← 返回线下交流
Clinical AI · Research · PublishingInside: the 6-step workflow · 六步工作流

OiOi MedAI


临床 AI 科研写作特刊 · November 2026 · Issue 01

Perioperative Medicine journal cover
IN REVIEW在审
Brazilian Journal of Anesthesiology, Vol 76 No 4 issue cover
IN REVIEW在审
Scientific Reports journal cover

ACCEPTED已接收

Scientific Reports · IF 4.9 · JCR Q1
BMC Anesthesiology journal cover

ACCEPTED已接收

BMC Anesthesiology · IF 3.2 · JCR Q2

In review5 manuscripts across 4 journals · 5 篇在审

From Question to Acceptance从研究问题到期刊接收

2SCI ACCEPTED2 篇接收

How an anesthesiologist built an AI-assisted, author-verified workflow — and took two papers to acceptance in 2026. 一位麻醉医生如何用 AI 辅助、作者全程核验的工作流,2026 年拿下两篇 SCI 接收。

Lixin Yin, MD · 殷利鑫

Journal covers © their respective publishers, shown for identification only.

Read the issue 翻开本期
A night operating-room monitor with waveforms and floating manuscript pages

From Question to Acceptance从研究问题到期刊接收

From Research Question to Accepted Paper从研究问题,到 SCI 接收


Two SCI acceptances in 2026, five manuscripts in peer review — built with an AI-assisted, author-verified writing workflow. 2026 年 2 篇 SCI 接收、5 篇在审——由 AI 辅助、作者全程核验的科研写作工作流完成。

AI-generated illustration

2 SCI papers accepted已接收
5 Manuscripts in peer review在审稿件
4.9 Highest journal IF · Scientific Reports, JCR 2025最高影响因子
4 Open clinical databases: VitalDB · INSPIRE · MIMIC-IV · eICU-CRD公开临床数据库

Accepted已接收

Two papers, two journals, one year.一年之内,两本期刊,两篇接收

A glass kidney linked to two databases

Accepted

AI-generated illustration

Scientific ReportsNature Portfolio

Data Readiness and Cross-Database Transportability of Creatinine-Based AKI Prediction in the ICU ICU 中基于肌酐的急性肾损伤预测:数据就绪度与跨数据库可迁移性

DOI 10.1038/s41598-026-77681-5 · Online publication pending / 即将在线发表

In plain language

Asks whether an AKI prediction model built on serum creatinine in one ICU database still works in another — and shows that data readiness (what is recorded, when, and how it is mapped) decides transportability as much as the model itself. 研究基于血肌酐的 AKI 预测模型能否从一个 ICU 数据库迁移到另一个,指出数据“就绪度”(记录什么、何时记录、如何映射)与模型本身同样决定可迁移性。

Journal profile

Journal Impact Factor
4.9 · 2025, JCR 2026 release
5-year IF
4.8
JCR quartile
Q1 · Multidisciplinary Sciences

Nature Portfolio open-access journal; 2nd most-cited journal in the world (1,061,414 citations in 2025).

Review timeline · 3 rounds

  1. Minor revision
  2. Major revision
  3. Accepted
  4. Licence & proofs completed
A blood-pressure cuff, an arterial line and an anesthesia monitor

Accepted

AI-generated illustration

BMC AnesthesiologyBMC, Springer Nature

Measurement of Intraoperative Hypotension Exposure and ICU Resource-Use Associations: A Retrospective Methodological Cohort Study Using VitalDB 术中低血压暴露的测量方式与 ICU 资源使用关联:基于 VitalDB 的回顾性方法学队列研究

In plain language

Using 5,825 surgical cases from VitalDB, shows how the choice of MAP threshold, invasive vs. non-invasive monitoring and sampling frequency changes who is labeled “hypotensive” — and why severity, duration and cumulative burden matter more than a single binary cutoff. 基于 VitalDB 5,825 例手术,揭示 MAP 阈值、有创/无创监测与采样频率如何改变“低血压”判定,强调严重度、持续时间与累积负荷比单一阈值更重要。

“All nine of Reviewer 1’s points and all six of Reviewer 2’s points … appear to have been substantively addressed, not just acknowledged.”

— Peer reviewer, BMC Anesthesiology, Sep 2026

Journal profile

Journal Impact Factor
3.2 · 2025
5-year IF
3.2
JCR quartile
Q2 · Anesthesiology
First decision
15 days, median

Indexed in MEDLINE, PubMed Central and SCIE.

Review timeline · 5 rounds

  1. Major revision
  2. Minor revisions: 21 Sep, 1 Oct, 19 Oct, 6 Nov 2026
  3. Accepted (Beijing time)
  4. Publishing agreement signed

Under review在审

Five more manuscripts in the pipeline.另有五篇稿件正在同行评审中

Research questions only — results stay under embargo until publication.

  1. Under review在审BMC AnesthesiologyIF 3.2In re-review

    Event-Level Discordance Between Intermittent Non-Invasive and Continuous Invasive Arterial Blood Pressure for Intraoperative Hypotension Labeling in VitalDB

    Major revision 26 Oct 2026 → revised manuscript returned 27 Oct 2026

    Asks

    How often do cuff (NIBP) and arterial-line readings disagree on whether a hypotensive event happened?无创袖带血压与有创动脉压在“是否发生低血压事件”上的事件级不一致有多大。

  2. Under review在审Scientific ReportsIF 4.9Revision in progress

    Prediction-Time Provenance Stress Testing for Cross-Database ICU Models: Timestamp Availability, Semantic Mapping, and Bidirectional Validation in MIMIC-IV and eICU-CRD

    Major revision 30 Oct 2026

    Asks

    Stress-tests whether ICU models only use data truly available at prediction time, validated in both directions across MIMIC-IV and eICU-CRD.对 ICU 模型做“预测时点数据溯源”压力测试,检验时间戳可得性与语义映射,并在 MIMIC-IV 与 eICU-CRD 间双向验证。

  3. Under review在审Perioperative MedicineIF 2.5Under consideration

    Cross-Representation Agreement of a Frozen Operation-Window MAP Phenotype Rule: A Linked VitalDB–INSPIRE Study

    Submitted 18 Jul 2026

    Asks

    Does one pre-specified intraoperative MAP phenotype rule give consistent labels across two different data representations?同一条预先冻结的术中 MAP 表型规则,在 VitalDB 与 INSPIRE 两种数据表征下是否一致。

  4. Under review在审Scientific ReportsIF 4.9Awaiting editorial update

    LabTraj-ICU: A Staged Feasibility Study of Self-Supervised Laboratory Trajectory Representations Across MIMIC-IV and eICU

    Minor revision (Jul 2026)

    Asks

    Can self-supervised learning turn ICU lab trajectories into reusable representations across databases?自监督学习能否把 ICU 检验轨迹转化为跨库可复用的表征。

  5. Under review在审Brazilian Journal of AnesthesiologyElsevierCo-author · 合作作者Under review

    Incremental Predictive Value of Preoperative Coagulation Variables for Intraoperative Red-Cell Transfusion in Elective Non-Cardiac Surgery: A Grouped Cross-Validation Study

    Submitted Aug 2026

    Asks

    Do preoperative coagulation tests add real predictive value for intraoperative transfusion?术前凝血指标对术中红细胞输注的增量预测价值。

The workflow工作流

One Workflow. Every Number Verified.一套工作流,每个数字都经核验

A researcher's desk at dusk with a laptop and printed proofs
AI-generated illustration
  1. 01

    Question选题

    Gap scan of the literature and public data.

  2. 02

    Data数据

    Open databases (VitalDB, INSPIRE, MIMIC-IV, eICU-CRD), frozen analysis rules.

  3. 03

    Analysis分析

    Pre-specified, reproducible code; patient-grouped validation; sensitivity analyses.

  4. 04

    Drafting写作

    AI-assisted drafting in journal format (sci-writing-harness agent workflow).

  5. 05

    Evidence check证据核查

    Every number traced to output tables; every reference verified.

  6. 06

    Peer review审稿回复

    Point-by-point responses, revision tracking, through to acceptance.

Proof成果

Sci Rep: 3 review rounds → accepted · BMC Anesthesiology: 5 review rounds → acceptedSci Rep 3 轮审稿→接收 · BMC Anesthesiology 5 轮审稿→接收

AI assists; the author designs, verifies and takes responsibility for every analysis and sentence, in line with journal AI-use policies. AI 负责辅助,研究设计、每项分析与每句表述均由作者核验并负责,符合期刊 AI 使用政策。

Next step下一步

Want this workflow for your department?想把这套工作流带到你的科室?


Follow the work and get in touch through GitHub and OiOi MedAI Academy.