pediatric-ai-scribe-v3/src/routes/edEncounters.js
Daniel b53aa34248 feat: ED multi-stage UX, extensions polish, docs viewer + application-logic docs
Three concurrent themes from this session:

═══════════════════════════════════════════════════════════════════
ED ENCOUNTERS — per-stage cards + consolidate→MDM finalize
═══════════════════════════════════════════════════════════════════

UX redesign per Daniel's feedback ("every stage note should be shown,
if AI is told to modify that particular note then the modified version
is used in final mdm"):

- Each generated stage stays on screen as its own editable card with
  its own embedded "Don't Miss" panel. No more single rolling note
  element that gets replaced on each generation.
- gatherCurrentNotes() reads contenteditable text from each stage card
  before any operation (advance, finalize, persist) so inline edits
  flow into the next AI call and the final consolidate.
- Stage badge is now state-accurate. "Stage N (recording)" with yellow
  background after Add-more before generation; "Stage N" with gray
  after generation. Fixes the bug where the badge flipped to Stage 2
  the moment Add-more was clicked.
- Save & Done now runs TWO server-side AI calls in /finalize:
  1. edConsolidate (new prompt) → polished single final note that
     integrates every stage chronologically (HPI / ROS / PE / ED Course /
     A&P with disposition).
  2. edFinalize (rewritten with full inline 2023 AMA E/M element
     rubric — problems / data / risk definitions, level mapping with
     concrete examples) → MDM JSON.
- Two new cards render after finalize: blue-bordered Final Consolidated
  Note + green-bordered MDM. Stage cards become read-only.
- partial_data on the saved row now stores {stages, finalNote, mdm,
  finalized} so resume re-renders the full state.

Why two-call finalize: a single combined prompt makes the model cut
corners on one task. Two focused calls cost ~2× latency at the very end
of an encounter — acceptable since finalize is a one-time terminal
action, not a per-stage hot path.

Files: public/components/ed-encounter.html, public/js/ed-encounters.js,
src/routes/edEncounters.js, src/utils/prompts.js (edConsolidate added,
edFinalize rewritten).

═══════════════════════════════════════════════════════════════════
EXTENSIONS / PAGERS — visual polish
═══════════════════════════════════════════════════════════════════

Multiple iterations based on Daniel's feedback:

- Layout: align-items:flex-start so action buttons stay pinned top-right
  when long numbers wrap (was align-items:center → buttons drifted into
  the text area, causing visible overlap).
- Number: word-break:break-all + min-width:0 + font-feature-settings:tnum
  so long numbers wrap within their column instead of pushing under the
  buttons. Click-to-copy with a 0.55s green flash + ✓ copied badge.
- Phone/pager Font Awesome icon next to the number in the type color —
  at-a-glance type signal (replacing an earlier 3px left stripe that
  Daniel found visually bulky).
- Name: font-weight 700, font-size 14.5px, color g900, letter-spacing
  -0.012em — scan-target headline typography for long lists.
- Alternating subtle backgrounds by index (white vs #fafbfc) so a long
  list reads as distinct rows.
- Hover: card lifts 1px with a soft shadow; action buttons fade from
  55% to 100% opacity. Cubic-bezier transition on transform.
- Entrance: staggered fade-up animation per card (35ms × index, capped
  at 12). prefers-reduced-motion media query disables motion.
- Empty state: 48px FA icon + heading instead of plain gray text.

Files: public/js/extensions.js, public/css/styles.css.

═══════════════════════════════════════════════════════════════════
DOCS REORGANIZATION + APPLICATION-LOGIC DOCS + ADMIN VIEWER
═══════════════════════════════════════════════════════════════════

Document moves (preserving git history via git mv):
  BROWSER_WHISPER_SETUP.md          → docs/browser-whisper-setup.md
  BROWSER_WHISPER_TROUBLESHOOTING.md → docs/browser-whisper-troubleshooting.md
  DEVELOPER_GUIDE.md                → docs/developer-guide-extended.md
  EMBEDDINGS_SETUP.md               → docs/embeddings-setup.md
  FEATURES_EXPLAINED.md             → docs/features-explained.md
  IMPROVEMENTS.md                   → docs/improvements.md
  OPENID_SETUP.md                   → docs/openid-setup.md
  TRANSCRIPTION_OPTIONS.md          → docs/transcription-options.md
README.md updated with the new paths + a Documentation section that
links to docs/logic/ at the top.

New application-logic doc series (~8,300 lines total) at docs/logic/.
Built with 5 parallel doc-writing agents per Daniel's "use multiple
agents" directive. Each doc explains how a part of the app actually
works — application logic, data flow, design decisions, sacred zones,
how-to-extend recipes — at a depth that lets a new dev (or an AI
assistant) modify the code confidently.

  docs/logic/README.md                — index + recommended reading order
  docs/logic/architecture.md (2166 L) — frontend IIFE pattern, lazy tab
                                         load, backend route convention,
                                         schema, encryption, deployment
  docs/logic/clinical-notes.md (1546L) — every note tab + helper trio
  docs/logic/bedside-and-calculators.md (1373L) — bedside ES module
                                         pocket + calculators + PE Guide
                                         + suture selector
  docs/logic/auth-admin-learning.md (1281L) — auth (local+OIDC+2FA) +
                                         admin panel + Learning Hub
                                         (Quiz engine logic at sub-detail
                                         only — TODO follow-up)
  docs/logic/ai-and-voice.md (1128 L) — callAI 5-provider routing,
                                         prompts, voice/STT, helper trio
  docs/logic/ed-encounters.md (821 L) — multi-stage ED + MDM (this
                                         session's worked example)

Admin-only docs viewer:
- New route /api/admin/docs/{tree,file}: recursively walks docs/, returns
  the tree as JSON; /file?path=X validates path stays inside docs/ and
  renders markdown via marked. Both gated by req.user.role==='admin'.
- New tab "Docs" (book icon) in the sidebar, hidden by default and
  revealed in auth.js when user.role==='admin' (same pattern as the
  existing Admin and CMS tabs).
- New component public/components/admin-docs.html: split-pane layout
  with a tree sidebar + filter input + a markdown reader pane.
- New module public/js/admin-docs.js: lazy-loads the tree on first tab
  activation, renders collapsible folders, persists expanded state and
  last-opened path via UIState. Server-rendered HTML so no client
  markdown parser needed.
- CSS for the viewer (responsive split-pane, code-block styling, table
  scrolling, etc.).
- Mounted at /api/admin/docs (NOT /api) — important: mounting a router
  with router.use(authMiddleware) at /api accidentally 401s every other
  /api/* path (caught and fixed during testing — /api/health was 401'ing).

Files: docs/* (moved + new), README.md, public/components/admin-docs.html
(new), public/js/admin-docs.js (new), src/routes/adminDocs.js (new),
public/index.html (tab + section + script), public/js/auth.js (admin
gate + logout cleanup), public/css/styles.css (viewer styles), server.js
(mount).

═══════════════════════════════════════════════════════════════════
KNOWN GAPS (TODO follow-ups)
═══════════════════════════════════════════════════════════════════

- Learning Hub quiz engine (MCQ / multi-select / T-F scoring + attempt
  tracking + progress dashboard) is covered at the architectural level
  in docs/logic/auth-admin-learning.md but not drilled into the quiz
  data model and scoring flow. Worth a focused follow-up doc.
- ED finalize: if MDM step JSON parse fails, server returns 502 with
  the consolidated finalNote in the error payload, but client doesn't
  surface the partial result. Add a "MDM failed, retry" affordance.
- No e2e Playwright coverage for ED encounters or the new docs viewer.
2026-04-28 03:09:38 +02:00

212 lines
9.8 KiB
JavaScript

// ============================================================
// ED ENCOUNTER ROUTE — multi-stage emergency department note
// generation with don't-miss tooltips and 2023 E/M MDM finalize.
//
// Two endpoints, both auth-gated:
// POST /api/ed-encounters/generate — per-stage note + tooltips
// POST /api/ed-encounters/finalize — MDM block for billing
//
// Stage state lives client-side (localStorage + draft row in
// saved_encounters via the existing encounters route). This file
// is purely AI plumbing — it does not own persistence.
// ============================================================
var express = require('express');
var router = express.Router();
var { callAI } = require('../utils/ai');
var PROMPTS = require('../utils/prompts');
var { authMiddleware } = require('../middleware/auth');
var logger = require('../utils/logger');
var { wrapUserText, INJECTION_GUARD } = require('../utils/promptSafe');
router.use(authMiddleware);
// Models sometimes wrap JSON in ```json fences or prepend prose. Strip the
// fence and recover the JSON object between the first { and last }.
function extractJson(raw) {
var t = String(raw || '').trim();
t = t.replace(/^```(?:json)?\s*/i, '').replace(/\s*```$/i, '');
var s = t.indexOf('{');
if (s > 0) t = t.substring(s);
var e = t.lastIndexOf('}');
if (e > -1 && e < t.length - 1) t = t.substring(0, e + 1);
try { return JSON.parse(t); } catch (err) { return null; }
}
// ── POST /ed-encounters/generate — one stage of an ED encounter ─────
// Body:
// stage (1, 2, ...) — informational, drives "previous note" handling
// transcript — current stage's voice transcript (required)
// chiefComplaint — string, required
// patientAge — string, optional but strongly preferred
// patientGender — string, optional
// previousNote — string, only on stage > 1, the last generated note
// physicianMemories — string from /api/memories/context, optional
// model — admin/user-selected model id, optional
// Returns: { success, note, dontMiss: [{point, why}], model }
router.post('/ed-encounters/generate', async function (req, res) {
var start = Date.now();
try {
var {
stage, transcript, chiefComplaint,
patientAge, patientGender,
previousNote, physicianMemories, model
} = req.body;
if (!chiefComplaint || !chiefComplaint.trim()) {
return res.status(400).json({ error: 'Chief complaint is required' });
}
if (!transcript || !transcript.trim()) {
return res.status(400).json({ error: 'Transcript is required' });
}
var stageNum = parseInt(stage, 10) || 1;
var context = 'ED ENCOUNTER — STAGE ' + stageNum + '\n';
context += 'Patient: ' + (patientAge || 'Unknown age') + ', ' + (patientGender || 'Unknown gender') + '\n';
context += 'Chief Complaint: ' + wrapUserText('chief_complaint', chiefComplaint) + '\n\n';
context += 'CURRENT STAGE TRANSCRIPT (may include direct physician asides — preserve and route them per the prompt rules):\n'
+ wrapUserText('transcript', transcript) + '\n\n';
if (previousNote && previousNote.trim() && stageNum > 1) {
context += 'PREVIOUS-STAGE NOTE (baseline to integrate on top of — do not start fresh):\n'
+ wrapUserText('previous_note', previousNote) + '\n\n';
}
if (physicianMemories && physicianMemories.trim()) {
context += physicianMemories + '\n\n';
}
var result = await callAI([
{ role: 'system', content: PROMPTS.edEncounterStaged + INJECTION_GUARD },
{ role: 'user', content: context }
], { model: model, maxTokens: 4000 });
var parsed = extractJson(result.content);
if (!parsed || typeof parsed.note !== 'string') {
// Recovery: if the model returned plain prose, treat the whole response
// as the note and return an empty don't-miss list rather than 500-ing.
parsed = { note: String(result.content || '').trim(), dontMiss: [] };
}
if (!Array.isArray(parsed.dontMiss)) parsed.dontMiss = [];
parsed.dontMiss = parsed.dontMiss
.filter(function (d) { return d && d.point; })
.map(function (d) { return { point: String(d.point).trim(), why: String(d.why || '').trim() }; });
var dur = Date.now() - start;
logger.apiCall(req.user.id, '/api/ed-encounters/generate', {
model: result.model,
tokensInput: result.usage && result.usage.input_tokens,
tokensOutput: result.usage && result.usage.output_tokens,
duration: dur,
statusCode: 200
});
logger.audit(req.user.id, 'generate_ed_encounter', 'ED encounter stage ' + stageNum, req, { category: 'clinical' });
res.json({ success: true, note: parsed.note, dontMiss: parsed.dontMiss, model: result.model });
} catch (e) {
logger.error('[edEncounters] generate failed', e.message);
res.status(500).json({ error: 'Request failed' });
}
});
// ── POST /ed-encounters/finalize — consolidate stages + generate MDM ─
// Two-step server-side flow:
// 1. Consolidate every stage's note + transcript into one polished
// final clinical note (PROMPTS.edConsolidate, plain text out).
// 2. Generate the 2023 AMA E/M MDM block from that consolidated note
// (PROMPTS.edFinalize, JSON out).
// Two AI calls so each task gets its own focused prompt — more reliable
// than asking the model for both in one response.
// Body:
// stages — array of { transcript, note } in chronological order (required)
// chiefComplaint — string, optional but strongly preferred
// patientAge — string, optional
// patientGender — string, optional
// model — optional override
// Returns: { success, finalNote, mdm, model }
router.post('/ed-encounters/finalize', async function (req, res) {
var start = Date.now();
try {
var { stages, chiefComplaint, patientAge, patientGender, model } = req.body;
if (!Array.isArray(stages) || stages.length === 0) {
return res.status(400).json({ error: 'stages array is required' });
}
var validStages = stages.filter(function (s) { return s && (s.note || s.transcript); });
if (validStages.length === 0) {
return res.status(400).json({ error: 'at least one stage must have a note or transcript' });
}
// ── Step 1: consolidate ──────────────────────────────────────────
var consolidateContext = 'PATIENT: ' + (patientAge || 'Unknown age')
+ (patientGender ? ', ' + patientGender : '') + '\n';
if (chiefComplaint && chiefComplaint.trim()) {
consolidateContext += 'CHIEF COMPLAINT: ' + wrapUserText('chief_complaint', chiefComplaint) + '\n';
}
consolidateContext += '\nSTAGE-BY-STAGE NOTES AND TRANSCRIPTS (chronological):\n\n';
validStages.forEach(function (s, i) {
consolidateContext += '=== STAGE ' + (i + 1) + ' ===\n';
if (s.transcript && s.transcript.trim()) {
consolidateContext += 'Transcript:\n' + wrapUserText('transcript_' + i, s.transcript) + '\n';
}
if (s.note && s.note.trim()) {
consolidateContext += 'Working note (may include physician edits):\n' + wrapUserText('note_' + i, s.note) + '\n';
}
consolidateContext += '\n';
});
var consolidateResult = await callAI([
{ role: 'system', content: PROMPTS.edConsolidate + INJECTION_GUARD },
{ role: 'user', content: consolidateContext }
], { model: model, maxTokens: 4000 });
var finalNote = String(consolidateResult.content || '').trim();
if (!finalNote) {
return res.status(502).json({ error: 'Consolidate step returned empty note' });
}
// ── Step 2: MDM ──────────────────────────────────────────────────
var fullTranscript = validStages.map(function (s, i) {
return '--- Stage ' + (i + 1) + ' ---\n' + (s.transcript || '');
}).join('\n\n');
var mdmContext = 'PATIENT: ' + (patientAge || 'Unknown age') + '\n\n';
mdmContext += 'FINAL CLINICAL NOTE:\n' + wrapUserText('note', finalNote) + '\n\n';
mdmContext += 'FULL ENCOUNTER TRANSCRIPT (all stages):\n' + wrapUserText('transcript', fullTranscript) + '\n';
var mdmResult = await callAI([
{ role: 'system', content: PROMPTS.edFinalize + INJECTION_GUARD },
{ role: 'user', content: mdmContext }
], { model: model, maxTokens: 2000 });
var parsed = extractJson(mdmResult.content);
if (!parsed || !parsed.mdm) {
return res.status(502).json({
error: 'MDM step did not return a parseable JSON block',
finalNote: finalNote, // still return the consolidated note so the client doesn't lose work
raw: mdmResult.content
});
}
var dur = Date.now() - start;
var inTokens = ((consolidateResult.usage && consolidateResult.usage.input_tokens) || 0)
+ ((mdmResult.usage && mdmResult.usage.input_tokens) || 0);
var outTokens = ((consolidateResult.usage && consolidateResult.usage.output_tokens) || 0)
+ ((mdmResult.usage && mdmResult.usage.output_tokens) || 0);
logger.apiCall(req.user.id, '/api/ed-encounters/finalize', {
model: mdmResult.model,
tokensInput: inTokens,
tokensOutput: outTokens,
duration: dur,
statusCode: 200
});
logger.audit(req.user.id, 'finalize_ed_encounter', 'ED encounter consolidated + MDM (' + validStages.length + ' stages)', req, { category: 'clinical' });
res.json({ success: true, finalNote: finalNote, mdm: parsed.mdm, model: mdmResult.model });
} catch (e) {
logger.error('[edEncounters] finalize failed', e.message);
res.status(500).json({ error: 'Request failed' });
}
});
module.exports = router;