File indexing completed on 2024-11-24 04:44:41

 #!/usr/bin/env python3
 # SPDX-FileCopyrightText: 2022 Volker Krause <vkrause@kde.org>
 # SPDX-License-Identifier: LGPL-2.0-or-later
 
 import argparse
 import csv
 import json
 import os
 import requests
 import zipfile
 
 def writeToFile(fileName, content):
     f = open(fileName, 'wb')
     f.write(content)
     f.close();
 
 def countDepth(n):
     depth = 0
     while n[2*depth:].startswith('- '):
         depth += 1
     return depth + 2
 
 def findNodes(node, func):
     res = []
     if isinstance(node, list):
         for child in node:
             res += findNodes(child, func)
     else:
         if func(node):
             res.append(node)
         else:
             res += findNodes(node.get('children', []), func)
     return res;
 
 def toCodeMap(node):
     l = []
     if isinstance(node, list):
         for child in node:
             l += toCodeMap(child)
     else:
         if 'code' in node:
             l.append((node['code'], node['name']))
         l += toCodeMap(node.get('children', []))
     return l
 
 
 parser = argparse.ArgumentParser(description='Download and filter WHO ICD-11 data.')
 parser.add_argument('--output', type=str, required=True, help='Path to which the output should be written to')
 arguments = parser.parse_args()
 
 # download WHO ICD-11 archive
 icd11archive = 'simpletabulation.zip'
 req = requests.get('https://icd.who.int/browse11/Downloads/Download?fileName=simpletabulation.zip')
 writeToFile(icd11archive, req.content)
 
 # unpack the xls file in there
 zipFile = zipfile.ZipFile(icd11archive, 'r')
 icd11xls = 'simpletabulation.xlsx'
 writeToFile(icd11xls, zipFile.read(icd11xls))
 
 # convert xls to csv
 icd11csv = 'simpletabulation.csv'
 os.system(f"libreoffice --headless --convert-to csv --infilter=CSV:44,34,76 {icd11xls}")
 with open(icd11csv, newline='') as f:
     icd11table = list(csv.reader(f, delimiter=',', quotechar='"'))[1:]
 
 # load ICD-11 tree
 root = { 'children': [], 'kind': 'root', 'name': '' }
 stack = [root]
 for row in icd11table:
     kind = row[5];
     depth = countDepth(row[4])
 
     node = {}
     node['kind'] = kind
     node['name'] = row[4].lstrip('- ')
     if kind == 'category':
         node['code'] = row[2]
     elif kind == 'block':
         node['block'] = row[3]
     elif kind == 'chapter':
         node['chapter'] = row[8]
     else:
         continue
     if row[10] != 'True':
         node['children'] = []
 
     while len(stack) >= depth:
         stack.pop()
     parent = stack[-1]
     parent['children'].append(node)
     stack.append(node)
 
 # extract relevant disease codes
 # also: block == Virus?
 diseases = findNodes(root, lambda n: n['kind'] == 'chapter' and (n['chapter'] == '01' or n['chapter'] == '25'))
 diseaseCodes = toCodeMap(diseases)
 diseaseCodes = list(filter(lambda entry: len(entry[0]) == 4, diseaseCodes)) # drop sub-categories
 diseaseCodes.sort(key=lambda entry: entry[0])
 print("Diseases: ", len(diseaseCodes))
 writeToFile(os.path.join(arguments.output, 'diseases.json'), json.dumps(dict(diseaseCodes)).encode('utf-8'))
 
 # extract relevant medication codes
 vaccines = findNodes(root, lambda n: n['kind'] == 'block' and n['name'] == 'Vaccines')
 vaccineCodes = toCodeMap(vaccines)
 vaccineCodes.sort(key=lambda entry: entry[0])
 print("Vaccines:", len(vaccineCodes))
 writeToFile(os.path.join(arguments.output, 'vaccines.json'), json.dumps(dict(vaccineCodes)).encode('utf-8'))