Review MRI (AM) content by training parameter-to-conclusion reasoning: classify weighting from TR/TE relationships, select sequences by their documented trade-offs, run a fixed four-step safety chain on every vignette, and identify artifacts by encoding direction. Finish each weekly cycle with a scored practice set and an error log.
Reading TR and TE Values to Infer Image Weighting
Weighting is inferred from the relationship between TR and TE, not from memorized numeric ranges. A long TR with a long TE signals T2 weighting; a short TR with a short TE signals T1 weighting.
The relationship matters because each parameter controls one contrast axis. TR governs how completely tissues recover longitudinal magnetization between excitations, so it sets the amount of T1 contrast available. TE governs how much transverse decay occurs before the echo is read, so it sets how much T2 contrast is expressed. A labeled example: TR 2500 ms with TE 100 ms lets nearly all tissues recover and then samples late decay differences, producing T2 weighting; TR 500 ms with TE 15 ms samples early, before decay separates tissues, while recovery differences remain large, producing T1 weighting.
The mistake to avoid is anchoring to absolute cutoffs. Recovery and decay times shift with field strength and tissue state, so a TR that is 'long' at one field strength may not erase T1 contrast at another. Instead, ask two questions about any pair you meet in practice: did this TR let most tissues recover (minimizing T1 contribution), and did this TE allow meaningful decay differences to develop (expressing T2 contribution)? Answering both questions in order converts every parameter pair into a defensible weighting call.
- Long TR: T1 contrast largely erased between excitations.
- Long TE: T2 decay differences fully expressed at readout.
- Short TR: recovery differences preserved, driving T1 contrast.
- Short TE: decay differences barely developed, suppressing T2 contrast.
Why Proton Density Weighting Breaks the T1-or-T2 Habit
Proton density weighting arises when a long TR removes T1 contrast and a short TE removes T2 decay differences, leaving measured signal to track hydrogen proton density alone.
The trap is binary thinking. If you sort every image into 'T1 or T2,' a long TR paired with a short TE looks contradictory: the TR says T2, the TE says T1. Resolving it requires treating the two parameters as independent switches rather than one combined label. Long TR empties the T1 axis; short TE empties the T2 axis; whatever remains — the density of hydrogen protons per unit tissue volume — becomes the dominant source of contrast.
Train this deliberately, because the same two-switch logic reappears everywhere in MRI: flip angle and receiver bandwidth in gradient echo, echo train length in fast spin echo, and inversion time in inversion recovery all follow the pattern of one parameter controlling one contrast or sensitivity axis. Sort each parameter pair you encounter into three bins — T1, T2, or proton density — and write the one-line reason for each bin. The habit of naming the controlling parameter prevents the binary shortcut from resurfacing under time pressure.
Choosing Between Spin Echo, Gradient Echo, and EPI Trade-offs
Sequences trade speed, contrast control, and sensitivity to field imperfections. Spin echo offers robust refocused contrast; gradient echo is fast with reduced RF power; echo planar imaging trades image fidelity for extreme speed.
Apply trade-offs mechanically in vignettes. If a scenario mentions metal near the region of interest, a gradient echo sequence's sensitivity to field inhomogeneity becomes the deciding cost — refocused spin echo behavior handles the distortion better. If a scenario flags RF power concerns at high field, recall that gradient echo uses small flip angles, which lowers RF deposition, while the 180-degree refocusing pulse in conventional spin echo adds to it. If a scenario demands speed above all, echo planar imaging's rapid readout is the fit, accepting its pronounced susceptibility distortion as the price.
Study each sequence by writing three lines: one strength, one cost, one situation where the cost decides the choice. Then link the lines to parameters — flip angle, echo train length, effective TE — so sequence selection becomes parameter arithmetic rather than a memorized list. That linkage is what lets you answer a vignette that changes one variable, such as a higher flip angle or a longer echo train, without starting the reasoning over.
| Sequence family | Contrast behavior | Speed | Key vulnerability |
|---|---|---|---|
| Conventional spin echo | Refocused, well-controlled T1 or T2 weighting | Slowest of the three | Long acquisition times; 180-degree pulse raises RF deposition |
| Gradient echo (spoiled and balanced) | T1- or T2*-dominated depending on flip angle and TE | Fast | Sensitive to susceptibility and field inhomogeneity; no true T2 refocusing |
| Echo planar imaging | Rapid multi-contrast acquisition | Fastest of the three | Severe susceptibility distortion and geometric warping near interfaces |
Chaining MR Safety Screening Decisions in Order
MR safety vignettes test ordered reasoning: verify screening data, classify the item or implant, check any conditions attached to it, then apply zoned access control before anyone approaches the bore.
Worked scenario 1: a patient with a history of metalworking and an old eye injury is scheduled, and tells you a previous scan elsewhere 'went fine.' The tempting move is to proceed on the patient's reassurance, or to order a radiograph for later. The better decision is to treat the orbital metal history as unresolved until it is cleared through your facility's documented pathway, and to keep the patient outside restricted zones in the meantime. It matters because the projectile hazard near the magnet is catastrophic and irreversible; no scheduling convenience justifies scanning before classification is complete.
Generalize that vignette into a reusable chain: identify the object or implant by name and model, classify it as MR safe, conditional, or unsafe, check any stated conditions such as a field strength or RF power limit, and document each step. A conditional implant is where the chain earns its keep — its stated conditions determine the correct action, and both refusing reflexively and proceeding unchanged skip the conditions check. Practice writing the four steps in order on every safety vignette until the sequence is automatic.
Explaining Why Gadolinium Brightens T1-Weighted Images
Gadolinium is paramagnetic; its unpaired electrons shorten the T1 relaxation times of nearby hydrogen protons, so tissues that accumulate the agent appear brighter on T1-weighted images.
The mechanism is local: the agent's magnetic field fluctuations accelerate longitudinal recovery of protons in its immediate vicinity, so enhancement appears where the agent distributes — in vascular spaces and in tissue regions the agent reaches. When a vignette gives tissue properties, such as vascularity or barrier characteristics, predict where enhancement concentrates by tracing the chain: agent accumulates there, T1 shortens locally, T1-weighted readout shows increased signal.
Two habits convert this from a fact into a usable skill. First, name the weighting explicitly: the brightening is a T1 effect, so it requires T1-weighted parameters to be visible — the same accumulation would not read the same way on a T2-weighted image. Second, when a vignette changes the agent's distribution (for example, describing where it does or does not accumulate), predict the enhancement pattern before reading the options. Direction-of-change reasoning, rather than memorized enhancement lists, holds up when the vignette presents an unfamiliar tissue.
Recognizing Artifacts From Direction and Pattern Clues
Artifacts are identified by location and direction: motion ghosts repeat along the phase-encoding axis, susceptibility voids cluster near metal or air, and chemical shift displaces fat-water boundaries along the frequency-encoding axis.
Worked scenario 2: an image shows a dark band offset at a fat-water interface, displaced along the frequency-encoding direction, with no repetition elsewhere. The tempting call is motion, and the tempting fix is padding, coaching, or repeating with breath-hold instructions. The better call is chemical shift misregistration, because the clue pair — a fat-water interface plus displacement along the frequency axis — matches that mechanism; motion, by contrast, produces repeating ghosts along the phase axis. The better fix is a higher receiver bandwidth or fat suppression. It matters because the wrong fix costs scan time and leaves the artifact untouched, while the right fix addresses the actual mechanism.
Build the identification habit as a two-step lookup: first locate the artifact relative to anatomy and hardware, then determine its direction relative to the encoding axes, and only then map both observations to a cause. Write the two observations down before naming the artifact in practice sets — it forces you to use evidence instead of pattern-matching on a single striking feature.
- Motion: repeating ghosts oriented along the phase-encoding direction.
- Chemical shift: fat-water boundary offset along the frequency-encoding direction.
- Susceptibility: signal void and distortion near metal, air, or tissue interfaces.
- Wrap or aliasing: anatomy from outside the field of view appearing inside it.
A Weekly Preparation Cycle With a Self-Check Rubric
Cycle weekly through parameter inference, sequence trade-offs, safety chains, and artifact mapping, then close each cycle with a scored mixed practice set and a written log of items you misread.
An adaptable sequence: spend two early days on weighting drills, sorting TR/TE and flip angle pairs into T1, T2, and proton density bins with written reasons. Add a day on sequence comparison using the table above, a day on safety vignettes where you write the four-step chain in order, and a day on contrast and artifact two-step identification. Close the week with a mixed set, then review the error log rather than the items you answered correctly. Repeat the cycle with harder vignettes and stricter time limits rather than restarting from new material.
Score each practice item against this rubric, one point each: correct weighting call with a named controlling parameter; correct sequence choice with its deciding trade-off; complete safety chain written in order; artifact identified with both location and encoding direction stated. Expected observations: in early cycles, weighting calls are right but the written reasons stay vague, and safety chains skip the classification step. A reasonable milestone is that by the third cycle you state the controlling parameter and the full chain before looking at answer options. Hitting these milestones is a learning checkpoint, not a prediction of any score.
- Readiness check: you classify any TR/TE pair into T1, T2, or proton density with a stated reason in seconds.
- Readiness check: you write the four-step safety chain from memory and apply it to a conditional-implant vignette.
- Readiness check: you name the encoding axis for at least three artifact types from a written description.
- Readiness check: you explain gadolinium enhancement as a T1-shortening chain, not a memorized brightness rule.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.