruction algorithm effects: attributing increased noise to a sharper kernel, or expecting a smooth kernel to improve spatial resolution, and failing to recognize how iterative reconstruction changes noise texture rather than just lowering it.
Kernel and reconstruction choices are learned as settings rather than as trade-offs, so the direction of the noise-resolution relationship is guessed.
What to do instead
Make a two-column table for sharp versus smooth kernels and for filtered back projection versus iterative reconstruction, listing noise, resolution, and artifact effects. Practice predicting which kernel a bony or high-contrast study needs and why.
Ignoring artifacts caused by patient positioning, metal, or motion and instead blaming the equipment, for example calling beam-hardening streaks a detector failure or misidentifying out-of-field artifact from arms left in the scan range.
Artifact recognition is visual and pattern-based, and candidates who have not categorized artifacts by cause default to a generic equipment explanation.
What to do instead
Group artifacts by cause (physics, patient, scanner, reconstruction) and for each write the appearance, the mechanism, and one mitigation. Practice naming the cause from a written description before choosing a fix.
Overlooking radiation protection practices specific to CT: not using automatic exposure control correctly, not adjusting for patient size, and not shielding or collimating where appropriate, especially in pediatric and repeat examinations.
Protection content is often studied as general ALARA principles rather than as concrete CT workflow decisions, so candidates cannot translate principle into action.
What to do instead
For each protection tool (AEC, tube current modulation, bowtie filter, collimation, iterative reconstruction), write what it does, when it applies, and one pitfall. Practice justifying a dose-reduction choice for a specific patient scenario in two sentences.
Rushing the image evaluation step: accepting an image without checking for motion, incomplete coverage, contrast opacification, or correct labeling, and then answering questions about image quality without first establishing whether the image is diagnostic.
Candidates treat evaluation as a final formality and jump to technical explanations before confirming basic adequacy, which leads to choosing an answer that fixes the wrong problem.
What to do instead
Adopt a fixed evaluation order (coverage, contrast phase, motion, artifacts, labeling, window) and apply it aloud to every vignette before selecting an answer. Write down the first abnormal finding and only then consider its cause.