first commit
This commit is contained in:
liligeng111
parent
abc6526143
commit
26631f9b48
24 changed files with 1089 additions and 1 deletions
184
Code/4-ANOVA.py
Normal file
184
Code/4-ANOVA.py
Normal file
|
|
@ -0,0 +1,184 @@
|
|||
import openpyxl
|
||||
import csv
|
||||
import os.path
|
||||
|
||||
|
||||
def read_value(patient_dir, atlas, data_type):
|
||||
folder_name = os.path.basename(patient_dir)
|
||||
csv_file_path = os.path.join(patient_dir, f"{data_type}_{atlas}_result", "result_atlas.csv")
|
||||
|
||||
result = [folder_name]
|
||||
with open(csv_file_path, mode='r') as data_csv_file:
|
||||
csv_reader = csv.DictReader(data_csv_file)
|
||||
for row in csv_reader:
|
||||
mean = float(row["meanValue"])
|
||||
result.append(mean)
|
||||
|
||||
return result
|
||||
|
||||
|
||||
def read_header(patient_dir, atlas, data_type):
|
||||
header = ['编号']
|
||||
|
||||
csv_file_path = os.path.join(patient_dir, f"{data_type}_{atlas}_result", "result_atlas.csv")
|
||||
|
||||
with open(csv_file_path, mode='r') as data_csv_file:
|
||||
csv_reader = csv.DictReader(data_csv_file)
|
||||
for row in csv_reader:
|
||||
region_name = row["Chinese Name"]
|
||||
header.append(region_name)
|
||||
|
||||
return header
|
||||
|
||||
|
||||
def generate_stats_row(name, region_size, function, patient_size, extra_parameter):
|
||||
result = [name]
|
||||
|
||||
for i in range(region_size):
|
||||
letter = openpyxl.utils.cell.get_column_letter(i + 2)
|
||||
result.append(f'={function}({letter}2:{letter}{patient_size + 1}{extra_parameter})')
|
||||
|
||||
return result
|
||||
|
||||
|
||||
def write_to_work_sheet(work_sheet, patient_list, atlas, data_type):
|
||||
if len(patient_list) == 0:
|
||||
return
|
||||
|
||||
header = read_header(patient_list[0], atlas, data_type)
|
||||
|
||||
work_sheet.append(header)
|
||||
|
||||
for patient_dir in patient_list:
|
||||
print(data_type, patient_dir)
|
||||
patient_data = read_value(patient_dir, atlas, data_type)
|
||||
|
||||
work_sheet.append(patient_data)
|
||||
|
||||
region_size = len(header) - 1
|
||||
work_sheet.append([])
|
||||
work_sheet.append([])
|
||||
work_sheet.append(generate_stats_row('均值', region_size, 'AVERAGE', len(patient_list), ''))
|
||||
work_sheet.append(generate_stats_row('标准差', region_size, 'STDEV', len(patient_list), ''))
|
||||
work_sheet.append(generate_stats_row('最小值', region_size, 'MIN', len(patient_list), ''))
|
||||
work_sheet.append(generate_stats_row('5%', region_size, 'PERCENTILE', len(patient_list), ',0.05'))
|
||||
work_sheet.append(generate_stats_row('25%', region_size, 'PERCENTILE', len(patient_list), ',0.25'))
|
||||
work_sheet.append(generate_stats_row('50%', region_size, 'PERCENTILE', len(patient_list), ',0.50'))
|
||||
work_sheet.append(generate_stats_row('75%', region_size, 'PERCENTILE', len(patient_list), ',0.75'))
|
||||
work_sheet.append(generate_stats_row('95%', region_size, 'PERCENTILE', len(patient_list), ',0.95'))
|
||||
work_sheet.append(generate_stats_row('最大值', region_size, 'MAX', len(patient_list), ''))
|
||||
|
||||
|
||||
def write_sum(start_index, step, count, row_num):
|
||||
result = f"={openpyxl.utils.cell.get_column_letter(start_index)}{row_num}"
|
||||
for i in range(1, count):
|
||||
letter = openpyxl.utils.cell.get_column_letter(start_index + i * step)
|
||||
result += f'+{letter}{row_num}'
|
||||
return result
|
||||
|
||||
|
||||
def write_anova(workbook, work_sheet, groups, atlas, data_type):
|
||||
header_group = [""] * 14
|
||||
|
||||
for group_name in groups.keys():
|
||||
header_group += [group_name] * 5
|
||||
work_sheet.append(header_group)
|
||||
|
||||
header_cn = ['', '样本数量', '平均值', '标准差', '总平方和', '组间平方和', '组内平方和', '检验', '组间均方',
|
||||
'自由度', '组内均方', '自由度', '', '']
|
||||
for _ in groups.keys():
|
||||
header_cn += ['样本数量', '平均值', '标准差', '组间平方和', '组内平方和']
|
||||
work_sheet.append(header_cn)
|
||||
|
||||
header_en = ['', 'N', 'AVG', 'STD', 'TSS', 'BSS', 'WSS', 'check', 'BMSS', 'df', 'WMSS', 'df', 'F', 'p']
|
||||
for _ in groups.keys():
|
||||
header_en += ['N', 'AVG', 'STD', 'BSS', 'WSS']
|
||||
work_sheet.append(header_en)
|
||||
|
||||
sheet_dict = {}
|
||||
all_group_count = 0
|
||||
for group_name, group in groups.items():
|
||||
sheet_dict[group_name] = f'{group_name}_{atlas}_{data_type}'
|
||||
all_group_count += len(group)
|
||||
all_group_sheet_name = f'ALL_{atlas}_{data_type}'
|
||||
|
||||
first_group_sheet = workbook[list(sheet_dict.values())[0]]
|
||||
for column_num in range(2, first_group_sheet.max_column + 1):
|
||||
column_letter = openpyxl.utils.cell.get_column_letter(column_num)
|
||||
curr_row_number = column_num + 2
|
||||
data_row = [first_group_sheet.cell(row=1, column=column_num).value]
|
||||
data_row += [write_sum(15, 5, len(sheet_dict), curr_row_number)]
|
||||
data_row += [f'=AVERAGE(\'{all_group_sheet_name}\'!{column_letter}2:{column_letter}{all_group_count + 1})']
|
||||
data_row += [f'=STDEV(\'{all_group_sheet_name}\'!{column_letter}2:{column_letter}{all_group_count + 1})']
|
||||
data_row += [f'=DEVSQ(\'{all_group_sheet_name}\'!{column_letter}2:{column_letter}{all_group_count + 1})']
|
||||
data_row += [write_sum(18, 5, len(sheet_dict), curr_row_number)]
|
||||
data_row += [write_sum(19, 5, len(sheet_dict), curr_row_number)]
|
||||
data_row += [f'=E{curr_row_number}-F{curr_row_number}-G{curr_row_number}']
|
||||
data_row += [f'=F{curr_row_number}/J{curr_row_number}']
|
||||
data_row += [len(groups) - 1]
|
||||
data_row += [f'=G{curr_row_number}/L{curr_row_number}']
|
||||
data_row += [f'=B{curr_row_number}-J{curr_row_number}-1']
|
||||
data_row += [f'=I{curr_row_number}/K{curr_row_number}']
|
||||
data_row += [f'=FDIST(M{curr_row_number}, J{curr_row_number}, L{curr_row_number})']
|
||||
|
||||
for group_name, sheet_name in sheet_dict.items():
|
||||
data_count = len(groups[group_name])
|
||||
current_column_num = len(data_row) + 1
|
||||
count_letter = openpyxl.utils.cell.get_column_letter(current_column_num)
|
||||
avg_letter = openpyxl.utils.cell.get_column_letter(current_column_num + 1)
|
||||
data_row += [f'=COUNTA(\'{sheet_name}\'!{column_letter}2:{column_letter}{data_count + 1})']
|
||||
data_row += [f'=AVERAGE(\'{sheet_name}\'!{column_letter}2:{column_letter}{data_count + 1})']
|
||||
data_row += [f'=STDEV(\'{sheet_name}\'!{column_letter}2:{column_letter}{data_count + 1})']
|
||||
data_row += [f'={count_letter}{curr_row_number}*({avg_letter}{curr_row_number}-C{curr_row_number})^2']
|
||||
data_row += [f'=DEVSQ(\'{sheet_name}\'!{column_letter}2:{column_letter}{data_count + 1})']
|
||||
|
||||
work_sheet.append(data_row)
|
||||
|
||||
|
||||
def main():
|
||||
data_types = ['corr-CBF']
|
||||
atlas_list = ['AnImage_AAL3']
|
||||
|
||||
output_file = r'..\Data\csv-ANOVA.xlsx'
|
||||
group_file = r'..\Data\group.xlsx'
|
||||
patient_root = r'..\Data\csv-data'
|
||||
|
||||
groups = {}
|
||||
|
||||
group_workbook = openpyxl.load_workbook(group_file)
|
||||
for sheet_name in group_workbook.sheetnames:
|
||||
group = []
|
||||
group_sheet = group_workbook[sheet_name]
|
||||
for i in range(1, group_sheet.max_row + 1):
|
||||
patient_id = group_sheet.cell(row=i, column=1).value
|
||||
patient_path = os.path.join(patient_root, patient_id)
|
||||
group.append(patient_path)
|
||||
groups[sheet_name] = group
|
||||
|
||||
all_group = []
|
||||
for group in groups.values():
|
||||
all_group += group
|
||||
|
||||
work_book = openpyxl.Workbook()
|
||||
work_book.remove(work_book.active)
|
||||
|
||||
for data_type in data_types:
|
||||
for atlas in atlas_list:
|
||||
atlas_new_name = atlas.replace("AnImage_", "")
|
||||
for group_key in groups.keys():
|
||||
work_sheet = work_book.create_sheet(title=f'{group_key}_{atlas_new_name}_{data_type}')
|
||||
write_to_work_sheet(work_sheet, groups[group_key], atlas, data_type)
|
||||
|
||||
work_sheet = work_book.create_sheet(title=f'ALL_{atlas_new_name}_{data_type}')
|
||||
write_to_work_sheet(work_sheet, all_group, atlas, data_type)
|
||||
|
||||
for data_type in data_types:
|
||||
for atlas in atlas_list:
|
||||
atlas_new_name = atlas.replace("AnImage_", "")
|
||||
work_sheet = work_book.create_sheet(title=f'ANOVA_{atlas_new_name}_{data_type}')
|
||||
write_anova(work_book, work_sheet, groups, atlas_new_name, data_type)
|
||||
|
||||
work_book.save(output_file)
|
||||
|
||||
|
||||
main()
|
||||
Loading…
Add table
Add a link
Reference in a new issue