MedITron/PTSRaiAgent

Fork 0

Files

T

History

ElebobaandClaude Opus 5.5 6a77831603

CI / ruff · pytest (pull_request) Successful in 1m30s

Details

feat(llm): RAG knowledge base over clinical guidelines with cited extractive answers

- docs/source/rag: PDF sources (KR PTSD 2023, KR depression 2024, meds module spec),
  pdf->txt converter (tables as markdown, page markers, hand-transcribed figures),
  word-level check against pdftotext; fixes HAM-D/MADRS fake headers and split rows
- config/rag: corpus txt + sources.yaml (documents, audiences, retrieval params, synonyms)
- app/llm/rag: Snowball stemmer, section-aware chunker with page citations,
  BM25 + synonym expansion + densest-passage reranking, extractive answer()
  with coverage gate, patient dose/prescribing filter, optional LLM rephrasing
  rejected when it adds content; CLI python -m app.llm.rag
- tests: gold retrieval set (hit@3 >= 15/18), off-topic, safety, faithfulness
- ADR-011

Refs #60

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-04 04:11:06 +03:00

instruments

feat(questionnaire): integrate revised intake and follow-up monitoring

2026-10-03 18:19:34 +03:00

lexicon

feat(core): safety detector with RU risk lexicon and fixed crisis script

2026-10-03 15:01:35 +03:00

privacy

feat(core): implement scrub() and pseudo_id() with tests

2026-10-03 15:00:21 +03:00

rag

feat(llm): RAG knowledge base over clinical guidelines with cited extractive answers

2026-10-04 04:11:06 +03:00

scales

feat(questionnaire): integrate revised intake and follow-up monitoring

2026-10-03 18:19:34 +03:00

texts

feat(llm): RAG knowledge base over clinical guidelines with cited extractive answers

2026-10-04 04:11:06 +03:00

intake_runtime.yaml

feat(questionnaire): integrate revised intake and follow-up monitoring

2026-10-03 18:19:34 +03:00

modules.yaml

chore(core): stage 1 contracts — dialog, API schemas + fixtures, integrations, monitoring, modules

2026-10-03 15:22:25 +03:00

questionnaire.yaml

feat(questionnaire): integrate revised intake and follow-up monitoring

2026-10-03 18:19:34 +03:00

README.md

feat(llm): RAG knowledge base over clinical guidelines with cited extractive answers

2026-10-04 04:11:06 +03:00

rules.yaml

feat(core): rule engine for alerts/agent actions and virtual demo clock

2026-10-03 14:58:19 +03:00

scheduler.yaml

feat(backend): demo scheduler on the virtual clock — doses, missed, check-ins, scales, silence

2026-10-04 03:25:44 +03:00

README.md

config/

Конфигурация клинической логики (ADR-008). Схемы задаёт зона core, значения утверждает психиатр (clinical).

scales/ — шкалы: пункты, варианты, градации, флаги
rules.yaml — правила алертов и действий
modules.yaml — модули и расписания шкал
lexicon/risk_ru.yaml — лексикон риска
texts/ — тексты для пациента (кризисный скрипт, согласие, перефразировки)
rag/ — база знаний (ADR-011): sources.yaml (документы, параметры поиска, синонимы) и corpus/*.txt, собранные из PDF в docs/source/rag/