build_aidefansi_kb.py 26 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472
  1. #!/usr/bin/env python3
  2. """由采集 JSON 生成「艾德范思肠道微生态检测报告」指标知识库(JSON + Markdown 视图)。
  3. 用法:
  4. python build_aidefansi_kb.py [采集JSON] [输出目录]
  5. 输入:parse_aidefansi_report.py 产出的采集 JSON
  6. 输出:<输出目录>/艾德范思报告指标知识库.json、艾德范思报告指标知识库.md
  7. 约定:JSON 为单一事实源,MD 为派生视图;改数据只改 JSON 或采集源,随后重跑本脚本。
  8. """
  9. import json
  10. import re
  11. import sys
  12. from datetime import date
  13. from pathlib import Path
  14. DEFAULT_SRC = "docs/参考资料/艾德范思-肠道微生态检测报告-模板.json"
  15. DEFAULT_OUT = "docs/参考资料/知识库"
  16. # ---------------------------------------------------------------- 评分规则
  17. def tier(state, lo, hi):
  18. return {"状态": state, "最小值": lo, "最大值": hi}
  19. ABILITY_RANGE = {"最小值": 0, "最大值": 100}
  20. ABILITY_TIERS = [tier("弱", 0, 30), tier("稍弱", 30, 60), tier("正常", 60, 100)]
  21. ABX_TIERS = [tier("正常", 0, 70), tier("注意", 70, 90), tier("耐药", 90, 100)]
  22. RISK_TIERS = [tier("低风险", 0, 50), tier("中风险", 50, 70), tier("高风险", 70, 100)]
  23. SCORING_RULES = {
  24. "能力功能评分": {
  25. "适用": "肠道免疫力、营养物质代谢、短链脂肪酸合成、维生素、天然色素、氨基酸、"
  26. "谷胱甘肽、胆汁酸、矿物质、多轴系统、毒素降解",
  27. "标准范围": dict(ABILITY_RANGE),
  28. "状态分级": list(ABILITY_TIERS),
  29. "说明": "报告统一口径:<30 为弱,30-60 为稍弱,>60 为正常",
  30. },
  31. "抗生素耐药性评分": {
  32. "适用": "抗生素耐药性评估",
  33. "标准范围": dict(ABILITY_RANGE),
  34. "状态分级": list(ABX_TIERS),
  35. "说明": "<70 正常,70-90 注意,>90 耐药",
  36. },
  37. "风险指数评分": {
  38. "适用": "肠道菌群过敏风险、肠道菌群与疾病相关风险",
  39. "标准范围": dict(ABILITY_RANGE),
  40. "状态分级": list(RISK_TIERS),
  41. "说明": "0-50 低风险,50-70 中风险,70-100 高风险",
  42. },
  43. "肠道微生物健康指数(GMHI)": {
  44. "适用": "肠道菌群状态",
  45. "标准范围": {"最小值": -3, "最大值": 1},
  46. "状态分级": [tier("不健康", -3, -2), tier("不健康倾向", -2, -1), tier("亚健康", -1, 0),
  47. tier("健康倾向", 0, 0.5), tier("健康", 0.5, 1)],
  48. "说明": "GMHI>0 健康、GMHI<0 非健康;对应 2度/1度/0度/-1度/-2度 五档",
  49. },
  50. "肠道总评分": {
  51. "适用": "肠道菌群整体健康总览",
  52. "标准范围": dict(ABILITY_RANGE),
  53. "状态分级": [tier("需关注", 0, 40), tier("一般", 40, 60), tier("良好", 60, 80), tier("优秀", 80, 100)],
  54. "说明": "报告未给出总评分分级口径,此处沿用通用健康评分标准,供跨报告横向比较参考",
  55. },
  56. }
  57. # 各章节指标语义(用于生成偏高/偏低影响与调整建议)
  58. SECTION_SEMANTICS = {
  59. "肠道免疫力评估": ("能力", "肠道免疫力相关功能"),
  60. "营养物质代谢能力": ("能力", "营养物质代谢能力"),
  61. "短链脂肪酸合成能力": ("合成", "短链脂肪酸合成"),
  62. "维生素评估": ("合成", "维生素合成能力"),
  63. "天然色素合成能力": ("合成", "天然色素合成能力"),
  64. "氨基酸合成能力": ("合成", "氨基酸合成能力"),
  65. "谷胱甘肽合成能力": ("合成", "谷胱甘肽合成能力"),
  66. "胆汁酸合成能力": ("合成", "胆汁酸合成能力"),
  67. "健康相关矿物质评估": ("合成", "矿物质吸收/合成能力"),
  68. "多轴系统风险评估": ("合成", "多轴系统物质合成能力"),
  69. "毒素降解能力评估": ("降解", "毒素降解能力"),
  70. "抗生素耐药性评估": ("耐药", "抗生素耐药性"),
  71. "肠道菌群过敏风险评估": ("风险", "过敏风险"),
  72. "肠道菌群与疾病相关风险评估": ("风险", "疾病风险"),
  73. }
  74. # 指标名归一化(分项说明页标题 -> 整体健康页名称)
  75. ALIAS = {
  76. "甲硫氨酸": "蛋氨酸",
  77. "精神分裂": "精神分裂症",
  78. "抑郁、焦虑": "抑郁症",
  79. "糖(碳水化合物)": "糖(碳水化合物)",
  80. "厚壁菌门/拟杆菌门(F/B)": "F/B比值",
  81. }
  82. ABILITY_HINT = {
  83. "合成": "可通过增加膳食纤维与益生元(全谷物、豆类、洋葱、香蕉)及发酵食品摄入改善",
  84. "能力": "建议结合饮食结构调整与益生菌/益生元补充,必要时咨询专业人士",
  85. "降解": "建议减少相关环境暴露,增加膳食纤维与发酵食品,促进肠道菌群解毒通路",
  86. "耐药": "提示该类抗生素耐药基因水平偏高,建议减少或避免此类抗生素使用(遵医嘱)",
  87. "风险": "维持健康饮食与生活方式,定期复查肠道菌群变化",
  88. "风险": "维持健康饮食与生活方式,定期复查肠道菌群变化",
  89. }
  90. def norm(name):
  91. return ALIAS.get(name, name).replace("(", "(").replace(")", ")").replace("、", "、").strip()
  92. def first_sentence(text, limit=110):
  93. if not text:
  94. return ""
  95. t = re.sub(r"\s+", "", text)
  96. m = re.split(r"[。;]", t)
  97. out = next((s for s in m if len(s) > 8), t)
  98. return out[:limit]
  99. def clean_para(text, limit=200):
  100. return re.sub(r"\s+", "", text or "")[:limit]
  101. def make_indicator(name, value, status, kind, verb, detail=None):
  102. """按既有 KB schema 构造单条指标定义。"""
  103. detail = detail or {}
  104. ranges = ABX_TIERS if kind == "耐药" else (RISK_TIERS if kind == "风险" else ABILITY_TIERS)
  105. std = dict(ABILITY_RANGE)
  106. if kind == "风险":
  107. high = f"{name}风险指数偏高,提示需重点关注并预防,建议结合临床检查进一步评估。"
  108. low = f"{name}风险指数较低,保持现有健康状态即可。"
  109. elif kind == "耐药":
  110. high = f"{name}耐药性评估为注意/耐药,为防止抗生素耐药,建议减少或避免此类抗生素使用。"
  111. low = f"{name}耐药性在正常范围,合理用药即可。"
  112. else:
  113. high = f"{name}{verb}水平偏高,可能提示该项{verb}偏强或代偿性升高,结合其他指标综合判断。"
  114. low = f"{name}{verb}水平偏低,提示该项{verb}不足,建议关注并针对性干预。"
  115. hint = ABILITY_HINT.get("能力" if kind == "能力" else
  116. ("合成" if kind == "合成" else ("降解" if kind == "降解" else
  117. ("耐药" if kind == "耐药" else "风险"))), "")
  118. intro = clean_para(detail.get("相关介绍") or detail.get("疾病介绍", ""), 160) or f"{name}{verb}能力评估"
  119. advice = (clean_para(detail.get("调理建议") or detail.get("食物来源")
  120. or detail.get("注意事项") or detail.get("温馨提示"), 160) or hint)
  121. return {
  122. "名称": name,
  123. "标准范围": std,
  124. "状态分级": list(ranges),
  125. "说明": intro,
  126. "偏高影响": high,
  127. "偏低影响": low,
  128. "调整建议": advice,
  129. "模板值": value,
  130. "模板状态": status,
  131. }
  132. def build_section(rows, section):
  133. kind, verb = SECTION_SEMANTICS.get(section, ("能力", "能力"))
  134. return {
  135. "标准范围": dict(ABILITY_RANGE),
  136. "状态分级": (ABX_TIERS if kind == "耐药" else (RISK_TIERS if kind == "风险" else ABILITY_TIERS)),
  137. "指标": [make_indicator(r["名称"], r["数值"], r["状态"], kind, verb) for r in rows],
  138. }
  139. # ---------------------------------------------------------------- 主构建
  140. def build(src_path):
  141. data = json.loads(Path(src_path).read_text(encoding="utf-8"))
  142. # 分项说明页按归一化名称索引
  143. details = {}
  144. for d in data.get("指标详解", []):
  145. details[norm(d["名称"])] = d
  146. kb = {
  147. "元数据": {
  148. "名称": "艾德范思肠道微生态检测报告指标知识库",
  149. "版本": "1.0",
  150. "报告类型": "肠道微生态检测报告(宏基因组学,Greengene 数据库注释)",
  151. "报告机构": data.get("报告机构", {}),
  152. "数据来源": data.get("文件名", "") + ".pdf(模板报告,123页)",
  153. "建立日期": date.today().isoformat(),
  154. "用途": "用于艾德范思(Advanced Bios)报告指标值的自动化状态比对与解读;"
  155. "取值规则:遍历状态分级,若值落在 [最小值, 最大值) 区间内则匹配该状态",
  156. "关联文件": {
  157. "采集源": "docs/参考资料/艾德范思-肠道微生态检测报告-模板.json",
  158. "解析脚本": "docs/scripts/parse_aidefansi_report.py",
  159. "姊妹库": "报告指标知识库.json(谷禾健康等 5 份报告口径,风险指数为 0-1 量表,勿混用)",
  160. },
  161. "口径警示": "艾德范思报告的风险指数为 0-100(低风险 0-50 / 中风险 50-70 / 高风险 70-100),"
  162. "与既有 报告指标知识库.json(0-1 量表)不同,解析时须先判定报告厂商再套用分级",
  163. },
  164. "通用评分规则": SCORING_RULES,
  165. }
  166. sections = ["肠道免疫力评估", "营养物质代谢能力", "短链脂肪酸合成能力", "维生素评估",
  167. "天然色素合成能力", "氨基酸合成能力", "谷胱甘肽合成能力", "胆汁酸合成能力",
  168. "健康相关矿物质评估", "多轴系统风险评估", "毒素降解能力评估",
  169. "抗生素耐药性评估", "肠道菌群过敏风险评估", "肠道菌群与疾病相关风险评估"]
  170. for sec in sections:
  171. rows = data.get(sec, [])
  172. if not rows:
  173. continue
  174. kb[sec] = build_section(rows, sec)
  175. # 用分项说明页补充 说明/调整建议
  176. for ind in kb[sec]["指标"]:
  177. d = details.get(norm(ind["名称"]))
  178. if d:
  179. if d.get("相关介绍"):
  180. ind["说明"] = clean_para(d["相关介绍"], 160)
  181. advice = d.get("食物来源") or d.get("调理建议") or d.get("注意事项") or d.get("温馨提示")
  182. if advice:
  183. ind["调整建议"] = clean_para(advice, 160)
  184. # 疾病风险:合并整体健康页 14 项 + 分项说明页独有项
  185. risk_names = {}
  186. for r in kb["肠道菌群与疾病相关风险评估"]["指标"]:
  187. risk_names[norm(r["名称"])] = r
  188. notes = data.get("疾病介绍与调理建议", {})
  189. for name, note in notes.items():
  190. key = norm(name)
  191. if key in risk_names:
  192. if note.get("调理建议"):
  193. risk_names[key]["调整建议"] = clean_para(note["调理建议"], 160)
  194. if note.get("疾病介绍"):
  195. risk_names[key]["疾病介绍"] = clean_para(note["疾病介绍"], 260)
  196. risk_names[key]["说明"] = clean_para(note["疾病介绍"], 160)
  197. else:
  198. risk_names[key] = make_indicator(key, "", "", "风险", "风险",
  199. {"相关介绍": note.get("疾病介绍", ""),
  200. "调理建议": note.get("调理建议", "")})
  201. kb["肠道菌群与疾病相关风险评估"]["指标"] = list(risk_names.values())
  202. # 肠道菌群整体健康(首页总览)
  203. overview = data.get("报告概述", {})
  204. kb["肠道菌群整体健康"] = {
  205. "标准范围": dict(ABILITY_RANGE),
  206. "状态分级": list(ABILITY_TIERS),
  207. "指标": [
  208. {"名称": "肠道总评分", "说明": "肠道菌群整体健康综合评分",
  209. "标准范围": dict(ABILITY_RANGE), "状态分级": SCORING_RULES["肠道总评分"]["状态分级"],
  210. "偏高影响": "总分越高代表肠道菌群整体状态越好。",
  211. "偏低影响": "总分偏低提示肠道菌群整体状态有待改善。",
  212. "调整建议": "增加膳食纤维多样性与发酵食品摄入,规律作息与运动,必要时益生菌干预。",
  213. "模板值": overview.get("肠道总评分", {}).get("数值", ""),
  214. "模板状态": overview.get("肠道总评分", {}).get("状态", "")},
  215. {"名称": "肠道菌群健康状态(GMHI)", "说明": "基于物种分类评估健康状况的稳健指标,代表菌为健康相关与疾病相关两组微生物的平衡",
  216. "标准范围": SCORING_RULES["肠道微生物健康指数(GMHI)"]["标准范围"],
  217. "状态分级": SCORING_RULES["肠道微生物健康指数(GMHI)"]["状态分级"],
  218. "偏高影响": "GMHI>0 代表健康,值越大越健康。",
  219. "偏低影响": "GMHI<0 代表非健康,值越小疾病可能性越大。",
  220. "调整建议": "合理用药避免抗生素滥用,减少加工食品,适当补充益生元与益生菌。",
  221. "模板值": overview.get("肠道菌群健康状态", {}).get("数值", ""),
  222. "模板状态": overview.get("肠道菌群健康状态", {}).get("状态", "")},
  223. {"名称": "肠道微生物定植抗力(B/E)", "说明": "双歧杆菌/肠杆菌比值,反映肠道原籍菌群抑制外籍致病菌定植的能力",
  224. "标准范围": {"最小值": 0.05778, "最大值": 100},
  225. "状态分级": [tier("偏低", 0, 0.05778), tier("正常", 0.05778, 100)],
  226. "偏高影响": "定植抗力高,肠道抵御病原菌定植能力强。",
  227. "偏低影响": "定植抗力低时病原菌易大量增殖并可能突破肠黏膜,导致感染风险升高。",
  228. "调整建议": "减少抗生素使用,适当补充双歧杆菌等益生菌,增加肠道定植抗力。",
  229. "模板值": overview.get("肠道微生物定植抗力", {}).get("数值", ""),
  230. "模板状态": overview.get("肠道微生物定植抗力", {}).get("状态", "")},
  231. {"名称": "肠道菌群多样性", "说明": "菌群种类与均匀度的综合指标,参考范围 3-10",
  232. "标准范围": {"最小值": 3, "最大值": 10},
  233. "状态分级": [tier("偏低", 0, 3), tier("正常", 3, 10)],
  234. "偏高影响": "多样性高通常提示菌群生态稳定。",
  235. "偏低影响": "多样性低提示微生态稳定性差、抵御外源菌入侵能力弱。",
  236. "调整建议": "摄入多样化饮食与蔬果,增加体育锻炼与户外活动,合理用药,补充乳酸杆菌/双歧杆菌及果胶、菊粉、低聚果糖等益生元。",
  237. "模板值": overview.get("肠道菌群多样性", {}).get("数值", ""),
  238. "模板状态": overview.get("肠道菌群多样性", {}).get("状态", "")},
  239. {"名称": "肠道微生物种类", "说明": "检出菌种数量,人群中参考范围 200-2000 种",
  240. "标准范围": {"最小值": 200, "最大值": 2000},
  241. "状态分级": [tier("偏低", 0, 200), tier("正常", 200, 2000)],
  242. "偏高影响": "种类数多提示多样性高。",
  243. "偏低影响": "种类数偏少提示菌群多样性不足。",
  244. "调整建议": "增加膳食种类与膳食纤维,避免低纤维饮食。",
  245. "模板值": overview.get("肠道微生物种类", {}).get("数值", ""),
  246. "模板状态": overview.get("肠道微生物种类", {}).get("状态", "")},
  247. {"名称": "肠型", "说明": "类似血型的肠道菌群组成分型;肠型Ⅰ偏向拟杆菌属,擅长分解碳水化合物、利用脂肪与蛋白质",
  248. "标准范围": None, "状态分级": None,
  249. "偏高影响": "—", "偏低影响": "—",
  250. "调整建议": "肠型由长期饮食生活习惯形成,短期改变不易转换,应以膳食结构长期调整为主。",
  251. "模板值": overview.get("肠型", {}).get("数值", ""),
  252. "模板状态": overview.get("肠型", {}).get("状态", "")},
  253. {"名称": "厚壁菌门/拟杆菌门(F/B)", "说明": "F/B 大致反映菌群平衡状态;高脂高糖高蛋白饮食通常 F/B 升高,高纤维饮食使其降低",
  254. "标准范围": {"最小值": 0, "最大值": 100}, "状态分级": None,
  255. "偏高影响": "F/B 偏高(偏向厚壁菌门)常见于高脂高蛋白饮食模式。",
  256. "偏低影响": "F/B 偏低(偏向拟杆菌门)常见于高纤维/植物性饮食模式。",
  257. "调整建议": "以增加膳食纤维、减少精制脂肪的膳食调整为主。",
  258. "模板值": overview.get("厚壁菌门/拟杆菌门(F/B)", {}).get("数值", ""),
  259. "模板状态": overview.get("厚壁菌门/拟杆菌门(F/B)", {}).get("状态", "")},
  260. {"名称": "有益菌", "说明": "定植于胃肠道、产生活性物质并对宿主有益的微生物总量",
  261. "标准范围": dict(ABILITY_RANGE), "状态分级": list(ABILITY_TIERS),
  262. "偏高影响": "有益菌充足,菌群平衡维持良好。",
  263. "偏低影响": "有益菌偏低提示抵抗疾病能力弱,易诱发菌群失调。",
  264. "调整建议": "摄入酸奶等发酵食品与益生元(洋葱、豆类、香蕉);餐后或睡前服用益生菌效果更佳;减少加工食品与不必要的抗生素。",
  265. "模板值": overview.get("有益菌", {}).get("数值", ""),
  266. "模板状态": overview.get("有益菌", {}).get("状态", "")},
  267. {"名称": "致病菌", "说明": "正常情况下肠道内不应检出的致病性微生物",
  268. "标准范围": None, "状态分级": [tier("正常", 0, 0), tier("检出", 0.0001, 100)],
  269. "偏高影响": "检出致病菌提示急性感染风险,应及时就医。",
  270. "偏低影响": "—",
  271. "调整建议": "注意饮食卫生与手部清洁,避免生食与不洁水源;出现症状及时就医。",
  272. "模板值": overview.get("致病菌", {}).get("数值", ""),
  273. "模板状态": overview.get("致病菌", {}).get("状态", "")},
  274. ],
  275. }
  276. # 细菌参考库
  277. for key, desc in (("有益菌检出", "有益菌参考"), ("条件致病菌检出", "条件致病菌参考"),
  278. ("致病菌检出", "致病菌参考")):
  279. rows = data.get(key, [])
  280. if not rows:
  281. continue
  282. kb[desc] = {
  283. "说明": "检出:表示肠道中检测出相应微生物;结果为 -:表示未检出",
  284. "条目": [{"中文名": r["中文名"], "拉丁名": r["拉丁名"], "功能说明": clean_para(r.get("功能说明", ""), 200),
  285. "模板检测结果%": r.get("检测结果%", ""), "模板结果评价": r.get("结果评价", "")}
  286. for r in rows],
  287. }
  288. # 菌群组成参考范围(门/属/特征菌)
  289. comp = data.get("肠道菌群组成", {})
  290. kb["菌群组成参考范围"] = {
  291. "说明": "相对丰度参考范围(%),↑ 表示高于参考上限、↓ 表示低于参考下限",
  292. "门水平": comp.get("门水平", []),
  293. "属水平": comp.get("属水平", []),
  294. "特征菌科属种": comp.get("特征菌科属种", []),
  295. }
  296. # 疾病 / 过敏 相关菌参考
  297. kb["疾病相关菌参考"] = {
  298. "说明": "各疾病相关菌种及方向(正相关=丰度升高与风险同向;补充=建议补充的益生菌)",
  299. "疾病": data.get("疾病相关菌", {}),
  300. }
  301. kb["过敏相关菌参考"] = {
  302. "说明": "各过敏原相关菌种及方向,用于菌群角度解读过敏风险",
  303. "过敏原": data.get("过敏相关菌", {}),
  304. }
  305. kb["疾病介绍与调理建议"] = {
  306. "说明": "报告原文疾病介绍与调理建议(用于生成解读文案,不构成诊疗建议)",
  307. "疾病": {k: {"疾病介绍": clean_para(v.get("疾病介绍", ""), 400),
  308. "调理建议": clean_para(v.get("调理建议", ""), 300)}
  309. for k, v in (data.get("疾病介绍与调理建议") or {}).items()},
  310. }
  311. return kb, data
  312. # ---------------------------------------------------------------- MD 视图
  313. def md_table(headers, rows):
  314. out = ["| " + " | ".join(headers) + " |", "| " + " | ".join(["---"] * len(headers)) + " |"]
  315. for r in rows:
  316. cells = [str(c).replace("|", "\\|").replace("\n", " ") for c in r]
  317. out.append("| " + " | ".join(cells) + " |")
  318. return "\n".join(out)
  319. def tiers_text(tiers):
  320. if not tiers:
  321. return "—"
  322. return ";".join(f"{t['状态']} {t['最小值']}~{t['最大值']}" for t in tiers)
  323. _CN = "零一二三四五六七八九"
  324. def cn_num(n):
  325. """阿拉伯数字转中文数字(1-99)。"""
  326. if n < 10:
  327. return _CN[n]
  328. if n < 20:
  329. return "十" + (_CN[n % 10] if n % 10 else "")
  330. if n < 100:
  331. return _CN[n // 10] + "十" + (_CN[n % 10] if n % 10 else "")
  332. return str(n)
  333. def render_md(kb):
  334. meta = kb["元数据"]
  335. L = [f"# {meta['名称']}(视图)", "",
  336. "> 由艾德范思报告指标知识库.json 派生,勿手工编辑;改动请改 JSON 后重跑 build_aidefansi_kb.py。",
  337. "",
  338. f"- 版本:{meta['版本']}",
  339. f"- 报告类型:{meta['报告类型']}",
  340. f"- 数据来源:{meta['数据来源']}",
  341. f"- 建立日期:{meta['建立日期']}",
  342. f"- 报告机构:{meta['报告机构'].get('名称','')} {meta['报告机构'].get('官网','')}",
  343. "",
  344. f"⚠ {meta['口径警示']}", "",
  345. "## 一、通用评分规则(0-100 为主的能力分 vs 风险指数,切勿混用)", "",
  346. md_table(["规则", "适用范围", "标准范围", "状态分级", "说明"],
  347. [[k, v.get("适用", "—"), f"{v['标准范围']['最小值']}~{v['标准范围']['最大值']}",
  348. tiers_text(v.get("状态分级")), v.get("说明", "")]
  349. for k, v in kb["通用评分规则"].items()]),
  350. ""]
  351. order = ["肠道菌群整体健康", "肠道免疫力评估", "营养物质代谢能力", "短链脂肪酸合成能力", "维生素评估",
  352. "天然色素合成能力", "氨基酸合成能力", "谷胱甘肽合成能力", "胆汁酸合成能力",
  353. "健康相关矿物质评估", "多轴系统风险评估", "毒素降解能力评估", "抗生素耐药性评估",
  354. "肠道菌群过敏风险评估", "肠道菌群与疾病相关风险评估"]
  355. n = 2
  356. for sec in order:
  357. if sec not in kb:
  358. continue
  359. s = kb[sec]
  360. L += [f"## {cn_num(n)}、{sec}", "",
  361. f"标准范围:{s['标准范围']['最小值']}~{s['标准范围']['最大值']};"
  362. f"状态分级:{tiers_text(s.get('状态分级'))}", "",
  363. md_table(["指标", "说明", "偏高影响", "偏低影响", "调整建议", "模板值", "模板状态"],
  364. [[i["名称"], i["说明"][:60], i["偏高影响"][:50], i["偏低影响"][:50],
  365. i["调整建议"][:60], i.get("模板值", ""), i.get("模板状态", "")]
  366. for i in s["指标"]]), ""]
  367. n += 1
  368. for key in ("有益菌参考", "条件致病菌参考", "致病菌参考"):
  369. if key not in kb:
  370. continue
  371. L += [f"## {cn_num(n)}、{key}", "", kb[key]["说明"], "",
  372. md_table(["中文名", "拉丁名", "功能说明", "模板检出%", "模板评价"],
  373. [[e["中文名"], e["拉丁名"], e["功能说明"][:70], e["模板检测结果%"], e["模板结果评价"]]
  374. for e in kb[key]["条目"]]), ""]
  375. n += 1
  376. comp = kb["菌群组成参考范围"]
  377. L += [f"## {cn_num(n)}、菌群组成参考范围", "", comp["说明"], "", "### 门水平", "",
  378. md_table(["拉丁名", "中文名", "参考范围%"],
  379. [[r["拉丁名"], r["中文名"], r["参考范围%"]] for r in comp["门水平"]]), "",
  380. "### 属水平", "",
  381. md_table(["拉丁名", "中文名", "参考范围%"],
  382. [[r["拉丁名"], r["中文名"], r["参考范围%"]] for r in comp["属水平"]]), "",
  383. "### 特征菌科属种", "",
  384. md_table(["拉丁学名", "中文名称", "阈值%"],
  385. [[r["拉丁学名"], r["中文名称"], r["阈值%"]] for r in comp["特征菌科属种"]]), ""]
  386. n += 1
  387. for key, label in (("疾病相关菌参考", "疾病"), ("过敏相关菌参考", "过敏原")):
  388. block = kb[key]
  389. items = block.get(label, {})
  390. L += [f"## {cn_num(n)}、{key}", "", block["说明"], ""]
  391. for name, rows in items.items():
  392. L += [f"### {name}({len(rows)} 项)", "",
  393. md_table(["菌种", "拉丁名", "相关性", "参考范围"],
  394. [[r["菌种"], r["拉丁名"], r["相关性"], r["参考范围"]] for r in rows]), ""]
  395. n += 1
  396. dn = kb["疾病介绍与调理建议"]
  397. L += [f"## {cn_num(n)}、疾病介绍与调理建议(报告原文摘录)", "", dn["说明"], "",
  398. md_table(["疾病", "疾病介绍", "调理建议"],
  399. [[k, v["疾病介绍"][:150], v["调理建议"][:100]] for k, v in dn["疾病"].items()]), ""]
  400. return "\n".join(L)
  401. def main():
  402. src = Path(sys.argv[1]) if len(sys.argv) > 1 else Path(DEFAULT_SRC)
  403. out_dir = Path(sys.argv[2]) if len(sys.argv) > 2 else Path(DEFAULT_OUT)
  404. kb, data = build(src)
  405. out_dir.mkdir(parents=True, exist_ok=True)
  406. jp = out_dir / "艾德范思报告指标知识库.json"
  407. mp = out_dir / "艾德范思报告指标知识库.md"
  408. jp.write_text(json.dumps(kb, ensure_ascii=False, indent=1), encoding="utf-8")
  409. mp.write_text(render_md(kb) + "\n", encoding="utf-8")
  410. print(f"wrote {jp}")
  411. print(f"wrote {mp}")
  412. for k, v in kb.items():
  413. if isinstance(v, dict) and "指标" in v:
  414. print(f" {k}: {len(v['指标'])} 项")
  415. elif isinstance(v, dict) and "条目" in v:
  416. print(f" {k}: {len(v['条目'])} 条")
  417. elif isinstance(v, dict) and "门水平" in v:
  418. print(f" {k}: 门{len(v['门水平'])} 属{len(v['属水平'])} 特征菌{len(v['特征菌科属种'])}")
  419. if __name__ == "__main__":
  420. main()