2023年研究生数学建模竞赛E题第一问代码.zip

Python 3.11.5 (tags/v3.11.5:cce6ba9, Aug 24 2023, 14:38:34) [MSC v.1936 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license()" for more information. >>> import pandas as pd ... import numpy as np ... import matplotlib.pyplot as plt ... import seaborn as sns ... import warnings ... ... # 忽略未来警告 ... warnings.simplefilter(action='ignore', category=FutureWarning) ... ... # 设置字体为’SimHei‘，以显示中文标签 ... plt.rcParams['font.sans-serif'] = ['SimHei'] ... plt.rcParams['axes.unicode_minus'] = False ... ... # 导入给定的数据 ... df1 = pd.read_excel('表1-患者列表及临床信息.xlsx') ... df2 = pd.read_excel('表2-患者影像信息血肿及水肿的体积及位置.xlsx') ... df3 = pd.read_excel('附表1-检索表格-流水号vs时间.xlsx') ... ... # 将附表进行转化 ... ls1 = [] ... ls2 = [] ... ls3 = [] ... ls4 = [] ... ls5 = [] ... ls6 = [] ... ... for value in df3.values: ... ID = value[0] ... freq = value[1] ... for f in range(freq): ... ls1.append(ID) ... ls2.append(freq) ... ls3.append(value[f*2+2]) ... ls4.append(value[f*2+3]) ... ls5.append(f'第{f}次检查') ... ls6.append(value[3]) ... df3 = pd.DataFrame({'ID': ls1, 'first_code': ls6, '重复次数': ls2, 'code': ls4, 'time': ls3, '检查次数': ls5}) df3['code'] = df3['code'].astype('int64') # 调整df1中的入院首次影像检查流水号 ls = [] for i in df1['入院首次影像检查流水号'].values: code = df3[df3.code == i]['first_code'].values[0] ls.append(code) n = 0 for i, o in zip(df1['入院首次影像检查流水号'].values, ls): if i != o: n += 1 df1['入院首次影像检查流水号'] = ls print(n) # 将df2转换为易读的dataframe ls1 = [] ls2 = [] ls3 = [] for value in df2.values: ID = value[0] code = value[1] freq = df3[df3.first_code == value[1]]['重复次数'].values[0] for f in range(min(freq, 9)): ls1.append(ID) ls2.append(code) ls3.append(f'第{f}次检查') tem = pd.DataFrame({'ID': ls1, 'first_code': ls2, '检查次数': ls3}) # 添加时间代码 time_code = [] for value in df2.values: freq = df3[df3.first_code == value[1]]['重复次数'].values[0] for f in range(min(freq, 9)): time_code.append(value[23*f + 1]) tem['code'] = time_code # 逐列添加其它特征 columns = df2.columns for delta in range(22): ls = [] for value in df2.values: freq = df3[df3.first_code == value[1]]['重复次数'].values[0] for f in range(min(freq, 9)): ls.append(value[23*f + 2 + delta]) tem[columns[2 + delta]] = ls df2 = tem # 合并df2和df3的特定列 df2 = pd.merge(df3[['code', 'time']], df2, on='code') # 将df2中的HM_volume由10^-3ml转换成ml单位 df2['HM_volume'] = df2['HM_volume'] / 1000 # 计算每个患者是否发生血肿扩张及发生时间 ls1 = [] ls2 = [] ls3 = [] for code, ID in zip(df1['入院首次影像检查流水号'].values, df1['Unnamed: 0'].values): delta = df1[df1.入院首次影像检查流水号 == code]['发病到首次影像检查时间间隔'].values[0] first_code_time = df3[df3.code == code]['time'].values[0] error_time = first_code_time - np.timedelta64(int(delta * 60), 'm') max_time = error_time + np.timedelta64(int(48 * 60), 'm') origin_value = df2[df2.code == code].values[0] compare_value = df2[(df2.first_code == code) & (df2.time < max_time)][1:].values condition = 0 for value in compare_value: if value[5] - origin_value[5] >= 6 or (value[5] - origin_value[5]) / value[5] >= 0.33: condition = 1 time = str(value[1] - origin_value[1]).split(' ')[2] hours = int(time.split(':')[0]) minute = int(time.split(':')[1]) time_delta = hours + minute / 60 ls1.append(condition) if condition == 0: ls2.append(0) else: ls2.append(round(time_delta, 2)) ls3.append(ID) result = pd.DataFrame({'ID': ls3, '是否扩张': ls1, '发生时间': ls2}) # 保存结果到Excel文件 result.to_excel('1a结果.xlsx', index=False) # 绘制箱线图显示血肿扩张发生时间 plt.figure(figsize=(10, 6)) sns.boxplot(x='是否扩张', y='发生时间', data=result) plt.xlabel('是否发生血肿扩张') plt.ylabel('血肿扩张发生时间（小时）') plt.title('血肿扩张发生时间分布') plt.show() # 绘制条形图显示各患者血肿扩张发生时间 plt.figure(figsize=(10, 6)) plt.bar(result['ID'], result['发生时间']) plt.xlabel('患者编号') plt.ylabel('血肿扩张发生时间（小时）') plt.title('各患者血肿扩张发生时间') plt.show()