01Principles and purposeThe professional or clinical skill and the decisions it supports.
Imaging is a test, not a photograph of truth. Its value depends on whether the chosen examination can detect the suspected process in this patient at this stage. Plain radiography shows projectional density differences; ultrasound depends on acoustic windows and operator technique; CT uses ionising radiation and can resolve anatomy rapidly; MRI provides tissue contrast without ionising radiation but has access, time and safety constraints. Asking “what decision will this change?” prevents unfocused requests and incidental findings from displacing the presenting problem.
A disciplined reading begins outside the anatomy. Confirm identifiers, acquisition date and clinical indication, then modality, view or sequence, side marker and comparison. Assess whether the whole required region is included and whether positioning, inspiration, penetration, motion or contrast phase is sufficient. Technical defects are findings because they change sensitivity. If a portable anteroposterior chest radiograph magnifies the heart, apparent cardiomegaly should not be interpreted as if it were an erect posteroanterior film.
Separate observation from inference. “A peripheral triangular opacity at the right base” is an observation; pulmonary infarction, infection and atelectasis are interpretations requiring context. State site, size, number, margin, density or signal, associated effects and interval change. Then synthesize: which diagnosis best explains the pattern, what dangerous alternative remains, and is the image concordant with the patient? This sequence exposes assumptions and creates a report another clinician can challenge.
Structured search reduces omissions but does not remove cognitive bias. Satisfaction of search occurs when an obvious abnormality stops further review; framing occurs when the request anchors the reader; confirmation bias privileges supportive signs. Counter these by completing the search after finding a lesion, reviewing hidden areas, checking priors and asking what would be expected if the leading hypothesis were false. Incidental findings need proportionate action rather than reflex investigation.
Responsibility continues after acquisition. The requesting team must identify who will review the result, how an urgent alert will be received, what happens after transfer or discharge, and how recommended follow-up is tracked. A radiology report expresses the radiologist’s synthesis and limitations; clinicians must integrate it with physiology and trajectory. When report and bedside picture conflict, direct radiologist discussion can clarify protocol, uncertainty and the next best test.
Key points
- Start with identity, date, modality, body region, projection or phase and comparison studies; an image belonging to the wrong person or episode cannot answer the question.
- Define the clinical question before inspecting abnormalities, because modality, protocol and pre-test probability determine what a negative or positive study can mean.
- Judge technical adequacy before interpretation: coverage, positioning, exposure, motion, contrast timing and artefact can conceal disease or simulate it.
- Use a fixed search pattern that covers expected anatomy and common review areas, then return to the clinical question and examine any discordant region again.
- Describe observations before naming a diagnosis, localise precisely, compare with prior imaging and rank a short differential by probability and consequence.
- A normal image does not exclude disease outside the modality’s sensitivity or the imaged time point; escalate when clinical concern remains high.
- Read the authorised report, act on urgent communication, document the plan and close every recommendation for further imaging or specialist review.
02Situations and prioritiesThe context, relevant information and actions that matter most.
Verify patient, date, modality, body part, side, projection, phase and whether the study is preliminary or final before interpreting anatomy.
Look for missing coverage, rotation, poor inspiration, incorrect exposure, motion, metal artefact or mistimed contrast that changes diagnostic confidence.
Use a reproducible modality-specific sequence and deliberately inspect review areas where abnormalities are commonly overlooked.
Describe location, morphology, distribution and secondary effects before attaching a pathological label or causal story.
Test whether the imaging explains the symptoms, examination, laboratory data and trajectory; discordance is a prompt to reassess.
Distinguish an unreported image, preliminary interpretation, authorised report, critical alert and recommendation awaiting action.
03Assessment and interpretationHow to gather information, assess the situation and recognise uncertainty.
Consider the information, its meaning and its limitations before deciding what follows.
- 01
Clinical question and prior probability - Why
- Define the diagnosis or decision the image should address and identify time-critical alternatives.
- Interpretation and limitations
- A focused question lets the radiologist choose protocol and determines whether a negative result reduces probability enough to stop or requires another test.
- 02
Acquisition and quality check - Why
- Confirm projection, contrast phase, coverage, positioning and artefact before reading pathology.
- Interpretation and limitations
- If quality is inadequate, explicitly limit interpretation and discuss repeat or alternative imaging rather than calling the study normal.
- 03
Systematic image review - Why
- Inspect expected anatomy in a fixed order, including devices, soft tissues, bones and common blind spots.
- Interpretation and limitations
- Record objective findings and complete the entire search even after a striking abnormality is found.
- 04
Comparison with prior studies - Why
- Compare like modality and technique where available, using dates and interval treatment.
- Interpretation and limitations
- New, stable, resolving and progressive appearances have different implications; technique differences can mimic interval change.
- 05
Authorised report and alerts - Why
- Read the final report and determine whether an urgent finding or follow-up recommendation was communicated.
- Interpretation and limitations
- Translate the report into an owned clinical action, document who is responsible and reconcile later amendments.
- 06
Radiation justification - Why
- For ionising examinations, ensure the request is justified and optimised through the designated IR(ME)R roles.
- Interpretation and limitations
- A clinician supplies accurate information and the clinical question; justification is not replaced by a generic request or patient pressure.
04Worked approachesCases with ordered reasoning, an action and a check of the outcome.
01Worked caseA discordant chest radiographA 71-year-old develops sudden pleuritic pain and hypoxaemia. A portable AP chest radiograph is labelled “no focal consolidation”; pulmonary embolism remains clinically plausible.+
- 1Verify identity, timing, AP projection and image quality, then inspect lungs, pleura, mediastinum, bones and devices systematically rather than equating no consolidation with no acute disease.
- 2Represent the question as acute hypoxaemia with pleuritic pain and a radiograph that lacks an alternative explanation; the study was not designed to exclude pulmonary embolism.
- 3Reassess physiology and calculate the NICE two-level pulmonary embolism Wells score. If PE is likely, arrange immediate CTPA, or assess suitability for V/Q imaging when contrast allergy, severe renal impairment or high radiation risk changes the choice; follow the NICE interim anticoagulation pathway if imaging cannot be obtained immediately.
- 4Read and act on the definitive report, record the decision and contingency, and verify that any recommended imaging or treatment decision has an identified owner before transfer.
02Applied approachReading a new focal opacityAn erect chest radiograph shows a previously unreported left basal opacity in a patient with fever and weight loss.+
- 1Confirm technique and compare prior studies before deciding the opacity is truly new.
- 2Describe its exact site, margin, air bronchograms, volume loss, pleural change and any associated hilar or bony abnormality.
- 3Integrate tempo and risk factors to rank infection, collapse and malignancy rather than choosing solely from the request wording.
- 4Act on immediate illness and arrange the report-recommended follow-up, with a named person responsible for checking resolution.
03Safety processClosing an imaging recommendationA CT report recommends interval imaging after discharge, but the patient has moved between teams.+
- 1Read the authorised report and clarify the recommendation, urgency and intended clinical question.
- 2Contact the responsible team directly and document transfer of the task rather than assuming an electronic result is enough.
- 3Tell the patient what follow-up is planned in accessible language and how they will hear the result.
- 4Check the tracking system until booking, completion and review are confirmed, escalating any failure.
05Feedback, follow-up and evidenceReview outcomes, seek feedback and identify what to improve.
- Recheck the final authorised report after any preliminary bedside interpretation and reconcile discrepancies explicitly.
- Track critical alerts, addenda and recommendations through to documented action, including after discharge or team transfer.
- Review clinical trajectory after a negative or indeterminate examination; worsening physiology can invalidate reassurance.
- Audit rejected, cancelled and technically limited examinations because each may leave the original clinical question unanswered.
- For repeated ionising imaging, reconsider whether the question, interval and modality justify another exposure.
06Special situationsVariants, exceptions and circumstances that change the usual approach.
Negative is conditional
A negative test is reassuring only to the degree allowed by the protocol, technical quality, disease stage and pre-test probability.
Projection changes appearance
Magnification, rotation and supine positioning alter apparent heart size, fluid distribution and lung volumes; name the projection before inference.
Incidental findings need ownership
An incidental abnormality can be low probability yet still require planned review; record the action and responsible clinician.
Comparison is an investigation
Interval change can discriminate active from chronic disease more strongly than a single appearance, provided technique is comparable.
Direct dialogue resolves ambiguity
A radiologist can explain wording, protocol limits and the most informative next test when the report and bedside concern diverge.
Examination transfer
In PACES or CPSA, narrate quality and observations briefly, then give synthesis, key limitation and the next action without overcalling certainty.
Protocol is part of the test
A CT with or without contrast and at different phases answers different questions; state the intended decision so acquisition matches it.
Devices require a second question
After checking position, look for device-related complication and whether the device path or tip can be assessed confidently on this study.
07Common pitfallsFrequent interpretation and management errors.
- 01
Calling an image normal before checking identifiers, projection and adequacy.
- 02
Using the request wording as the diagnosis and searching only for confirming signs.
- 03
Stopping after the first abnormality and missing a second lesion or complication.
- 04
Treating absence of a radiographic sign as exclusion of an early or poorly visualised disease.
- 05
Quoting a report without integrating current observations, physiology and interval change.
- 06
Leaving an urgent alert, amended report or follow-up recommendation without a named owner.