### 短时心电图房颤检测的Python代码示例 以下是使用Python实现短时心电图（ECG）房颤（AFib）检测的一个简单代码示例。此代码主要依赖于`numpy`、`scipy`以及`wfdb`库来处理信号并提取特征，最终利用机器学习模型完成分类。 #### 数据预处理与特征提取 为了检测房颤，通常需要先对原始心电信号进行滤波和降噪处理[^4]。这里采用带通滤波器去除高频噪声和低频漂移： ```python import numpy as np from scipy.signal import butter, lfilter def butter_bandpass(lowcut, highcut, fs, order=5): nyquist = 0.5 * fs low = lowcut / nyquist high = highcut / nyquist b, a = butter(order, [low, high], btype='band') return b, a def butter_bandpass_filter(data, lowcut, highcut, fs, order=5): b, a = butter_bandpass(lowcut, highcut, fs, order=order) y = lfilter(b, a, data) return y ``` #### R波检测与RR间期计算 R波的位置可以通过差分法或小波变换找到[^4]。下面是一个基于差分法的简化版本： ```python def detect_r_peaks(signal, sampling_rate): diff_signal = np.diff(np.abs(signal)) squared_diff = diff_signal ** 2 moving_average = np.convolve(squared_diff, np.ones(int(0.1 * sampling_rate)) / (0.1 * sampling_rate), mode='same') threshold = 0.75 * np.max(moving_average) r_peak_indices = np.where(moving_average > threshold)[0] min_distance = int(0.2 * sampling_rate) # Minimum distance between peaks filtered_r_peaks = [] last_index = -min_distance for index in r_peak_indices: if index - last_index >= min_distance: filtered_r_peaks.append(index) last_index = index return np.array(filtered_r_peaks) # Example usage with MIT-BIH dataset using WFDB library import wfdb record = wfdb.rdrecord('data/MIT-BIH/100', sampfrom=0, sampto=10000, physical=False, channels=[0]) signal = record.p_signal[:, 0] filtered_signal = butter_bandpass_filter(signal, lowcut=0.5, highcut=40, fs=record.fs, order=3) r_peaks = detect_r_peaks(filtered_signal, record.fs) rr_intervals = np.diff(r_peaks) / record.fs ``` #### 特征工程与分类模型训练 对于房颤检测，常用的特征包括RR间期变异性、Poincaré plot分析等。以下展示如何构建基本的特征矩阵并与逻辑回归模型配合使用： ```python from sklearn.linear_model import LogisticRegression from sklearn.model_selection import train_test_split def extract_features(rr_intervals): features = { 'mean_rr': np.mean(rr_intervals), 'std_rr': np.std(rr_intervals), 'rmssd': np.sqrt(np.mean(np.square(np.diff(rr_intervals)))) } return list(features.values()) X = np.array([extract_features(rr_intervals)]) y = np.array([1]) # Label: AFib present or not # For demonstration purposes only; real-world datasets should be used here. model = LogisticRegression() X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) model.fit(X_train, y_train) print(f"Model score on test set: {model.score(X_test, y_test)}") ``` 以上代码仅为演示目的设计，在实际应用中需扩展数据集规模，并优化特征选择过程。 --- ### 动态图表制作补充说明 如果希望进一步增强可视化效果，可参考动态图表制作的相关技术[^2]。例如，借助Matplotlib动画功能实时更新RR间期变化曲线，从而更直观地观察潜在异常模式。 ```python import matplotlib.pyplot as plt from matplotlib.animation import FuncAnimation fig, ax = plt.subplots() x_data, y_data = [], [] line, = ax.plot([], []) def update(frame): x_data.append(len(x_data)) y_data.append(rr_intervals[frame % len(rr_intervals)]) # Simulate RR intervals dynamically line.set_data(x_data[-50:], y_data[-50:]) ax.relim() ax.autoscale_view(True, True, True) return line, ani = FuncAnimation(fig, update, interval=200) plt.show() ``` ---