feat(langgraph): pdf_parser.py 北京菌群 related_bacteria/gutStatusLevel 解析 + report_parse_agent.py schema 扩展

Ultraworked with [Sisyphus](https://github.com/obra/sisyphus) - Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

cfc-langgraph/app/agents/report_parse_agent.py

@@ -51,8 +51,9 @@ REPORT_PARSE_GUT_TEMPLATE = """你是一个肠道菌群检测报告解析专家
 {{
     "overview": {{ "person_name": "", "report_number": "", "age": 0, "gender": "male/female",
         "overallScore": 0, "gutHealthScore": 0, "chronicDiseaseScore": 0, "nutritionScore": 0,
-        "gutAge": "", "gutType": "" }},
-    "disease_risks": [{{"name": "", "value": "", "status": ""}}],
+        "gutAge": "", "gutType": "", "gutStatusLevel": "" }},
+    "disease_risks": [{{"name": "", "value": "", "status": "",
+        "related_bacteria": [{{"bacteria": "", "correlation": "", "value": "", "status": ""}}]}}],
     "nutrition": [{{"name": "", "value": "", "status": ""}}],
     "amino_acids": [{{"name": "", "value": "", "status": ""}}],
     "vitamins": [{{"name": "", "value": "", "status": ""}}],

cfc-langgraph/app/parsers/pdf_parser.py

@@ -1519,8 +1519,23 @@ def _parse_bj_core_bacteria(text):
     return bacteria
 
 
+def _parse_related_bacteria_in_chunk(chunk):
+    """从文本片段提取关联菌列表（如：乳杆菌属（缺乏）、另枝菌属（超标））"""
+    bacteria = []
+    for bm in re.finditer(
+            r'([\u4e00-\u9fa5]{2,10}?(?:菌|属|科|杆菌|球菌|梭菌))[（(]\s*(超标|缺乏|偏高|偏低|异常|正常)\s*[）)]',
+            chunk):
+        bacteria.append({
+            'bacteria': bm.group(1),
+            'correlation': bm.group(2),
+            'value': '',
+            'status': '',
+        })
+    return bacteria
+
+
 def _parse_bj_disease_risks(text):
-    """提取疾病风险评估"""
+    """提取疾病风险评估（含关联菌 P0）"""
     diseases = []
     for keyword in DISEASE_KEYWORDS_BJ:
         idx = text.find(keyword)
@@ -1529,11 +1544,38 @@ def _parse_bj_disease_risks(text):
         val_m = re.search(r'(\d+\.\d+)', chunk)
         level_m = re.search(r'(低风险|较低风险|中度风险|较高风险|高风险)', chunk)
         if val_m:
-            diseases.append({
+            entry = {
                 '疾病': keyword,
                 '风险指数': float(val_m.group(1)),
                 '风险等级': level_m.group(1) if level_m else '',
-            })
+            }
+            bacteria = _parse_related_bacteria_in_chunk(chunk)
+            if bacteria:
+                entry['related_bacteria'] = bacteria
+            diseases.append(entry)
+
+    # 补充：从「疾病肠菌评估（异常项）」章节提取关联菌，按疾病名匹配补挂
+    ab_idx = text.find('疾病肠菌评估')
+    if ab_idx != -1:
+        ab_sec = text[ab_idx:ab_idx + 1800]
+        for entry in diseases:
+            if entry.get('related_bacteria'):
+                continue
+            kw_idx = ab_sec.find(entry['疾病'])
+            if kw_idx == -1:
+                continue
+            # 截断到下一个疾病关键字，避免跨疾病串扰
+            end = len(ab_sec)
+            for other in DISEASE_KEYWORDS_BJ:
+                if other == entry['疾病']:
+                    continue
+                oi = ab_sec.find(other, kw_idx + len(entry['疾病']))
+                if oi != -1 and oi < end:
+                    end = oi
+            chunk = ab_sec[kw_idx:end]
+            bacteria = _parse_related_bacteria_in_chunk(chunk)
+            if bacteria:
+                entry['related_bacteria'] = bacteria
     return diseases
 
 
@@ -1931,6 +1973,11 @@ def _parse_beijing_report(file_path: str) -> dict:
     if pathogens:
         result['胃肠道感染病原体'] = pathogens
 
+    # 肠道菌群状态 → overview.gutStatusLevel（P1）
+    status = result.get('肠道微生物健康指数', {}).get('肠道菌群状态')
+    if status:
+        result['overview']['gutStatusLevel'] = status
+
     # 营养物质 → 标准字段映射
     metabolism = result.get('营养物质及营养素代谢评估', {})
     if metabolism.get('主要营养物质代谢'):