def synyi_id_icd10(file_name, code_file):
if code_file == '去重':
file = codecs.open(path + '\内部质量同步表\\' + file_name +
'\synyi_code匹配去重text_code.csv',
'r',
encoding='utf-8')
elif code_file == '不去重':
file = codecs.open(path + '\内部质量同步表\\' + file_name +
'\synyi_code匹配不去重text_code.csv',
'r',
encoding='utf-8')
dic = get_standard_ICD("系统对照表")
dic_list = []
for i in dic.values():
dic_list.extend(i)
if code_file == '去重':
file_out = codecs.open(path + '\内部质量同步表\\' + file_name +
'\synyi_code_icd_去重.csv',
'w',
encoding='utf-8')
elif code_file == '不去重':
file_out = codecs.open(path + '\内部质量同步表\\' + file_name +
'\synyi_code_icd_不去重.csv',
'w',
encoding='utf-8')
for line in file:
line = line.strip()
if line.split(',')[0] in dic_list:
file_out.write(line + '\n')
file.close()
file_out.close()
def true_false_diag_code(file_name, code_file):
if code_file == '去重':
file = codecs.open(path + '\内部质量同步表\\' + file_name +
'\synyi_code_icd_去重.csv',
'r',
encoding='utf-8')
elif code_file == '不去重':
file = codecs.open(path + '\内部质量同步表\\' + file_name +
'\synyi_code_icd_不去重.csv',
'r',
encoding='utf-8')
if code_file == '去重':
file_out_true = codecs.open(path + '\内部质量同步表\\' + file_name +
'\去重synyi_icd表中code和诊断描述匹配对的实例.csv',
'w',
encoding='utf-8')
file_out_false = codecs.open(path + '\内部质量同步表\\' + file_name +
'\去重synyi_icd表中code和诊断描述匹配错误的实例.csv',
'w',
encoding='utf-8')
elif code_file == '不去重':
file_out_true = codecs.open(path + '\内部质量同步表\\' + file_name +
'\不去重synyi_icd表中code和诊断描述匹配对的实例.csv',
'w',
encoding='utf-8')
file_out_false = codecs.open(path + '\内部质量同步表\\' + file_name +
'\不去重synyi_icd表中code和诊断描述匹配错误的实例.csv',
'w',
encoding='utf-8')
dic = get_standard_ICD("系统对照表")
for line in file:
line = line.strip()
cause = line.split(',')[0]
icd = line.split(',')[1]
if cause in dic[icd]:
file_out_true.write(line + '\n')
else:
file_out_false.write(line + '\n')
file.close()
file_out_true.close()
file_out_false.close()
def sample_code(file_name):
file = pd.read_csv(path + '\内部质量同步表\\' + file_name +
'\只有code的数据没有包含在去重数据中的数量.csv',
encoding='utf-8')
code = file['诊断编码'].tolist()
file_out = codecs.open(path + '\内部质量同步表\\' + file_name +
'\sample\抽样只有诊断编码.csv',
'w',
encoding='utf-8')
dic = get_standard_ICD("系统对照表")
for i in code:
if i.upper() in dic.keys():
for v in dic[i.upper()]:
standard_synyi = post_new(v)
else:
standard_synyi = '[]'
file_out.write(i + ',' + str(post_name(standard_synyi)) + '\n')
def nlp_code(file_name):
# *********************************跑不一致的数据******************************************
# file_in = pd.read_csv(path + '\编码匹配\\' + file_name + '\编码匹配.csv', encoding='utf-8')
# cause = file_in['数据集中诊断名称'].tolist()
# icd = file_in['数据集中诊断编码'].tolist()
# file = codecs.open(path + '\编码匹配\\' + file_name + '\\nlp\使用nlp算法生成编码.csv', 'w', encoding='utf-8')
# file = codecs.open(path + '\用旧算法跑全部数据\\' + file_name + '\\使用nlp算法生成编码.csv', 'w', encoding='utf-8')
# file.write('诊断名称,诊断编码,算法结果' + '\n')
# for index in range(len(cause)):
# code = post(cause[index])
# file.write(cause[index] + ',' + icd[index] + ',' + code + '\n')
# ***********************************跑全体数据************************************************
cause_original, icd_original, _ = get_CDC_ICD(file_name)
icd = []
cause = []
for index in range(len(cause_original)):
if cause_original[index] != '^' and icd_original[index] != '^':
cause.append(cause_original[index])
icd.append(icd_original[index])
num_code = len(cause)
# 随机抽样100条
random_list = random.sample(range(0, len(cause)), 100)
dic = get_standard_ICD("系统对照表")
file = codecs.open(path + '\用旧算法跑全部数据\\' + file_name + '\\使用nlp算法生成编码.csv', 'w', encoding='utf-8')
file.write('诊断名称,诊断编码,诊断编码对应的标准诊断名称,算法结果,算法结果对应的标准诊断名称' + '\n')
for item in random_list:
code = post(cause[item])
if len(code) == 7 and icd[item] in dic.keys() and code[0:5] in dic.keys():
file.write(cause[item] + ',' + icd[item] + ',' + str(dic[icd[item]]) + ',' + code + ',' + str(dic[code[0:5]]) + '\n')
elif len(code) == 7 and icd[item] in dic.keys() and code[0:5] not in dic.keys():
file.write(cause[item] + ',' + icd[item] + ',' + str(dic[icd[item]]) + ',' + code + ',,' + '\n')
elif code == '0' and icd[item] in dic.keys():
file.write(cause[item] + ',' + icd[item] + ',' + str(dic[icd[item]]) + ',' + code + ',,' + '\n')
else:
file.write(cause[item] + ',' + icd[item] + ',' + str(dic[icd[item]]) + ',' + code + ',' + str(dic[code]) + '\n')
def synyi_id(file_name, code_file):
# cause_original, icd_original, _ = get_CDC_ICD(file_name)
# icd = []
# cause = []
# for index in range(len(cause_original)):
# if cause_original[index] != '^' and icd_original[index] != '^':
# # 按照诊断名称去重的
# if cause_original[index] not in cause:
# cause.append(cause_original[index])
# icd.append(icd_original[index])
# print(len(cause), len(icd))
table = pd.read_csv(path + '\内部质量同步表\\' + file_name + '\\' + code_file +
'.csv',
encoding='utf-8')
cause = table['CDC诊断名称'].tolist()
icd = table['CDC诊断编码'].tolist()
dic = get_standard_ICD("系统对照表")
file = codecs.open(path + '\内部质量同步表\\' + file_name + '\\synyi_code匹配' +
code_file + '.csv',
'w',
encoding='utf-8')
file.write(
'CDC中诊断名称,CDC中诊断编码,CDC诊断名称对应的synyi_code,CDC诊断编码对应的标准诊断名称对应的synyi_code'
+ '\n')
for index in range(len(cause)):
cdc_synyi = post_new(cause[index])
if icd[index].upper() in dic.keys():
for i in dic[icd[index].upper()]:
standard_synyi = post_new(i)
else:
standard_synyi = '[]'
file.write(cause[index] + ',' + icd[index] + ',' + cdc_synyi + ',' +
standard_synyi + '\n')
import os
import codecs
from read_file import get_standard_ICD
path = os.getcwd()
file = pd.read_csv(path + '\编码不同.csv', encoding='utf-8')
file = file.drop_duplicates(subset=['诊断名称', '诊断编码', '算法结果'], keep='first')
file_out = codecs.open(path + '\编码不同的数据对照表V2.csv', 'w', encoding='utf-8')
cause = file['诊断名称'].tolist()
code_cdc = file['诊断编码'].tolist()
code_nlp = file['算法结果'].tolist()
dic = get_standard_ICD('系统对照表')
# file_out.write('CDC诊断名称,CDC诊断编码,CDC诊断编码对应的标准诊断名称,算法结果,算法结果对应的标准诊断名称' + '\n')
# for index in range(len(cause)):
# if len(code_nlp[index]) == 7 and code_cdc[index] in dic.keys() and code_nlp[index][0:5] in dic.keys():
# file_out.write(cause[index] + ',' + code_cdc[index] + ',' + str(dic[code_cdc[index]]) + ',' + code_nlp[index] + ',' + str(dic[code_nlp[index][0:5]]) + '\n')
# elif len(code_nlp[index]) == 7 and code_cdc[index] in dic.keys() and code_nlp[index][0:5] not in dic.keys():
# file_out.write(cause[index] + ',' + code_cdc[index] + ',' + str(dic[code_cdc[index]]) + ',' + code_nlp[index] + ',' + '\n')
# elif code_cdc[index] in dic.keys() and code_nlp[index] not in dic.keys() and len(code_nlp[index]) != 7:
# file_out.write(cause[index] + ',' + code_cdc[index] + ',' + str(dic[code_cdc[index]]) + ',' + code_nlp[index] + ',' + '\n')
# elif code_cdc[index] not in dic.keys() and code_nlp[index] in dic.keys():
# file_out.write(cause[index] + ',' + code_cdc[index] + ',' + ',' + code_nlp[index] + ',' + str(dic[code_nlp[index]]) + '\n')
# elif code_cdc[index] not in dic.keys() and code_nlp[index] not in dic.keys():
# file_out.write(cause[index] + ',' + code_cdc[index] + ',' + ',' + code_nlp[index] + ',' + '\n')
# else:
# file_out.write(cause[index] + ',' + code_cdc[index] + ',' + str(dic[code_cdc[index]]) + ',' + code_nlp[index] + ',' + str(dic[code_nlp[index]]) + '\n')