Radiology Writing

Radiology-Style Manuscript Writing

Use this skill to construct imaging-research prose — argument first, then sentences — in the exact shape Radiology requires. It is for drafting and restructuring, not only polishing (for sentence-level polish use radiology-polishing).

Core stance

• Author evidence first. Never invent results, metrics, p-values, cohort numbers, citations, mechanisms, or limitations. Missing input → explicit placeholder or a question.
• Write the argument before the sentences. One-sentence claim → section architecture → paragraph jobs → prose.
• Match the Radiology shape. Structured abstract, Summary statement (one sentence), Key Results (≤ 3, ≤ 75 words), structured Discussion, tight word/figure limits.
• Bound the claim. Ambitious but evidence-bounded; calibrate verbs to evidence (demonstrate → suggest → may reflect).
• Reporting-aware. Every Methods/Results element should satisfy the relevant checklist item (CLAIM/TRIPOD+AI/CLEAR/STARD) — cross-check with radiology-reporting.

When to use

• Draft/rebuild any section: title, structured abstract, Summary statement, Key Results, Introduction, Materials and Methods, Results, Discussion.
• Turn Chinese lab notes / mixed drafts into submission-ready English.
• Restructure a rejected draft to the Radiology argument shape.

When to open extra files

Intake (identify before drafting)

• Target venue/shape: Radiology-family (structured abstract, Summary statement, Key Results — default) or Nature-family (unstructured abstract, no Summary statement/Key Results → references/nature-family-shape.md). If undecided, draft the Radiology shape first and confirm venue before finalising the abstract (→ radiology-journal).
• Section(s) requested.
• Study type: diagnostic-accuracy, prediction model, radiomics, radiogenomics, reader study, observational, trial.
• Core claim: what the study actually shows.
• Evidence: cohorts (n, source, dates), metrics with CIs, comparisons, validation.
• Boundary: where the claim stops (single-centre? retrospective? prevalence?).
• Limits: target word/figure counts (verify against current author instructions).

If core claim, evidence, or boundary is missing, surface the gap and offer a scaffold with placeholders rather than inventing content.

Writing workflow

0. For full manuscripts or high-impact rebuilds, open argument-spine-and-stage-gates.md and establish the project context, contribution-first gate, and results-as-validation map before drafting long prose.
1. For target-venue writing taste, open journal-family-writing-style.md after confirming the venue family; use it to adjust article shape, title/abstract rhythm, key points, and clinical relevance language.
2. One-sentence argument: "In [population/modality], we show [advance] using [approach], supported by [key result with CI], with [boundary]."
3. Pick the architecture (article-architecture.md) by study type.
4. Map each paragraph to one job: context / gap / objective / design / cohort / technique / analysis / result / comparison / validation / interpretation / limitation.
5. For full sections, draft the topic-sentence chain first. If the claims do not flow, revise the chain before writing full paragraphs.
6. Draft from evidence outward — keep claims next to the numbers that support them.
7. Calibrate verbs to evidence; remove unsupported novelty/"first" claims.
8. Fit the Radiology shape — abstract headings, Summary statement, Key Results, structured Discussion.
9. Self-review against the relevant reporting checklist; flag unmet items.

Section defaults

• Title — concrete: population/condition + modality/method + finding/role. State AI/ radiomics if central. Avoid slogans and "novel."
• Structured abstract — Background → Purpose → Materials and Methods → Results → Conclusion; report design, cohort sizes/dates, primary metric(s) with CIs, a bounded conclusion. (structured-abstract.md)
• Summary statement — a single declarative sentence of the main finding.
• Key Results — up to 3 results/conclusions, ≤ 75 words, with summary data; don't repeat the Summary statement; avoid vague language and abbreviations.
• Introduction — field/clinical stakes → specific gap → objective/hypothesis. Short; no results dump.
• Materials and Methods — design + ethics/registration → participants/flow → imaging technique → image analysis/reference standard/readers → model/feature pipeline → statistical analysis. (methods.md)
• Results — patient flow + characteristics → primary performance with CIs → comparisons → validation/subgroups. Past tense, quantitative. (results.md)
• Discussion — first paragraph = concise summary of key findings, then relation to prior work, then limitations, then a bounded conclusion. (discussion.md)

Output format

1. Draft — the requested prose in Radiology shape.
2. Section outline — 3–7 compact bullets (for a full section).
3. Topic-sentence chain — for full sections or major rewrites, the claim sequence before paragraph expansion.
4. Claim–evidence map — Claim | Evidence (with CI) | Status: supported / needs input.
5. Stage gates — for full manuscripts: contribution gate, results-as-validation gate, citation/figure/reporting gates that are passed or still open.
6. Venue style check — if the target family is known: abstract shape, title/key-points logic, clinical relevance language, and any guide-derived limits or VERIFY FROM GUIDE.
7. Assumptions / missing inputs — only material gaps.
8. Reporting check — checklist items this draft does/doesn't satisfy (→ radiology-reporting).

For Chinese notes: polished English first, then brief Chinese notes on structural choices.

Handoffs

• Sentence-level polish / house style → radiology-polishing.
• Statistics/CIs/tests behind the numbers → radiology-stats.
• Checklist compliance → radiology-reporting.
• Figures/legends → radiology-figure.
• Finding/verifying citations to support a claim in Introduction/Discussion → radiology-citation.
• Draft is complete and ready for a harsh read before submission → radiology-prereview.