From 26631f9b4860ba2d2d05b6df7c46fb3dabaaa980 Mon Sep 17 00:00:00 2001 From: liligeng111 Date: Wed, 6 Aug 2025 02:04:35 +0800 Subject: [PATCH 1/2] first commit --- .gitignore | 4 + Code/1-extract.py | 72 +++++++++++++++ Code/10-registration.py | 53 +++++++++++ Code/2-simple-stat.py | 93 ++++++++++++++++++++ Code/3-group.py | 104 ++++++++++++++++++++++ Code/4-ANOVA.py | 184 +++++++++++++++++++++++++++++++++++++++ Code/5-ANOVA-SciPy.py | 150 +++++++++++++++++++++++++++++++ Code/6-cbf-graph.py | 68 +++++++++++++++ Code/7-box-graph.py | 50 +++++++++++ Code/8-subplot.py | 55 ++++++++++++ Code/9-polar-graph.py | 71 +++++++++++++++ Exercise/exercise-1.1.py | 13 +++ Exercise/exercise-1.2.py | 8 ++ Exercise/exercise-2.1.py | 9 ++ Exercise/exercise-2.2.py | 18 ++++ Exercise/exercise-2.3.py | 19 ++++ Exercise/exercise-3.1.py | 13 +++ Exercise/exercise-4.1.py | 33 +++++++ Exercise/exercise-4.2.py | 15 ++++ Exercise/exercise-4.3.py | 14 +++ Exercise/exercise-5.1.py | 10 +++ Exercise/exercise-5.2.py | 23 +++++ Exercise/exercise-5.3.py | 8 ++ README.md | 3 +- 24 files changed, 1089 insertions(+), 1 deletion(-) create mode 100644 Code/1-extract.py create mode 100644 Code/10-registration.py create mode 100644 Code/2-simple-stat.py create mode 100644 Code/3-group.py create mode 100644 Code/4-ANOVA.py create mode 100644 Code/5-ANOVA-SciPy.py create mode 100644 Code/6-cbf-graph.py create mode 100644 Code/7-box-graph.py create mode 100644 Code/8-subplot.py create mode 100644 Code/9-polar-graph.py create mode 100644 Exercise/exercise-1.1.py create mode 100644 Exercise/exercise-1.2.py create mode 100644 Exercise/exercise-2.1.py create mode 100644 Exercise/exercise-2.2.py create mode 100644 Exercise/exercise-2.3.py create mode 100644 Exercise/exercise-3.1.py create mode 100644 Exercise/exercise-4.1.py create mode 100644 Exercise/exercise-4.2.py create mode 100644 Exercise/exercise-4.3.py create mode 100644 Exercise/exercise-5.1.py create mode 100644 Exercise/exercise-5.2.py create mode 100644 Exercise/exercise-5.3.py diff --git a/.gitignore b/.gitignore index ab3e8ce..185ae7e 100644 --- a/.gitignore +++ b/.gitignore @@ -162,3 +162,7 @@ cython_debug/ # option (not recommended) you can uncomment the following to ignore the entire idea folder. #.idea/ +.venv/ +.idea/ +.ipynb_checkpoints/ +Data/ \ No newline at end of file diff --git a/Code/1-extract.py b/Code/1-extract.py new file mode 100644 index 0000000..fad1dd9 --- /dev/null +++ b/Code/1-extract.py @@ -0,0 +1,72 @@ +import openpyxl +import csv +import os.path +import glob + + +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 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, atlas, patient_dir) + patient_data = read_value(patient_dir, atlas, data_type) + + work_sheet.append(patient_data) + + +def main(): + # 选择导出的数据类型，请尝试增加VOL类型，并修改read_value、read_header函数以支持读取体积 + data_types = ['corr-CBF', 'ATT', 'ACBV', 'CBF1', 'CBF2', 'CBF3', 'CBF4', 'CBF5'] + # 选择导出后的图谱 + atlas_list = ['AnImage_WholeBrain', 'AnImage_BrainLobes', 'AnImage_AAL3'] + # 选择输出的位置 + output_file = r'..\Data\csv-extract.xlsx' + # 选择输入的数据 + patient_list = glob.glob(r'..\Data\csv-data\*') + # 创建新的excel workbook + work_book = openpyxl.Workbook() + # 删除默认worksheet + work_book.remove(work_book.active) + + for data_type in data_types: + for atlas in atlas_list: + work_sheet = work_book.create_sheet(title=f'{atlas}_{data_type}') + write_to_work_sheet(work_sheet, patient_list, atlas, data_type) + + # 保存workbook到xlsx文件 + work_book.save(output_file) + + +main() diff --git a/Code/10-registration.py b/Code/10-registration.py new file mode 100644 index 0000000..0abfd62 --- /dev/null +++ b/Code/10-registration.py @@ -0,0 +1,53 @@ +import SimpleITK as sitk + + +def command_iteration(method): + print(f"{method.GetOptimizerIteration()} = {method.GetMetricValue():.5f} {method.GetOptimizerPosition()}") + + +def registration(fixed_image_path, moving_image_path, new_image_path): + fixed_image = sitk.ReadImage(fixed_image_path, sitk.sitkFloat32) + moving_image = sitk.ReadImage(moving_image_path, sitk.sitkFloat32) + + registration_method = sitk.ImageRegistrationMethod() + + registration_method.SetMetricAsJointHistogramMutualInformation(numberOfHistogramBins=50) + registration_method.SetMetricSamplingStrategy(registration_method.REGULAR) + registration_method.SetMetricSamplingPercentage(0.2, 42) + registration_method.SetGlobalDefaultNumberOfThreads(1) + registration_method.SetOptimizerScalesFromPhysicalShift() + + registration_method.SetInterpolator(sitk.sitkLinear) + registration_method.AddCommand(sitk.sitkIterationEvent, lambda: command_iteration(registration_method)) + + registration_method.SetOptimizerAsRegularStepGradientDescent(minStep=0.0001, + learningRate=1.0, + numberOfIterations=400, + gradientMagnitudeTolerance=1e-8) + + initial_transform = sitk.CenteredTransformInitializer(fixed_image, moving_image, sitk.Euler3DTransform()) + registration_method.SetInitialTransform(initial_transform, inPlace=True) + + # 进行配准 + final_transform_v4 = registration_method.Execute(fixed_image, moving_image) + + # 打印结束条件 + stop_condition = registration_method.GetOptimizerStopConditionDescription() + print(stop_condition) + + # 重采样新的图像 + moving_resampled = sitk.Resample(moving_image, fixed_image, final_transform_v4, sitk.sitkLinear, 0.0, + moving_image.GetPixelID()) + + sitk.WriteImage(moving_resampled, new_image_path) + + +def main(): + fixed_image_path = r"..\Data\registration\t1.nii.gz" + moving_image_path = r"..\Data\registration\t2 flair.nii.gz" + new_image_path = r"..\Data\registration\t2 flair_resampled.nii.gz" + + registration(fixed_image_path, moving_image_path, new_image_path) + + +main() diff --git a/Code/2-simple-stat.py b/Code/2-simple-stat.py new file mode 100644 index 0000000..e75abe5 --- /dev/null +++ b/Code/2-simple-stat.py @@ -0,0 +1,93 @@ +import openpyxl +import csv +import os.path +import glob + + +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): + 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})') + + 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))) + # 请尝试在空格的位置添加 5% 25% 50% 75% 95% 分位数，使用excel的PERCENTILE函数 + work_sheet.append([]) + work_sheet.append([]) + work_sheet.append([]) + work_sheet.append([]) + work_sheet.append([]) + work_sheet.append(generate_stats_row('最大值', region_size, 'MAX', len(patient_list))) + + +def main(): + data_types = ['corr-CBF', 'ATT', 'ACBV', 'CBF1', 'CBF2', 'CBF3', 'CBF4', 'CBF5'] + atlas_list = ['AnImage_WholeBrain', 'AnImage_BrainLobes', 'AnImage_AAL3'] + + output_file = r'..\Data\csv-simple-stat.xlsx' + + patient_list = glob.glob(r'..\Data\csv-data\*') + + work_book = openpyxl.Workbook() + work_book.remove(work_book.active) + + for data_type in data_types: + for atlas in atlas_list: + work_sheet = work_book.create_sheet(title=f'{atlas}_{data_type}') + write_to_work_sheet(work_sheet, patient_list, atlas, data_type) + + work_book.save(output_file) + + +main() diff --git a/Code/3-group.py b/Code/3-group.py new file mode 100644 index 0000000..0bdc69c --- /dev/null +++ b/Code/3-group.py @@ -0,0 +1,104 @@ +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 main(): + data_types = ['corr-CBF', "ATT"] + atlas_list = ['AnImage_WholeBrain', 'AnImage_BrainLobes', 'AnImage_AAL3'] + + output_file = r'..\Data\csv-group.xlsx' + group_file = r'..\Data\group.xlsx' + patient_root = r'..\Data\csv-data' + + groups = {} + # 读取分组信息 + 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 + + work_book = openpyxl.Workbook() + work_book.remove(work_book.active) + + for group_key in groups.keys(): + for data_type in data_types: + for atlas in atlas_list: + work_sheet = work_book.create_sheet(title=f'{group_key}_{atlas}_{data_type}') + write_to_work_sheet(work_sheet, groups[group_key], atlas, data_type) + + work_book.save(output_file) + + +main() diff --git a/Code/4-ANOVA.py b/Code/4-ANOVA.py new file mode 100644 index 0000000..075e8db --- /dev/null +++ b/Code/4-ANOVA.py @@ -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() diff --git a/Code/5-ANOVA-SciPy.py b/Code/5-ANOVA-SciPy.py new file mode 100644 index 0000000..cceccad --- /dev/null +++ b/Code/5-ANOVA-SciPy.py @@ -0,0 +1,150 @@ +import numpy as np +import openpyxl +import csv +import os.path +from scipy.stats import f_oneway + + +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_anova(workbook, work_sheet, groups, atlas, data_type): + header = ['', 'F', 'p'] + work_sheet.append(header) + + sheet_dict = {} + for group_name, group in groups.items(): + sheet_dict[group_name] = f'{group_name}_{atlas}_{data_type}' + + first_group_sheet = workbook[list(sheet_dict.values())[0]] + for column_num in range(2, first_group_sheet.max_column + 1): + data_list = [] + data_row = [first_group_sheet.cell(row=1, column=column_num).value] + + for group_name, sheet_name in sheet_dict.items(): + group_data = [] + data_count = len(groups[group_name]) + for i in range(2, data_count + 2): + data_value = workbook[sheet_name].cell(row=i, column=column_num).value + if not np.isnan(data_value): + group_data.append(data_value) + data_list.append(group_data) + + if len(data_list[0]) == 0: + continue + + f_value, p_value = f_oneway(*data_list) + data_row += [f_value, p_value] + work_sheet.append(data_row) + + +def main(): + data_types = ['corr-CBF'] + atlas_list = ['AnImage_AAL3'] + + output_file = r'..\Data\csv-ANOVA-SciPy.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() diff --git a/Code/6-cbf-graph.py b/Code/6-cbf-graph.py new file mode 100644 index 0000000..5013607 --- /dev/null +++ b/Code/6-cbf-graph.py @@ -0,0 +1,68 @@ +import numpy as np +import openpyxl +import os.path +import matplotlib.pyplot as plt + + +def draw_cbf_graph(region, data, save_path): + bar_x = np.array(['0.5', '1', '1.5', '2', '2.5', 'cCBF']) + bar_y = np.array(data) + color = ['#4CAF50', '#4CAF50', '#4CAF50', '#4CAF50', '#4CAF50', '#478FC1'] + + # Plotting the Graph + font_color = '#FFFFFF' + plt.rcParams['text.color'] = font_color + plt.rcParams['axes.labelcolor'] = font_color + plt.rcParams['xtick.color'] = font_color + plt.rcParams['ytick.color'] = font_color + + plt.figure(facecolor='#000000') + ax = plt.axes() + ax.set_facecolor('#000000') + ax.spines['left'].set_color(font_color) + ax.spines['right'].set_color(font_color) + ax.spines['bottom'].set_color(font_color) + ax.spines['top'].set_color(font_color) + + plt.rcParams["font.sans-serif"] = ["SimHei"] # 设置字体 + plt.bar(bar_x, bar_y, width=0.5, color=color) + + for i in range(len(bar_x)): + plt.text(i - 0.2, bar_y[i] - 2, f"{bar_y[i]:.2f}") + + plt.title(f"221例健康人西门子5延迟数据矫正前后{region}CBF对比") + plt.xlabel("PLD") + plt.ylabel("CBF") + plt.savefig(save_path, dpi=200) + plt.cla() + plt.close('all') + + +def main(): + data_types = ['CBF1', 'CBF2', 'CBF3', 'CBF4', 'CBF5', 'corr-CBF'] + # 请尝试添加'AnImage_AAL3'图谱,并修复错误 + atlas_list = ['AnImage_WholeBrain', 'AnImage_BrainLobes'] + output_dir = r'..\Data\cbf-graph' + os.makedirs(output_dir, exist_ok=True) + + input_file = r'..\Data\csv-simple-stat.xlsx' + workbook = openpyxl.load_workbook(input_file, data_only=True) + + for atlas in atlas_list: + cbf_mean_dict = {} + for data_type in data_types: + worksheet = workbook[f'{atlas}_{data_type}'] + for column_num in range(2, worksheet.max_column + 1): + name = worksheet.cell(row=1, column=column_num).value + mean = worksheet.cell(row=worksheet.max_row - 8, column=column_num).value + print(atlas, data_type, name, mean) + + if name not in cbf_mean_dict: + cbf_mean_dict[name] = [] + cbf_mean_dict[name].append(mean) + + for region, data in cbf_mean_dict.items(): + draw_cbf_graph(region, data, os.path.join(output_dir, f'{region}.png')) + + +main() diff --git a/Code/7-box-graph.py b/Code/7-box-graph.py new file mode 100644 index 0000000..313a1f4 --- /dev/null +++ b/Code/7-box-graph.py @@ -0,0 +1,50 @@ +import openpyxl +import os.path +import matplotlib.pyplot as plt + + +def draw_box_graph(region, data, groups, data_type, save_path): + plt.rcParams["font.sans-serif"] = ["SimHei"] # 设置字体 + plt.boxplot(data, + patch_artist=True, # 是否填色 + labels=groups) + + plt.title(f"各个分组{region}{data_type}对比") + plt.xlabel("分组") + plt.ylabel(data_type) + plt.savefig(save_path, dpi=200) + plt.cla() + plt.close('all') + + +def main(): + data_types = ['corr-CBF', "ATT"] + # 请尝试添加'AnImage_AAL3'图谱,并修复错误 + atlas_list = ['AnImage_WholeBrain', 'AnImage_BrainLobes'] + output_dir = r'..\Data\box-graph' + os.makedirs(output_dir, exist_ok=True) + + input_file = r'..\Data\csv-group.xlsx' + workbook = openpyxl.load_workbook(input_file, data_only=True) + groups = ['A', 'B', 'C'] + + for data_type in data_types: + for atlas in atlas_list: + value_dict = {} + for group in groups: + worksheet = workbook[f'{group}_{atlas}_{data_type}'] + for column_num in range(2, worksheet.max_column + 1): + data = [] + name = worksheet.cell(row=1, column=column_num).value + for row in range(2, worksheet.max_row - 10): + data.append(worksheet.cell(row=row, column=column_num).value) + + if name not in value_dict: + value_dict[name] = [] + value_dict[name].append(data) + + for region, data in value_dict.items(): + draw_box_graph(region, data, groups, data_type, os.path.join(output_dir, f'{region}{data_type}.png')) + + +main() diff --git a/Code/8-subplot.py b/Code/8-subplot.py new file mode 100644 index 0000000..03f955a --- /dev/null +++ b/Code/8-subplot.py @@ -0,0 +1,55 @@ +import openpyxl +import os.path +import matplotlib.pyplot as plt + + +def draw_box_graph(data_dict, groups, data_type, save_path): + plt.rcParams["font.sans-serif"] = ["SimHei"] # 设置字体 + + grid_height = 2 + grid_width = 4 + + _, ax = plt.subplots(grid_height, grid_width, figsize=(20, 15), sharey=True) + keys = list(data_dict.keys()) + for x in range(grid_width): + for y in range(grid_height): + region_name = keys[x + y * grid_width] + ax[y][x].boxplot(data_dict[region_name], patch_artist=True, labels=groups) + ax[y][x].set_title(f"{region_name}{data_type}对比") + ax[y][x].set_ylabel(data_type) + + plt.tight_layout() + plt.savefig(save_path, dpi=200) + plt.cla() + plt.close('all') + + +def main(): + data_types = ['corr-CBF', "ATT"] + atlas_list = ['AnImage_BrainLobes'] + output_dir = r'..\Data\subplot' + os.makedirs(output_dir, exist_ok=True) + + input_file = r'..\Data\csv-group.xlsx' + workbook = openpyxl.load_workbook(input_file, data_only=True) + groups = ['A', 'B', 'C'] + + for data_type in data_types: + for atlas in atlas_list: + value_dict = {} + for group in groups: + worksheet = workbook[f'{group}_{atlas}_{data_type}'] + for column_num in range(2, worksheet.max_column + 1): + data = [] + name = worksheet.cell(row=1, column=column_num).value + for row in range(2, worksheet.max_row - 10): + data.append(worksheet.cell(row=row, column=column_num).value) + + if name not in value_dict: + value_dict[name] = [] + value_dict[name].append(data) + + draw_box_graph(value_dict, groups, data_type, os.path.join(output_dir, f'{atlas}_{data_type}.png')) + + +main() diff --git a/Code/9-polar-graph.py b/Code/9-polar-graph.py new file mode 100644 index 0000000..1b8f2a4 --- /dev/null +++ b/Code/9-polar-graph.py @@ -0,0 +1,71 @@ +import numpy as np +import openpyxl +import os.path +import matplotlib.pyplot as plt + + +def draw_cbf_graph(cbf_mean_dict, data_type, save_path): + plt.rcParams["figure.figsize"] = [32, 18] + plt.rcParams["figure.autolayout"] = True + plt.rcParams["font.sans-serif"] = ["SimHei"] # 设置字体 + subplot = plt.subplot(1, 1, 1, projection='polar') + + count = len(cbf_mean_dict) + theta = np.linspace(0.0, 2 * np.pi, count, endpoint=False) + data = np.array(list(cbf_mean_dict.values())) + width = 1.8 * np.pi / count + subplot.set_theta_zero_location('N') + subplot.set_theta_direction(-1) + subplot.xaxis.set_ticks(theta) + subplot.yaxis.set_ticks([]) + + bars = subplot.bar(x=theta, height=data, width=width) + for data_value, bar in zip(data, bars): + bar.set_facecolor(plt.cm.turbo(data_value / 60)) + bar.set_alpha(0.8) + for theta_value, data_value in zip(theta, data): + margin = data.max() * 0.08 + subplot.annotate(str(round(data_value, 2)), xy=(theta_value, data_value + margin)) + + subplot.set_xticklabels(list(cbf_mean_dict.keys())) + for label, angle in zip(subplot.get_xticklabels(), theta): + x, y = label.get_position() + lab = subplot.text(x, y - 0.1, label.get_text(), transform=label.get_transform(), ha=label.get_ha(), + va=label.get_va()) + rotation = - np.rad2deg(angle) + 90 + if rotation < -90: + rotation += 180 + lab.set_rotation(rotation) + subplot.set_xticklabels([]) + + plt.title(f"221例健康人西门子5延迟数据各脑区{data_type}") + plt.savefig(save_path, dpi=200) + plt.cla() + plt.close('all') + + +def main(): + # 请尝试添加att参数，并调整色阶范围使得CBF=40时为红色，60时为蓝色 + data_types = ['corr-CBF'] + atlas_list = ['AnImage_AAL3'] + output_dir = r'..\Data\polar-graph' + os.makedirs(output_dir, exist_ok=True) + + input_file = r'..\Data\csv-simple-stat.xlsx' + workbook = openpyxl.load_workbook(input_file, data_only=True) + + for data_type in data_types: + for atlas in atlas_list: + cbf_mean_dict = {} + worksheet = workbook[f'{atlas}_{data_type}'] + for column_num in range(2, worksheet.max_column + 1): + name = worksheet.cell(row=1, column=column_num).value + mean = worksheet.cell(row=worksheet.max_row - 8, column=column_num).value + print(atlas, data_type, name, mean) + if type(mean) is float: + cbf_mean_dict[name] = mean + + draw_cbf_graph(cbf_mean_dict, data_type, os.path.join(output_dir, f'{atlas}_{data_type}.png')) + + +main() diff --git a/Exercise/exercise-1.1.py b/Exercise/exercise-1.1.py new file mode 100644 index 0000000..4fcde8f --- /dev/null +++ b/Exercise/exercise-1.1.py @@ -0,0 +1,13 @@ + +# 计算 12879 * 36521 并打印 +# product = ??? +# print(product) + + +# 计算 100 模 32 并打印 +# mod = ??? +# print(mod) + +# 打印Hello\ "World!" +# msg = ??? +# print(msg) diff --git a/Exercise/exercise-1.2.py b/Exercise/exercise-1.2.py new file mode 100644 index 0000000..eec36a0 --- /dev/null +++ b/Exercise/exercise-1.2.py @@ -0,0 +1,8 @@ +import math + +# 请输出以下关于圆的周长和面积的计算，保留3位小数 +radius = 3 +pi = math.pi + +# 应当输出 当圆的的半径是3时，它的周长是18.850，他的面积是28.274 +print('当圆的的半径是时，它的周长是，他的面积是') diff --git a/Exercise/exercise-2.1.py b/Exercise/exercise-2.1.py new file mode 100644 index 0000000..7648898 --- /dev/null +++ b/Exercise/exercise-2.1.py @@ -0,0 +1,9 @@ +# 请尝试修改list以满足输出 + +list_a = [0] + +list_a += ['Hello'] * 5 +list_a += [4, 5] * 3 + +# 输出[0, 'Hello', 'Hello', 'Hello', 'Hello', 'Hello', 4, 5, 4, 5, 4, 5] +print(list_a) diff --git a/Exercise/exercise-2.2.py b/Exercise/exercise-2.2.py new file mode 100644 index 0000000..c2bf5ad --- /dev/null +++ b/Exercise/exercise-2.2.py @@ -0,0 +1,18 @@ + +# 我们用如下数组存储一些患者的信息 +patient_info = [ + ["Zhang San", 52, "M"], + ["Li Si", 19, "F"], + ["Wang Wu", 47, "F"], +] + +# 检查后发现，张三的年龄和李四的年龄写反了，请将他们调整过来 + + + + +# 应当为 19 +print(patient_info[0][1]) + +# 应当为 52 +print(patient_info[1][1]) diff --git a/Exercise/exercise-2.3.py b/Exercise/exercise-2.3.py new file mode 100644 index 0000000..87c0ba9 --- /dev/null +++ b/Exercise/exercise-2.3.py @@ -0,0 +1,19 @@ +age_dict = { + "Zhang San": 52, + "Li Si": 19, + "Wang Wu": 47 +} + +# 检查后发现，张三的年龄和李四的年龄写反了，请将他们调整过来。并添加Zhao Liu，年龄为29 + + + + +# 应当为 19 +print(age_dict['Zhang San']) + +# 应当为 52 +print(age_dict['Li Si']) + +# 应当为 29 +print(age_dict['Zhao Liu']) diff --git a/Exercise/exercise-3.1.py b/Exercise/exercise-3.1.py new file mode 100644 index 0000000..5664c63 --- /dev/null +++ b/Exercise/exercise-3.1.py @@ -0,0 +1,13 @@ +# 请根据用户输入的年龄输出在学校的阶段： +# 0-5 输出 学龄前 +# 6-11 输出 小学 +# 12-14 输出 初中 +# 15-17 输出 高中 +# 18-21 输出 大学 +# 22-59 输出 打工人 +# 60+ 输出 退休了 + +print('请输入你的年龄：') +age = int(input()) +if age < 6: + print('学龄前') diff --git a/Exercise/exercise-4.1.py b/Exercise/exercise-4.1.py new file mode 100644 index 0000000..23777f1 --- /dev/null +++ b/Exercise/exercise-4.1.py @@ -0,0 +1,33 @@ + + +# 我们希望将患者按照信息，分别按照性别和年龄分成4个组别，分别存储他们的姓名 +patient_info = [ + ["Zhang San", 52, "M"], + ["Li Si", 19, "F"], + ["Wang Wu", 47, "F"], + ["Zhao Liu", 29, "M"], +] +group_M_under_30 = [] +group_M_over_30 = [] +group_F_under_30 = [] +group_F_over_30 = [] + +for patient in patient_info: + print('new patient:', patient) + # 补充代码，使用条件判断，将患者分为4组，分别存储他们的姓名即可 + + + + + +# 应当为 ["Zhao Liu"] +print(group_M_under_30) + +# 应当为 ["Zhang San"] +print(group_M_over_30) + +# 应当为 ["Li Si"] +print(group_F_under_30) + +# 应当为 ["Wang Wu"] +print(group_F_over_30) diff --git a/Exercise/exercise-4.2.py b/Exercise/exercise-4.2.py new file mode 100644 index 0000000..6b3548b --- /dev/null +++ b/Exercise/exercise-4.2.py @@ -0,0 +1,15 @@ +# 给定一组整数，计算他们的和并输出等式 + +magic_number_list = [4, 8, 15, 16, 23, 42] + +sum = 0 +expression = '' +for index in range(len(magic_number_list)): + print(index) + + + + + +# 应输出4 + 8 + 15 + 16 + 23 + 42 = 108 +print(expression) \ No newline at end of file diff --git a/Exercise/exercise-4.3.py b/Exercise/exercise-4.3.py new file mode 100644 index 0000000..1255156 --- /dev/null +++ b/Exercise/exercise-4.3.py @@ -0,0 +1,14 @@ +import random + +# 我们来写一个猜数字的小游戏 + +# 生成一个随机的1-100整数作为答案 +answer = random.randint(1, 100) +game_over = False +count = 0 +while not game_over: + print(f"你已经猜了{count}次，请输入下一次猜测") + count += 1 + guess = int(input()) + print(f"你的猜测是{guess}") + # 如果猜对了，输出“猜对了”并结束游戏。否则输出“猜大了”或者“猜小了” \ No newline at end of file diff --git a/Exercise/exercise-5.1.py b/Exercise/exercise-5.1.py new file mode 100644 index 0000000..2894ecf --- /dev/null +++ b/Exercise/exercise-5.1.py @@ -0,0 +1,10 @@ +# 我们来输出100以内的素数 + +def is_prime(n): + # 检查n是否是素数，一个简易的办法是，遍历小于n的自然数，判断是否有自然数可以整除n + return False + + +for i in range(100): + # 遍历100以内的自然数，检查他们是不是素数，并输出其中的素数 + result = is_prime(i) diff --git a/Exercise/exercise-5.2.py b/Exercise/exercise-5.2.py new file mode 100644 index 0000000..955de4f --- /dev/null +++ b/Exercise/exercise-5.2.py @@ -0,0 +1,23 @@ +import random + + +# 概率计算表明，23个人中有至少两人生日相同的概率高达50.7%，听起来有些反直觉，我们来通过计算估计这一概率。 +def contains_same_birthday(birthdays): + # 修改代码，判断birthdays这个list中是否有相同的元素 + return True + + +# 我们进行10000次测试 +test_count = 10000 +# 有至少两人生日相同的次数 +true_count = 0 + +for i in range(test_count): + birth_days = [] + for j in range(23): + birth_days.append(random.randint(1, 365)) + + # 统计满足条件的测试次数 + +# 输出测试结果，保留4位小数 +print('23个人中有至少两人生日相同的概率是:') diff --git a/Exercise/exercise-5.3.py b/Exercise/exercise-5.3.py new file mode 100644 index 0000000..06f1fde --- /dev/null +++ b/Exercise/exercise-5.3.py @@ -0,0 +1,8 @@ + + +# 哥德巴赫猜想（Goldbach's conjecture）是数论中存在最久的未解问题之一。这个猜想最早出现在1742年普鲁士数学家克里斯蒂安·哥德巴赫与瑞士数学家莱昂哈德·欧拉的通信中。用现代的数学语言，哥德巴赫猜想可以陈述为： +# 任一大于2的偶数，都可表示成两个素数之和 +# 验证1000以内偶数是否满足这一猜想，并输出每一个结果，例如 +# 4 = 2 + 2 +# 6 = 3 + 3 +# 8 = 3 + 5 \ No newline at end of file diff --git a/README.md b/README.md index 86813d1..63bda06 100644 --- a/README.md +++ b/README.md @@ -1,2 +1,3 @@ -# python_for_asl +# ASL成像基于Python的定量数据处理基础 + From da9dc81248a106123224732146b8b84f0bed105a Mon Sep 17 00:00:00 2001 From: liligeng111 Date: Wed, 6 Aug 2025 02:13:58 +0800 Subject: [PATCH 2/2] update requirements.txt --- requirements.txt | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) create mode 100644 requirements.txt diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000..43df810 --- /dev/null +++ b/requirements.txt @@ -0,0 +1,32 @@ +attr==0.3.2 +ConfigParser==7.2.0 +contextlib2==21.6.0 +cryptography==45.0.5 +Cython==3.1.2 +dl==0.1.0 +docutils==0.22 +HTMLParser==0.0.2 +ipython==7.34.0 +ipywidgets==8.1.7 +Jinja2==3.0.3 +jnius==1.1.0 +keyring==25.6.0 +lockfile==0.12.2 +matplotlib==3.3.4 +mock==5.2.0 +numpy==1.21.6 +openpyxl==3.1.3 +Pillow==9.5.0 +Pillow==11.3.0 +protobuf==3.14.0 +pyOpenSSL==25.1.0 +pywin32==308 +railroad==0.5.0 +scipy==1.5.4 +Sphinx==8.2.3 +thread==2.0.5 +tornado==6.2 +trove_classifiers==2025.5.9.12 +truststore==0.10.3 +urllib3_secure_extra==0.1.0 +xmlrpclib==1.0.1