Multi-Echo fMRI Posters at OHBM 2026

A searchable static index of every multi-echo fMRI poster at the OHBM 2026 Annual Meeting (Bordeaux, France · June 14–18, 2026), for the multi-echo community.

👉 Open index.html — one HTML file, no build step (fonts and icons load from CDN).

Currently 114 posters match out of 3,219 total. Each card shows the poster number, title, authors (lead author highlighted), a collapsible abstract (with its figures), the poster's session days, "why it matched" tags, and a link to the official OHBM floq.live entry (and poster PDF once uploaded).

The interface follows an editorial, Dieter Rams–style minimalism (DM Serif Display titles, hairline-separated cards, amber-as-signal). It offers live search (title / author / keyword), filters by session day and match method, sort by number or title, an accent-color picker, and auto / light / dark theme — accent and theme persist in localStorage.

Unofficial community resource. Not affiliated with or endorsed by OHBM. Data is pulled from the public OHBM 2026 conference app.

How it works

build_site.py (Python standard library only — no dependencies) does everything:

1. Fetch every "Poster Abstract" from the OHBM conference AppSync GraphQL API (getTimeslotsWithType, paginated; the full raw set is cached to posters_raw.json.gz, which is committed so --no-fetch works after a clone).
2. Filter titles + abstracts for multi-echo fMRI content (see below).
3. Enrich matches with session day / keywords / categories (resolved via the API's classifier batchGet).
4. Generate posters.json (clean data) and index.html (the self-contained site).

python build_site.py            # fetch (or reuse cache), filter, build the site
python build_site.py --refetch  # force a fresh fetch from the API (~75 s, ~65 pages)
python build_site.py --no-fetch # rebuild from the cache only (instant — for tuning)

Matching logic

"multi-echo" alone is a generic MRI term (QSM, relaxometry, MRS), so the filter is two-tiered to target multi-echo fMRI:

• ME-fMRI-specific (always qualifies): tedana, ME-ICA, ME-fMRI / ME-EPI.
• fMRI-gated (qualifies only when the abstract also has fMRI/BOLD context): multi-echo, echo/optimal combination, T2*/R2* mapping, TE-dependent, S0.

This biases toward recall — a missed poster is worse than one extra to skim — and every card shows why it matched so you can judge at a glance. To tune the patterns, edit multiecho_reasons() in build_site.py and re-run --no-fetch (no re-fetch).

A note on scheduling

Exact per-poster presentation times aren't published this far out (the API's start/end are empty). Each poster's session days come from the conference's Date classifier; use the floq.live link on a card for the authoritative schedule.

Publish to GitHub Pages

index.html is at the repo root, so Pages can serve it directly:

git remote add origin git@github.com:<you>/ohbm-2026-multi-echo-posters.git
git push -u origin main

Then on GitHub: Settings → Pages → Build and deployment → Source: Deploy from a branch → Branch: main / / (root) → Save. The site appears at https://<you>.github.io/ohbm-2026-multi-echo-posters/ within a minute.

Files

File	Description
index.html	The self-contained website (poster data inlined).
posters.json	The 114 matched posters with parsed fields.
build_site.py	Fetch + filter + generate script (stdlib only).
posters_raw.json.gz	Full raw dump of all 3,219 posters (~14 MB, committed for future use).