Cohort and question framing
从临床问题、样本来源、终点和可解释性出发,先确定分析边界,再进入数据处理。
Medical bioinformatics
Jiang Congxin
整形外科 · 单细胞与空间组学 · 医学 AI
I work at the intersection of plastic surgery, wound repair, scar biology, public datasets, and reproducible omics analysis. My current focus is turning clinical questions into traceable analysis workflows, publication-ready figures, and practical research tools.
A practical toolkit for medical research projects that need clear data provenance, interpretable biology, and figures that can survive review.
从临床问题、样本来源、终点和可解释性出发,先确定分析边界,再进入数据处理。
处理单细胞、空间组学、转录组和公开数据库结果,保留可追溯的中间表和 QC 记录。
把 marker、通路、通讯、轨迹和模型结果整理成论文级图表、方法记录和复用代码。
Daily literature notes updated on 2026-05-12.
Plastic surgery and wound repair
A burns and plastic surgery-linked team reports a diabetic wound dressing that responds to glucose and ROS while suppressing macrophage cGAS-STING inflammation.
Open notes02Scar and keloid biology
A high-value JID review reframes keloids as a multi-compartment disease involving fibroblasts, inflammation, adipocytes, genetic risk, and skin-of-color biology.
Open notes03Medical AI and clinical decision support
A large pragmatic Nature Medicine trial shows that AI worklist prioritization can shorten report time yet still fail to move downstream CT, diagnosis, treatment, or staging outcomes.
Open notes04Public database mining
CGPA is useful for public database mining because it makes prognosis queries more explicit across cancer type, context, and validation logic instead of relying on isolated Kaplan-Meier screenshots.
Open notes05Single-cell and multi-omics analysis
A Nature four-omics paper pushes single-cell analysis toward simultaneous regulatory measurement, useful for moving scar/wound atlases from cell labels to causal hypotheses.
Open notes06AI-assisted scientific writing
A recent Nature Communications RAG paper is relevant to scientific writing because it treats hallucination as a retrieval-structure problem, not just a prompting problem.
Open notesOpen to clinical omics projects, reproducible figure workflows, and method-oriented collaborations in wound repair, scar biology, and medical AI.