Feature-Extraction/ecg_calculate_areas.py

import numpy as np

from scipy.integrate import simpson

from global_var import global_var_init
cycle, fs_ecg, fs_ppg, record_name, record_name_csv, data_path = global_var_init()

from ecg_time_interval import compute_interval
(intervals_ecg, ratio_pp_tt)= compute_interval()

from processing import processing
(signal_dwt , waves_dwt , rpeaks , ecg_signal,
 signal, signal.v, signal.fs, rows_to_extract, t_ecg, t_ppg,
 on_column_list, on_column_list_toArea, on_values, on_values_toArea,
 sp_column_list, sp_values, dn_column_list, dn_values, dp_column_list,
 dp_values, u_column_list, u_column_list_toArea, u_values, u_values_toArea,
 v_column_list, v_values, w_column_list, w_values, a_column_list, a_column_list_toArea,
 a_values, a_values_toArea, b_column_list, b_values, c_column_list, c_values,
 e_column_list, e_values, f_column_list, f_values,
 signal.ppg, signal.vpg, signal.apg, signal.jpg) = processing()

from ecg_peaks_val import ecg_peaks_val
(P_peaks, Q_peaks, R_peaks, S_peaks, T_peaks,
 P_onsets, P_offsets, T_onsets, T_offsets,
 P_peaks_values, Q_peaks_values, R_peaks_values, S_peaks_values, T_peaks_values, 
 P_onsets_values, P_offsets_values, T_onsets_values, T_offsets_values,
 PQ_baseline) = ecg_peaks_val()

num = len(R_peaks)

#--------------------------------------------------------------------------------------------------------
#--------------------------------------------------------------------------------------------------------
#--------------------------------------------------------------------------------------------------------
#--------------------------------------------------------------------------------------------------------
#-----------------------------------          计算面积            ----------------------------------------
#--------------------------------------------------------------------------------------------------------
#--------------------------------------------------------------------------------------------------------
#--------------------------------------------------------------------------------------------------------
#--------------------------------------------------------------------------------------------------------


# 计算每个周期相对于基线的面积
def ecg_areas_calculate():

    ecg_areas = []

    for i in range(num):
        # 获取当前周期的起点和终点，并将其转换为整数索引
        start = int(P_onsets[i])  
        end = int(T_offsets[i])  

        # print("此时的P_onsets为：", P_onsets[i])
        # print("此时的T_offsets为：", T_offsets[i])
        
        # 提取当前周期的ECG信号片段
        signal_segment = ecg_signal[start:end]
        
        # 提取当前周期的时间片段
        time_segment = t_ecg[start:end]

        # 当前周期的基线值（使用p_offset点的纵坐标值）
        baseline_value = P_offsets_values[i]
        
        # 创建一个与信号片段大小相同的基线片段
        baseline_segment = np.full_like(signal_segment.shape, baseline_value)
        
        # 计算相对于基线的信号
        relative_signal = signal_segment - baseline_segment

        # 相对于基线的信号进行积分，计算面积
        area = simpson(relative_signal, time_segment)


        
        # 将计算得到的面积添加到列表中
        ecg_areas.append(area)

    # # 输出所有周期的面积列表
    # print("ecg_areas：", ecg_areas)


    # 计算P_areas
    P_areas = []

    for i in range(num):
        # 获取当前周期的起点和终点，并将其转换为整数索引
        start = int(P_onsets[i])  
        end = int(P_offsets[i])  

        
        # 提取当前周期的ECG信号片段
        signal_segment = ecg_signal[start:end]
        
        # 提取当前周期的时间片段
        time_segment = t_ecg[start:end]
        
        # 当前周期的基线值（使用p_offset点的纵坐标值）
        baseline_value = P_offsets_values[i]
        
        # 创建一个与信号片段大小相同的基线片段
        baseline_segment = np.full_like(signal_segment.shape, baseline_value)
        
        # 计算相对于基线的信号
        relative_signal = signal_segment - baseline_segment
        
        # 相对于基线的信号进行积分，计算面积
        area = simpson(relative_signal, time_segment)
        
        # 将计算得到的面积添加到列表中
        P_areas.append(area)

    # # 输出所有周期的面积列表
    # print("P_areas:", P_areas)


    # 计算T_areas
    T_areas = []

    for i in range(num):
        # 获取当前周期的起点和终点，并将其转换为整数索引
        start = int(T_onsets[i])  
        end = int(T_offsets[i]) 

        # print("此时的T_onsets为：", T_onsets[i])
        # print("此时的T_offsets为：", T_offsets[i])

        # 提取当前周期的ECG信号片段
        signal_segment = ecg_signal[start:end]
        
        # 提取当前周期的时间片段
        time_segment = t_ecg[start:end]
        
        # 当前周期的基线值（使用T_offset点的纵坐标值）
        baseline_value = T_offsets_values[i]
        
        # 创建一个与信号片段大小相同的基线片段
        baseline_segment = np.full_like(signal_segment.shape, baseline_value)
        
        # 计算相对于基线的信号
        relative_signal = signal_segment - baseline_segment
        
        # 相对于基线的信号进行积分，计算面积
        area = simpson(relative_signal, time_segment)
        
        # 将计算得到的面积添加到列表中
        T_areas.append(area)

    # # 输出所有周期的面积列表
    # print("T_areas:", T_areas)


    # 计算QRS_areas
    QRS_areas = []

    for i in range(num):
        # 获取当前周期的起点和终点，并将其转换为整数索引
        start = int(Q_peaks[i])  
        end = int(S_peaks[i])  

        # 提取当前周期的ECG信号片段
        signal_segment = ecg_signal[start:end]
        
        # 提取当前周期的时间片段
        time_segment = t_ecg[start:end]
        
        # 当前周期的基线值（使用S_peaks点的纵坐标值）
        baseline_value = S_peaks_values[i]
        
        # 创建一个与信号片段大小相同的基线片段
        baseline_segment = np.full_like(signal_segment.shape, baseline_value)
        
        # 计算相对于基线的信号
        relative_signal = signal_segment - baseline_segment
        
        # 相对于基线的信号进行积分，计算面积
        area = simpson(relative_signal, time_segment)
        
        # 将计算得到的面积添加到列表中
        QRS_areas.append(area)

    # # 输出所有周期的面积列表
    # print("QRS_areas:", QRS_areas)


    # values = [  ecg_areas, P_areas, T_areas, QRS_areas]

    # # 遍历并打印每个列表的长度
    # for idx, values in enumerate(values):
    #     print(f"List {idx+1}: {len(values)}")

    return ecg_areas, P_areas, T_areas, QRS_areas


if __name__ == "__main__":
    ecg_areas_calculate()