feat: traçabilité source systématique + viewer interactif

Ajoute source_page/source_excerpt à tous les types (biologie, imagerie,
traitements, actes CCAM, antécédents, complications). Convertit antecedents
et complications en types structurés (Antecedent/Complication) avec
validators backward-compat pour les vieux JSON. Étend _apply_source_tracking
à tous les éléments du dossier. Ajoute un endpoint /api/source-text/ et un
modal interactif dans le viewer avec surlignage du texte source.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

src/config.py

@@ -7,7 +7,7 @@ from pathlib import Path
 from typing import Optional
 
 from dotenv import load_dotenv
-from pydantic import BaseModel, Field
+from pydantic import BaseModel, Field, field_validator
 
 load_dotenv()
 
@@ -125,24 +125,44 @@ class ActeCCAM(BaseModel):
     date: Optional[str] = None
     validite: Optional[str] = None  # "valide" | "obsolete" | "non_verifie"
     alertes: list[str] = Field(default_factory=list)
+    source_page: Optional[int] = None
+    source_excerpt: Optional[str] = None
 
 
 class Traitement(BaseModel):
     medicament: str
     posologie: Optional[str] = None
     code_atc: Optional[str] = None
+    source_page: Optional[int] = None
+    source_excerpt: Optional[str] = None
 
 
 class BiologieCle(BaseModel):
     test: str
     valeur: Optional[str] = None
     anomalie: Optional[bool] = None
+    source_page: Optional[int] = None
+    source_excerpt: Optional[str] = None
 
 
 class Imagerie(BaseModel):
     type: str
     conclusion: Optional[str] = None
     score: Optional[str] = None
+    source_page: Optional[int] = None
+    source_excerpt: Optional[str] = None
+
+
+class Antecedent(BaseModel):
+    texte: str
+    source_page: Optional[int] = None
+    source_excerpt: Optional[str] = None
+
+
+class Complication(BaseModel):
+    texte: str
+    source_page: Optional[int] = None
+    source_excerpt: Optional[str] = None
 
 
 class DossierMedical(BaseModel):
@@ -152,17 +172,45 @@ class DossierMedical(BaseModel):
     diagnostic_principal: Optional[Diagnostic] = None
     diagnostics_associes: list[Diagnostic] = Field(default_factory=list)
     actes_ccam: list[ActeCCAM] = Field(default_factory=list)
-    antecedents: list[str] = Field(default_factory=list)
+    antecedents: list[Antecedent] = Field(default_factory=list)
     traitements_sortie: list[Traitement] = Field(default_factory=list)
     biologie_cle: list[BiologieCle] = Field(default_factory=list)
     imagerie: list[Imagerie] = Field(default_factory=list)
-    complications: list[str] = Field(default_factory=list)
+    complications: list[Complication] = Field(default_factory=list)
     alertes_codage: list[str] = Field(default_factory=list)
     source_files: list[str] = Field(default_factory=list)
     ghm_estimation: Optional[GHMEstimation] = None
     controles_cpam: list[ControleCPAM] = Field(default_factory=list)
     processing_time_s: float | None = None
 
+    @field_validator("antecedents", mode="before")
+    @classmethod
+    def _coerce_antecedents(cls, v):
+        """Backward compat : convertit les anciennes list[str] en list[Antecedent]."""
+        if not isinstance(v, list):
+            return v
+        result = []
+        for item in v:
+            if isinstance(item, str):
+                result.append({"texte": item})
+            else:
+                result.append(item)
+        return result
+
+    @field_validator("complications", mode="before")
+    @classmethod
+    def _coerce_complications(cls, v):
+        """Backward compat : convertit les anciennes list[str] en list[Complication]."""
+        if not isinstance(v, list):
+            return v
+        result = []
+        for item in v:
+            if isinstance(item, str):
+                result.append({"texte": item})
+            else:
+                result.append(item)
+        return result
+
 
 # --- Rapport d'anonymisation ---
 

src/control/cpam_response.py

@@ -470,10 +470,10 @@ def _build_cpam_prompt(
         dossier_lines.append(f"- Traitements de sortie : {', '.join(trt_parts)}")
 
     if dossier.antecedents:
-        dossier_lines.append(f"- Antécédents : {', '.join(dossier.antecedents[:10])}")
+        dossier_lines.append(f"- Antécédents : {', '.join(a.texte for a in dossier.antecedents[:10])}")
 
     if dossier.complications:
-        dossier_lines.append(f"- Complications : {', '.join(dossier.complications)}")
+        dossier_lines.append(f"- Complications : {', '.join(c.texte for c in dossier.complications)}")
 
     dossier_str = "\n".join(dossier_lines) if dossier_lines else "Non disponible"
 

src/medical/cim10_extractor.py

@@ -14,7 +14,9 @@ from .ccam_dict import lookup as ccam_lookup, validate_code as ccam_validate
 from .das_filter import clean_diagnostic_text, is_valid_diagnostic_text, correct_known_miscodes
 from ..config import (
     ActeCCAM,
+    Antecedent,
     BiologieCle,
+    Complication,
     Diagnostic,
     DossierMedical,
     Imagerie,
@@ -180,10 +182,10 @@ def _extract_das_llm(text: str, dossier: DossierMedical) -> None:
             "age": dossier.sejour.age,
             "duree_sejour": dossier.sejour.duree_sejour,
             "imc": dossier.sejour.imc,
-            "antecedents": dossier.antecedents[:5],
+            "antecedents": [a.texte for a in dossier.antecedents[:5]],
             "biologie_cle": [(b.test, b.valeur, b.anomalie) for b in dossier.biologie_cle],
             "imagerie": [(i.type, (i.conclusion or "")[:200]) for i in dossier.imagerie],
-            "complications": dossier.complications,
+            "complications": [c.texte for c in dossier.complications],
         }
 
         # DAS existants (texte + code)
@@ -532,7 +534,8 @@ _ANTECEDENT_NOISE = (
     "item de", "surveillance", "température", "signes vitaux",
     "pouls", "type de note", "aucune donnée", "renseignée",
     "habitudes de vie", "systolique", "diastolique", "saturation",
-    "texte libre", "mode de vie", "n° rpps",
+    "texte libre", "mode de vie", "n° rpps", "secrétariat",
+    "aucune aide",
 )
 
 _SURVEILLANCE_SINGLE_WORDS = frozenset({
@@ -569,8 +572,14 @@ def _is_valid_antecedent(line: str) -> bool:
     # Deux mots identiques
     if len(words) == 2 and len(set(words)) == 1:
         return False
-    # Identifiants administratifs isolés
+    # Lignes commençant par un tag médecin (artefact colonne gauche CRH)
-    if re.match(r'^\[MEDECIN\]\s', line) and len(line) < 30:
+    if re.match(r'^\[MEDECIN', line):
+        return False
+    # Lignes commençant par "Dr [MEDECIN" ou "Dr[PERSONNE" (nom de médecin)
+    if re.match(r'^Dr\s*\[', line):
+        return False
+    # Fragment de localisation : "de Bordeaux", "de Lyon", "de Paris"
+    if re.match(r'^de [A-ZÀ-Ú]', line) and len(line) < 25:
         return False
     return True
 
@@ -578,7 +587,7 @@ def _is_valid_antecedent(line: str) -> bool:
 def _extract_antecedents(text: str, dossier: DossierMedical) -> None:
     """Extrait les antécédents."""
     m = re.search(
-        r"Antécédents?\s*[:：]?\s*\n?(.*?)(?=\n\s*(?:Traitements?\s*[:：]|Allergie|Histoire de la maladie|Examen clinique|Signes\s+[Vv]itaux|Observations?\s+m[eé]dicale|Passage aux|\n\n))",
+        r"Antécédents?\s*[:：]?\s*\n?(.*?)(?=\n\s*(?:Traitements?\s*[:：]|Allergie|Histoire de la maladie|Examen clinique|Signes\s+[Vv]itaux|Observations?\s+m[eé]dicale|Passage aux|Mode de vie|\n\n))",
         text,
         re.DOTALL | re.IGNORECASE,
     )
@@ -587,7 +596,7 @@ def _extract_antecedents(text: str, dossier: DossierMedical) -> None:
         for line in block.split("\n"):
             line = line.strip().lstrip("- •")
             if _is_valid_antecedent(line):
-                dossier.antecedents.append(line)
+                dossier.antecedents.append(Antecedent(texte=line))
 
 
 def _extract_traitements(
@@ -778,7 +787,7 @@ def _extract_complications(
             # Fallback regex pour la négation
             pattern = rf"(?:pas de|sans|absence de|aucun[e]?)\s+{re.escape(term)}"
             if not re.search(pattern, text_lower):
-                dossier.complications.append(term.capitalize())
+                dossier.complications.append(Complication(texte=term.capitalize()))
 
 
 def _is_negated_by_edsnlp(term: str, negated_terms: set[str]) -> bool:
@@ -1028,34 +1037,84 @@ def _is_abnormal(test: str, value: str) -> bool | None:
     return None
 
 
-def _apply_source_tracking(dossier: DossierMedical, page_tracker, search_text: str) -> None:
+def _track_item(item, search_key: str, page_tracker, search_text: str) -> bool:
-    """Ajoute la traçabilité source (page + extrait) à chaque diagnostic.
+    """Cherche la page source et l'extrait pour un item avec source_page/source_excerpt."""
+    if item.source_page is not None:
+        return False
+    if not search_key:
+        return False
+    page = page_tracker.find_page_for_text(search_key, search_text)
+    if page:
+        item.source_page = page
+        item.source_excerpt = page_tracker.extract_excerpt(search_key, search_text)
+        return True
+    return False
 
-    Cherche le texte du diagnostic dans le texte source pour retrouver
+
+def _apply_source_tracking(dossier: DossierMedical, page_tracker, search_text: str) -> None:
+    """Ajoute la traçabilité source (page + extrait) à tous les éléments du dossier.
+
+    Cherche le texte de chaque élément dans le texte source pour retrouver
     la page d'origine et extraire un passage contextualisé.
     """
+    tracked = 0
+    total = 0
+
+    # Diagnostics (DP + DAS)
     all_diags: list[Diagnostic] = []
     if dossier.diagnostic_principal:
         all_diags.append(dossier.diagnostic_principal)
     all_diags.extend(dossier.diagnostics_associes)
 
-    tracked = 0
     for diag in all_diags:
-        if diag.source_page is not None:
+        total += 1
-            continue  # déjà renseigné
+        if _track_item(diag, diag.texte, page_tracker, search_text):
+            tracked += 1
 
-        texte = diag.texte
+    # Biologie
-        if not texte:
+    for b in dossier.biologie_cle:
-            continue
+        total += 1
+        search_key = f"{b.test}: {b.valeur}" if b.valeur else b.test
+        if _track_item(b, search_key, page_tracker, search_text):
+            tracked += 1
+        elif b.valeur and _track_item(b, b.test, page_tracker, search_text):
+            tracked += 1
 
-        page = page_tracker.find_page_for_text(texte, search_text)
+    # Imagerie
-        if page:
+    for img in dossier.imagerie:
-            diag.source_page = page
+        total += 1
-            diag.source_excerpt = page_tracker.extract_excerpt(texte, search_text)
+        search_key = img.type
+        if _track_item(img, search_key, page_tracker, search_text):
+            tracked += 1
+        elif img.conclusion and _track_item(img, img.conclusion[:50], page_tracker, search_text):
+            tracked += 1
+
+    # Traitements
+    for t in dossier.traitements_sortie:
+        total += 1
+        if _track_item(t, t.medicament, page_tracker, search_text):
+            tracked += 1
+
+    # Actes CCAM
+    for a in dossier.actes_ccam:
+        total += 1
+        if _track_item(a, a.texte, page_tracker, search_text):
+            tracked += 1
+
+    # Antécédents
+    for ant in dossier.antecedents:
+        total += 1
+        if _track_item(ant, ant.texte, page_tracker, search_text):
+            tracked += 1
+
+    # Complications
+    for comp in dossier.complications:
+        total += 1
+        if _track_item(comp, comp.texte, page_tracker, search_text):
             tracked += 1
 
     if tracked:
-        logger.info("  Traçabilité source : %d/%d diagnostics localisés", tracked, len(all_diags))
+        logger.info("  Traçabilité source : %d/%d éléments localisés", tracked, total)
 
 
 def _validate_justifications(dossier: DossierMedical) -> None:

src/medical/clinical_context.py

@@ -166,10 +166,10 @@ def build_enriched_context(dossier: DossierMedical) -> dict:
         "age": dossier.sejour.age,
         "duree_sejour": dossier.sejour.duree_sejour,
         "imc": dossier.sejour.imc,
-        "antecedents": dossier.antecedents[:5],
+        "antecedents": [a.texte for a in dossier.antecedents[:5]],
         "biologie_cle": [(b.test, b.valeur, b.anomalie) for b in dossier.biologie_cle],
         "imagerie": [(i.type, (i.conclusion or "")[:200]) for i in dossier.imagerie],
-        "complications": dossier.complications,
+        "complications": [c.texte for c in dossier.complications],
     }
 
     # Interprétations biologiques

src/medical/fusion.py

@@ -10,7 +10,9 @@ import logging
 
 from ..config import (
     ActeCCAM,
+    Antecedent,
     BiologieCle,
+    Complication,
     Diagnostic,
     DossierMedical,
     Imagerie,
@@ -251,7 +253,7 @@ def merge_dossiers(dossiers: list[DossierMedical]) -> DossierMedical:
     ant_seen: set[str] = set()
     for d in dossiers:
         for a in d.antecedents:
-            key = a.lower().strip()
+            key = a.texte.lower().strip()
             if key not in ant_seen:
                 merged.antecedents.append(a)
                 ant_seen.add(key)
@@ -260,7 +262,7 @@ def merge_dossiers(dossiers: list[DossierMedical]) -> DossierMedical:
     comp_seen: set[str] = set()
     for d in dossiers:
         for c in d.complications:
-            key = c.lower().strip()
+            key = c.texte.lower().strip()
             if key not in comp_seen:
                 merged.complications.append(c)
                 comp_seen.add(key)

src/viewer/app.py

@@ -16,7 +16,7 @@ from werkzeug.utils import secure_filename
 from collections import Counter
 
 from ..config import (
-    STRUCTURED_DIR, OLLAMA_URL, CCAM_DICT_PATH, DossierMedical,
+    ANONYMIZED_DIR, STRUCTURED_DIR, OLLAMA_URL, CCAM_DICT_PATH, DossierMedical,
     ALLOWED_EXTENSIONS, UPLOAD_MAX_SIZE_MB,
     CIM10_PDF, GUIDE_METHODO_PDF, CCAM_PDF, CIM10_DICT_PATH, CIM10_SUPPLEMENTS_PATH,
 )
@@ -594,6 +594,27 @@ def create_app() -> Flask:
             logger.exception("Erreur lors du retraitement")
             return jsonify({"error": str(e)}), 500
 
+    # ------------------------------------------------------------------
+    # API texte source anonymisé
+    # ------------------------------------------------------------------
+
+    @app.route("/api/source-text/<path:dossier_id>")
+    def source_text(dossier_id: str):
+        """Retourne le contenu texte anonymisé de tous les fichiers d'un dossier."""
+        safe_dir = (ANONYMIZED_DIR / dossier_id).resolve()
+        if not safe_dir.is_relative_to(ANONYMIZED_DIR.resolve()):
+            abort(403)
+        if not safe_dir.is_dir():
+            abort(404)
+
+        result = {}
+        for txt_path in sorted(safe_dir.glob("*_anonymized.txt")):
+            try:
+                result[txt_path.name] = txt_path.read_text(encoding="utf-8")
+            except Exception:
+                logger.warning("Impossible de lire %s", txt_path)
+        return jsonify(result)
+
     # ------------------------------------------------------------------
     # Routes admin référentiels
     # ------------------------------------------------------------------

src/viewer/templates/base.html

@@ -233,6 +233,79 @@
     border-radius: 50%;
     animation: spin 0.8s linear infinite;
   }
+
+  /* Source tracking badges */
+  .src-btn {
+    display: inline-block;
+    padding: 1px 6px;
+    border-radius: 9999px;
+    font-size: 0.65rem;
+    font-weight: 600;
+    background: #e0f2fe;
+    color: #0369a1;
+    border: 1px solid #bae6fd;
+    cursor: pointer;
+    margin-left: 0.3rem;
+    vertical-align: middle;
+    transition: background 0.15s;
+  }
+  .src-btn:hover { background: #bae6fd; }
+
+  /* Source modal */
+  #source-modal {
+    display: none;
+    position: fixed;
+    inset: 0;
+    z-index: 9999;
+    background: rgba(0,0,0,0.5);
+    padding: 2rem;
+  }
+  #source-modal-inner {
+    background: #fff;
+    border-radius: 12px;
+    max-width: 900px;
+    margin: 0 auto;
+    max-height: 90vh;
+    display: flex;
+    flex-direction: column;
+    box-shadow: 0 8px 30px rgba(0,0,0,0.2);
+  }
+  #source-header {
+    padding: 1rem 1.25rem;
+    border-bottom: 1px solid #e2e8f0;
+    font-weight: 700;
+    font-size: 0.9rem;
+    color: #0f172a;
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+  }
+  #source-content {
+    flex: 1;
+    overflow-y: auto;
+    padding: 1.25rem;
+    font-size: 0.85rem;
+    line-height: 1.6;
+    white-space: pre-wrap;
+    word-break: break-word;
+    color: #334155;
+  }
+  #source-content mark {
+    background: #fef08a;
+    padding: 2px 0;
+    border-radius: 2px;
+  }
+  #source-close-btn {
+    padding: 0.4rem 1rem;
+    background: #64748b;
+    color: #fff;
+    border: none;
+    border-radius: 6px;
+    cursor: pointer;
+    font-size: 0.8rem;
+    font-weight: 600;
+  }
+  #source-close-btn:hover { background: #475569; }
 </style>
 </head>
 <body>

src/viewer/templates/detail.html

@@ -287,7 +287,10 @@
 {% set dp = dossier.diagnostic_principal %}
 <div class="card section">
   <h3>Diagnostic principal</h3>
-  <div style="font-size:0.95rem;margin-bottom:0.5rem;">{{ dp.texte }}</div>
+  <div style="font-size:0.95rem;margin-bottom:0.5rem;">
+    {{ dp.texte }}
+    {% if dp.source_page %}<button class="src-btn" onclick="showSource('{{ dp.source_excerpt|default('',true)|e }}', {{ dp.source_page }})">p.{{ dp.source_page }}</button>{% endif %}
+  </div>
   {% if dp.cim10_suggestion %}
     <span class="badge" style="background:#dbeafe;color:#1d4ed8;font-size:0.85rem;">{{ dp.cim10_suggestion }}</span>
     {{ dp.cim10_confidence | confidence_badge }}
@@ -355,12 +358,7 @@
             <span class="badge" style="background:#e0e7ff;color:#3730a3;font-size:0.7rem;">{{ das.source }}</span>
           {% endif %}
           {% if das.source_page %}
-            <span style="font-size:0.7rem;color:#64748b;">p.{{ das.source_page }}</span>
+            <button class="src-btn" onclick="showSource('{{ das.source_excerpt|default('',true)|e }}', {{ das.source_page }})">p.{{ das.source_page }}</button>
-          {% endif %}
-          {% if das.source_excerpt %}
-            <details style="margin-top:0.2rem;"><summary style="font-size:0.7rem;color:#94a3b8;cursor:pointer;">extrait</summary>
-              <pre style="font-size:0.7rem;white-space:pre-wrap;max-width:300px;color:#475569;">{{ das.source_excerpt }}</pre>
-            </details>
           {% endif %}
         </td>
         <td style="font-size:0.8rem;color:#475569;">
@@ -410,7 +408,7 @@
 <div class="card section">
   <h3>Actes CCAM ({{ dossier.actes_ccam|length }})</h3>
   <table>
-    <thead><tr><th>Texte</th><th>Code CCAM</th><th>Regroupement</th><th>Date</th><th>Validité</th></tr></thead>
+    <thead><tr><th>Texte</th><th>Code CCAM</th><th>Regroupement</th><th>Date</th><th>Validité</th><th>Source</th></tr></thead>
     <tbody>
     {% for a in dossier.actes_ccam %}
       <tr>
@@ -432,6 +430,7 @@
             <div style="font-size:0.7rem;color:#dc2626;">{{ alerte }}</div>
           {% endfor %}
         </td>
+        <td>{% if a.source_page %}<button class="src-btn" onclick="showSource('{{ a.source_excerpt|default('',true)|e }}', {{ a.source_page }})">p.{{ a.source_page }}</button>{% endif %}</td>
       </tr>
     {% endfor %}
     </tbody>
@@ -444,13 +443,14 @@
 <div class="card section">
   <h3>Biologie clé ({{ dossier.biologie_cle|length }})</h3>
   <table>
-    <thead><tr><th>Test</th><th>Valeur</th><th>Anomalie</th></tr></thead>
+    <thead><tr><th>Test</th><th>Valeur</th><th>Anomalie</th><th>Source</th></tr></thead>
     <tbody>
     {% for b in dossier.biologie_cle %}
       <tr{% if b.anomalie %} class="anomalie"{% endif %}>
         <td>{{ b.test }}</td>
         <td>{{ b.valeur or '' }}</td>
         <td>{% if b.anomalie %}<span class="badge" style="background:#fee2e2;color:#dc2626;">Oui</span>{% else %}—{% endif %}</td>
+        <td>{% if b.source_page %}<button class="src-btn" onclick="showSource('{{ b.source_excerpt|default('',true)|e }}', {{ b.source_page }})">p.{{ b.source_page }}</button>{% endif %}</td>
       </tr>
     {% endfor %}
     </tbody>
@@ -466,6 +466,7 @@
   <div style="margin-bottom:0.5rem;">
     <strong>{{ img.type }}</strong>
     {% if img.score %} — Score : {{ img.score }}{% endif %}
+    {% if img.source_page %}<button class="src-btn" onclick="showSource('{{ img.source_excerpt|default('',true)|e }}', {{ img.source_page }})">p.{{ img.source_page }}</button>{% endif %}
     {% if img.conclusion %}
       <div style="font-size:0.85rem;color:#475569;">{{ img.conclusion }}</div>
     {% endif %}
@@ -479,13 +480,14 @@
 <div class="card section">
   <h3>Traitements de sortie ({{ dossier.traitements_sortie|length }})</h3>
   <table>
-    <thead><tr><th>Médicament</th><th>Posologie</th><th>Code ATC</th></tr></thead>
+    <thead><tr><th>Médicament</th><th>Posologie</th><th>Code ATC</th><th>Source</th></tr></thead>
     <tbody>
     {% for t in dossier.traitements_sortie %}
       <tr>
         <td>{{ t.medicament }}</td>
         <td>{{ t.posologie or '' }}</td>
         <td>{% if t.code_atc %}<span class="badge" style="background:#e0e7ff;color:#3730a3;">{{ t.code_atc }}</span>{% endif %}</td>
+        <td>{% if t.source_page %}<button class="src-btn" onclick="showSource('{{ t.source_excerpt|default('',true)|e }}', {{ t.source_page }})">p.{{ t.source_page }}</button>{% endif %}</td>
       </tr>
     {% endfor %}
     </tbody>
@@ -499,7 +501,7 @@
   <h3>Antécédents ({{ dossier.antecedents|length }})</h3>
   <ul class="bullet">
     {% for a in dossier.antecedents %}
-      <li>{{ a }}</li>
+      <li>{{ a.texte }}{% if a.source_page %} <button class="src-btn" onclick="showSource('{{ a.source_excerpt|default('',true)|e }}', {{ a.source_page }})">p.{{ a.source_page }}</button>{% endif %}</li>
     {% endfor %}
   </ul>
 </div>
@@ -511,16 +513,109 @@
   <h3>Complications ({{ dossier.complications|length }})</h3>
   <ul class="bullet">
     {% for c in dossier.complications %}
-      <li>{{ c }}</li>
+      <li>{{ c.texte }}{% if c.source_page %} <button class="src-btn" onclick="showSource('{{ c.source_excerpt|default('',true)|e }}', {{ c.source_page }})">p.{{ c.source_page }}</button>{% endif %}</li>
     {% endfor %}
   </ul>
 </div>
 {% endif %}
 
+{# ---- Modal source ---- #}
+<div id="source-modal">
+  <div id="source-modal-inner">
+    <div id="source-header">
+      <span id="source-title">Document source</span>
+      <button id="source-close-btn" onclick="closeSource()">Fermer</button>
+    </div>
+    <div id="source-content"></div>
+  </div>
+</div>
+
 {% endblock %}
 
 {% block scripts %}
 <script>
+/* --- Source modal --- */
+let _sourceCache = null;
+
+function getDossierId() {
+  // filepath = "103_23056749/103_23056749_fusionne_cim10.json"
+  // dossier_id = "103_23056749"
+  const fp = '{{ filepath }}';
+  const parts = fp.split('/');
+  return parts.length > 1 ? parts.slice(0, -1).join('/') : '';
+}
+
+async function loadSourceTexts() {
+  if (_sourceCache !== null) return _sourceCache;
+  const dossierId = getDossierId();
+  if (!dossierId) { _sourceCache = {}; return _sourceCache; }
+  try {
+    const resp = await fetch('/api/source-text/' + dossierId);
+    if (resp.ok) { _sourceCache = await resp.json(); }
+    else { _sourceCache = {}; }
+  } catch (e) { _sourceCache = {}; }
+  return _sourceCache;
+}
+
+async function showSource(excerpt, page) {
+  const modal = document.getElementById('source-modal');
+  const content = document.getElementById('source-content');
+  const title = document.getElementById('source-title');
+
+  title.textContent = 'Document source — Page ' + page;
+  content.innerHTML = '<em style="color:#94a3b8;">Chargement...</em>';
+  modal.style.display = 'block';
+
+  const texts = await loadSourceTexts();
+  const allText = Object.values(texts).join('\n\n--- ---\n\n');
+
+  if (!allText) {
+    content.innerHTML = '<em style="color:#94a3b8;">Texte source non disponible</em>';
+    return;
+  }
+
+  // Chercher l'extrait dans le texte et le surligner
+  if (excerpt && excerpt.length > 10) {
+    const idx = allText.indexOf(excerpt);
+    if (idx >= 0) {
+      const before = allText.substring(0, idx);
+      const match = allText.substring(idx, idx + excerpt.length);
+      const after = allText.substring(idx + excerpt.length);
+      content.innerHTML = '';
+      content.appendChild(document.createTextNode(before));
+      const mark = document.createElement('mark');
+      mark.textContent = match;
+      mark.id = 'source-highlight';
+      content.appendChild(mark);
+      content.appendChild(document.createTextNode(after));
+      // Scroll vers le surlignage
+      setTimeout(() => {
+        const el = document.getElementById('source-highlight');
+        if (el) el.scrollIntoView({ behavior: 'smooth', block: 'center' });
+      }, 100);
+      return;
+    }
+  }
+
+  // Fallback : afficher le texte brut sans surlignage
+  content.textContent = allText;
+}
+
+function closeSource() {
+  document.getElementById('source-modal').style.display = 'none';
+}
+
+// Fermer le modal en cliquant sur le fond
+document.getElementById('source-modal').addEventListener('click', function(e) {
+  if (e.target === this) closeSource();
+});
+
+// Fermer avec Escape
+document.addEventListener('keydown', function(e) {
+  if (e.key === 'Escape') closeSource();
+});
+
+/* --- Reprocess --- */
 document.getElementById('reprocess-btn').addEventListener('click', async () => {
   const btn = document.getElementById('reprocess-btn');
   const status = document.getElementById('reprocess-status');

tests/test_medical.py

@@ -2,7 +2,7 @@
 
 import pytest
 
-from src.config import DossierMedical, Diagnostic
+from src.config import DossierMedical, Diagnostic, Antecedent, Complication
 from src.medical.cim10_extractor import (
     extract_medical_info,
     _lookup_cim10,
@@ -121,7 +121,7 @@ Devenir : sortie le 03/03."""
         assert any("Balthazar" in (i.score or "") for i in dossier.imagerie)
 
         # Complications
-        assert any("cutanée" in c.lower() for c in dossier.complications)
+        assert any("cutanée" in c.texte.lower() for c in dossier.complications)
 
     def test_extract_without_edsnlp(self):
         """Vérifie que l'extraction fonctionne sans résultat edsnlp."""
@@ -236,7 +236,7 @@ Devenir : sortie le 03/03."""
 
         dossier = extract_medical_info(parsed, text, edsnlp_result=edsnlp_result)
         # Fièvre et infection sont niées, ne doivent pas apparaître dans complications
-        complication_terms = [c.lower() for c in dossier.complications]
+        complication_terms = [c.texte.lower() for c in dossier.complications]
         assert "fièvre" not in complication_terms
         assert "infection" not in complication_terms
 
@@ -504,6 +504,44 @@ class TestIsValidAntecedent:
     def test_reject_texte_libre(self):
         assert not _is_valid_antecedent("(texte libre)")
 
+    # --- Artefacts CRH colonne gauche (médecins) ---
+    def test_reject_medecin_tag_start(self):
+        assert not _is_valid_antecedent(
+            "[MEDECIN] hospitalier - Syndrome anxio depressif suivi Dr [MEDECIN_39]"
+        )
+
+    def test_reject_medecin_assistant(self):
+        assert not _is_valid_antecedent(
+            "[MEDECIN] Assistant des Hôpitaux de Lyon - Bilan neurologique"
+        )
+
+    def test_reject_medecin_contractuel(self):
+        assert not _is_valid_antecedent("[MEDECIN] hospitalier contractuel")
+
+    def test_reject_dr_medecin_tag(self):
+        assert not _is_valid_antecedent("Dr [MEDECIN_7] (Caradoc)")
+
+    def test_reject_dr_chef_clinique(self):
+        assert not _is_valid_antecedent(
+            "Dr [MEDECIN_37] Chef de Clinique des Hôpitaux aucune aide"
+        )
+
+    def test_reject_de_bordeaux(self):
+        assert not _is_valid_antecedent("de Bordeaux")
+
+    def test_reject_de_lyon(self):
+        assert not _is_valid_antecedent("de Lyon")
+
+    def test_reject_secretariat(self):
+        assert not _is_valid_antecedent("Secrétariat : [TEL_3] - fracture en 2017")
+
+    def test_reject_aucune_aide(self):
+        assert not _is_valid_antecedent("aucune aide, pas d'ide, pas d'aide ménagère")
+
+    def test_accept_de_long_medical(self):
+        """'de' suivi d'une vraie description médicale longue passe."""
+        assert _is_valid_antecedent("dégénérescence maculaire liée à l'âge")
+
     # --- Cas limites ---
     def test_reject_too_short(self):
         assert not _is_valid_antecedent("de Bo")
@@ -541,3 +579,108 @@ class TestClassifierConfidence:
         result = classify(text)
         assert isinstance(result, str)
         assert result in ("crh", "trackare")
+
+
+class TestBackwardCompatAntecedent:
+    """Tests de rétrocompatibilité pour les antécédents et complications."""
+
+    def test_old_format_string_list(self):
+        """Charger un vieux JSON avec antecedents: ["HTA", "Diabète"]."""
+        d = DossierMedical.model_validate({
+            "antecedents": ["HTA", "Diabète type 2"],
+            "complications": ["Fièvre"],
+        })
+        assert len(d.antecedents) == 2
+        assert isinstance(d.antecedents[0], Antecedent)
+        assert d.antecedents[0].texte == "HTA"
+        assert d.antecedents[1].texte == "Diabète type 2"
+        assert len(d.complications) == 1
+        assert isinstance(d.complications[0], Complication)
+        assert d.complications[0].texte == "Fièvre"
+
+    def test_new_format_object_list(self):
+        """Charger un nouveau JSON avec antecedents: [{texte: "HTA", source_page: 1}]."""
+        d = DossierMedical.model_validate({
+            "antecedents": [{"texte": "HTA", "source_page": 2, "source_excerpt": "contexte HTA"}],
+            "complications": [{"texte": "Fièvre", "source_page": 3}],
+        })
+        assert d.antecedents[0].texte == "HTA"
+        assert d.antecedents[0].source_page == 2
+        assert d.antecedents[0].source_excerpt == "contexte HTA"
+        assert d.complications[0].source_page == 3
+
+    def test_mixed_format(self):
+        """Un mélange de strings et d'objets est converti correctement."""
+        d = DossierMedical.model_validate({
+            "antecedents": ["HTA", {"texte": "Diabète", "source_page": 1}],
+        })
+        assert len(d.antecedents) == 2
+        assert d.antecedents[0].texte == "HTA"
+        assert d.antecedents[0].source_page is None
+        assert d.antecedents[1].texte == "Diabète"
+        assert d.antecedents[1].source_page == 1
+
+    def test_empty_list(self):
+        d = DossierMedical.model_validate({"antecedents": [], "complications": []})
+        assert d.antecedents == []
+        assert d.complications == []
+
+    def test_antecedent_extraction_produces_objects(self):
+        """L'extraction produit bien des objets Antecedent."""
+        parsed = {
+            "type": "crh",
+            "patient": {"sexe": "M"},
+            "sejour": {},
+            "diagnostics": [],
+        }
+        text = "Antécédents :\n- Diabète type 2\n- Hypertension artérielle\n\nHistoire de la maladie"
+        dossier = extract_medical_info(parsed, text)
+        assert len(dossier.antecedents) >= 1
+        assert all(isinstance(a, Antecedent) for a in dossier.antecedents)
+        textes = [a.texte for a in dossier.antecedents]
+        assert "Diabète type 2" in textes
+
+    def test_complication_extraction_produces_objects(self):
+        """L'extraction produit bien des objets Complication."""
+        parsed = {
+            "type": "crh",
+            "patient": {"sexe": "M"},
+            "sejour": {},
+            "diagnostics": [],
+        }
+        text = "Patient avec fièvre post-opératoire."
+        dossier = extract_medical_info(parsed, text)
+        assert all(isinstance(c, Complication) for c in dossier.complications)
+
+
+class TestSourceTrackingFields:
+    """Tests que les champs source_page/source_excerpt existent sur les modèles."""
+
+    def test_biologie_source_fields(self):
+        from src.config import BiologieCle
+        b = BiologieCle(test="CRP", valeur="45", source_page=2, source_excerpt="CRP=45")
+        assert b.source_page == 2
+        assert b.source_excerpt == "CRP=45"
+
+    def test_imagerie_source_fields(self):
+        from src.config import Imagerie
+        i = Imagerie(type="TDM", source_page=3)
+        assert i.source_page == 3
+
+    def test_traitement_source_fields(self):
+        from src.config import Traitement
+        t = Traitement(medicament="Paracétamol", source_page=4)
+        assert t.source_page == 4
+
+    def test_acte_source_fields(self):
+        from src.config import ActeCCAM
+        a = ActeCCAM(texte="Cholécystectomie", source_page=5)
+        assert a.source_page == 5
+
+    def test_antecedent_source_fields(self):
+        a = Antecedent(texte="HTA", source_page=1, source_excerpt="Antécédents: HTA")
+        assert a.source_page == 1
+
+    def test_complication_source_fields(self):
+        c = Complication(texte="Fièvre", source_page=2)
+        assert c.source_page == 2

tests/test_viewer.py

@@ -143,3 +143,15 @@ class TestDetailPageLoads:
         """Un fichier inexistant retourne 404."""
         response = client.get("/dossier/nonexistent.json")
         assert response.status_code == 404
+
+
+class TestSourceTextEndpoint:
+    def test_source_text_404_nonexistent(self, client):
+        """Un dossier inexistant retourne 404."""
+        response = client.get("/api/source-text/nonexistent_dossier")
+        assert response.status_code == 404
+
+    def test_source_text_security_path_traversal(self, client):
+        """Path traversal bloqué."""
+        response = client.get("/api/source-text/../../etc")
+        assert response.status_code in (403, 404)