Skip to content

Paper Digest

🟢 This week — 2026-05-04

9 new papers in Tiers A–F → · top theme: Imaging & WSS measurement

Weekly roundup of cardiovascular CFD, hemodynamics, and AI modelling (PINN · neural operators · surrogates · digital twins). Auto-curated from OpenAlex + arXiv, classified by Claude into six tiers (A = boundary conditions, B = turbulence, C = V&V/UQ, D = physiology, E = imaging/WSS, F = AI/ML).

Subscribe: RSS feed · or bookmark this page.

Archive

  • 2026-05-04 — 9 papers · Imaging & WSS measurement
  • 2026-04-27 — 5 papers · Imaging & WSS measurement
  • 2026-04-20 — 2 papers · Turbulence modelling (RANS, LES, DNS)
  • 2026-03-20 — 26 papers · Imaging & WSS measurement

Methodology

  • Sources: OpenAlex (peer-reviewed) + arXiv (preprints), filtered to the last 7 days with cardiovascular + CFD + AI keywords.
  • Keyword filter: each paper must match at least one context term (aorta, hemodynamics, WSS, …) and one method term (CFD, LES, PINN, digital twin, …) in its title or abstract.
  • Classifier: Claude Haiku 4.5 scores relevance (0–1), assigns one of six tiers (A–F), and writes a 2-sentence summary plus a one-sentence why it matters note.
  • RAG enrichment:
    • Overlap flag — DOI match against the 108-paper bibliography (aortacfd_corpus.json).
    • Related reading — BM25 search of the new abstract against 1,466 indexed units (57 full-text papers + abstract-only entries for the rest) in BibCorpusStore.
  • Relevance threshold: 0.5. Below that, a paper is classified but dropped from the digest.
  • Published: every Monday at 09:00 UTC. Source code and Action in aortacfd-agent/paper_digest.

The six tiers

Tier Focus
A Boundary conditions — inlet/outlet BCs, Windkessel, lumped parameter, fluid-structure coupling
B Turbulence & LES — RANS, DES, LES, DNS applied to cardiovascular flows
C V&V and UQ — grid convergence, uncertainty quantification, sensitivity analysis
D Physiology & scaling — Murray's law, Windkessel theory, biomechanics
E Imaging & WSS — 4D flow MRI, Doppler echo, WSS measurement techniques
F AI, ML & digital-twin pipelines — PINN, neural operators, GNN surrogates, diffusion/transformer/foundation models, AI segmentation

Scope and limitations

Be aware of the trade-offs baked into this digest:

  • Abstract-only summaries. Claude reads title + abstract, not the full PDF. Summaries describe what a paper claims; they do not replace reading the methods and figures.
  • Research papers only. The feed does not surface industry announcements (Siemens, Dassault, ANSYS, Philips product blogs), non-indexed conference talks, GitHub releases of open-source CFD/ML code, or social-media discussion (Twitter/X, LinkedIn).
  • Anglophone bias. OpenAlex and arXiv both skew toward English-language work.

For deeper analysis of a single paper, a direct chat with Claude (with the full PDF) will always go further than this digest. For industry and tooling news, follow vendor blogs and GitHub trending directly.

Found something the digest missed? Open an issue with the DOI.

Found an issue or have a suggestion for this page?

Open a GitHub issue