Brummel/alpha-id
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
17 Commits 1 Branch 0 Tags
6647a66bd08ec6d503d5a18540f85de36e4fcef3
Commit Graph

17 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Brummel 6647a66bd0 feat: Mode C pipeline end-to-end (lexical + tags + fusion) 
Implement Pipeline::load/suggest for Mode C: lexical search per segment,
cross-segment dedup via fusion, tag derivation, filter application.
Add meta_lookup fallback (exact → strip trailing 0 → 3-char root) to
bridge the ICD code granularity gap between the corpus (E11.90-style
6-char codes) and ClaML (which only stores E11, E11.0-style entries).
 2026-05-18 17:04:34 +02:00
Brummel 8d9ab0e299 feat: RRF and cross-segment max-score dedupe 2026-05-18 16:54:25 +02:00
Brummel a896c5bbde fix: store IndexReader once; sanitize all tantivy query specials 
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-18 16:51:48 +02:00
Brummel 6dd1c17306 feat: tantivy lexical index over Alpha-ID texts 2026-05-18 16:44:51 +02:00
Brummel ce7403fee9 feat: tag derivation from ClaML metadata + validity 2026-05-18 16:40:08 +02:00
Brummel 27d5144864 feat: dictation segmentation on blank lines 2026-05-18 16:38:21 +02:00
Brummel 9f3e79c874 feat: text normalization with medical abbreviation expansion 2026-05-18 16:36:06 +02:00
Brummel c8ea9a5fb3 fix: accumulate mixed-content ClaML labels (full titles) 
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-18 16:34:19 +02:00
Brummel bb42f07790 feat: streaming ClaML parser for ICD metadata and titles 
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-18 16:25:46 +02:00
Brummel 2fe9656c84 refactor: idiomatic find in primary_code (clippy) 2026-05-18 16:22:18 +02:00
Brummel 3e8284bfc0 feat: Alpha-ID-SE corpus parser with code normalization 2026-05-18 16:20:21 +02:00
Brummel 176aa87c0b refactor: derive Default for Filter (clippy) 2026-05-18 16:18:34 +02:00
Brummel 30799ddb0b feat: core domain types and Filter predicate 2026-05-18 16:16:08 +02:00
Brummel 70be6449db chore: scaffold alpha-id crate skeleton 2026-05-18 16:12:59 +02:00
Brummel bb5123aa4a chore: gitignore .claude/worktrees 
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-18 16:09:00 +02:00
Brummel 24a0db2359 Add implementation plan for alpha-id ICD suggestion PoC 
19 TDD tasks across Phase 0 (scaffold), Phase C (lexical baseline
+ data layer + eval, no IONOS), Phase A (IONOS bge-m3 hybrid +
Qwen3-VL reranker + cost-guarded resumable embedding + degrade).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-18 16:03:37 +02:00
Brummel b313c94ea3 Add design spec for alpha-id ICD code suggestion PoC 
Brainstormed design: free-text dictation (marker-segmented by
doctate) -> single prioritized top-10 ICD-10-GM list with tag
filtering. Two modes (C lexical baseline, A hybrid retrieval +
IONOS bge-m3 + Qwen3-VL-Reranker-8B), eval harness via synthetic
transcription errors, one-time cost-guarded embedding.

Licensed BfArM data dirs and local settings are gitignored.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-18 15:51:55 +02:00