{"id":92,"date":"2026-02-13T10:40:45","date_gmt":"2026-02-13T02:40:45","guid":{"rendered":"https:\/\/qdszxyy.com\/?p=92"},"modified":"2026-04-29T07:48:37","modified_gmt":"2026-04-28T23:48:37","slug":"%e8%84%91%e6%9c%ba%e6%8e%a5%e5%8f%a3%e6%95%b0%e6%8d%ae%e5%a4%84%e7%90%86%e8%bf%9e%e8%bd%bd%ef%bc%88%e4%ba%8c%ef%bc%89-%e6%95%b0%e6%8d%ae%e6%ba%90%e5%a4%b4%e8%a7%a3%e6%9e%90%ef%bc%9a%e8%84%91%e7%94%b5","status":"publish","type":"post","link":"https:\/\/qdszxyy.com\/?p=92","title":{"rendered":"\u8111\u673a\u63a5\u53e3\u6570\u636e\u5904\u7406\u8fde\u8f7d\uff08\u4e8c\uff09 \u6570\u636e\u6e90\u5934\u89e3\u6790\uff1a\u8111\u7535\u4fe1\u53f7\u7684\u91c7\u96c6\u539f\u7406\u4e0e\u6570\u636e\u7279\u6027"},"content":{"rendered":"

\u8111\u673a\u63a5\u53e3\uff08BCI\uff09\u7cfb\u7edf\u7684\u6838\u5fc3\u662f\u5bf9\u5927\u8111\u6d3b\u52a8\u4fe1\u53f7\u7684\u6709\u6548\u91c7\u96c6\u4e0e\u89e3\u8bfb\uff0c\u800c\u8111\u7535\u56fe\uff08EEG\uff09\u4f5c\u4e3a\u6700\u5e38\u7528\u7684\u975e\u4fb5\u5165\u5f0f\u8111\u4fe1\u53f7\u91c7\u96c6\u65b9\u5f0f\uff0c\u5176\u6570\u636e\u8d28\u91cf\u76f4\u63a5\u51b3\u5b9a\u4e86\u540e\u7eed\u5904\u7406\u4e0e\u5e94\u7528\u7684\u6548\u679c\u3002\u672c\u6587\u5c06\u4ece EEG \u4fe1\u53f7\u7684\u4ea7\u751f\u673a\u5236\u3001\u91c7\u96c6\u539f\u7406\u3001\u786c\u4ef6\u7ec4\u6210\u3001\u6570\u636e\u7279\u6027\u56db\u4e2a\u7ef4\u5ea6\u5c55\u5f00\u89e3\u6790\uff0c\u5e76\u7ed3\u5408 Python \u4ee3\u7801\u793a\u4f8b\u5c55\u793a\u5982\u4f55\u5904\u7406\u4e0e\u5206\u6790 EEG \u539f\u59cb\u6570\u636e\uff0c\u5e2e\u52a9\u8bfb\u8005\u6df1\u5165\u7406\u89e3 BCI \u7684 \u201c\u6570\u636e\u6e90\u5934\u201d\u3002<\/p>\n

\u4e00\u3001EEG \u4fe1\u53f7\u7684\u4ea7\u751f\u673a\u5236\uff1a\u5927\u8111\u6d3b\u52a8\u7684 \u201c\u7535\u6307\u7eb9\u201d
\n\u5927\u8111\u7531\u7ea6 1000 \u4ebf\u4e2a\u795e\u7ecf\u5143\u7ec4\u6210\uff0c\u795e\u7ecf\u5143\u901a\u8fc7\u7a81\u89e6\u4f20\u9012\u7535\u4fe1\u53f7\uff0c\u5f62\u6210\u590d\u6742\u7684\u795e\u7ecf\u7f51\u7edc\u6d3b\u52a8\u3002EEG \u4fe1\u53f7\u7684\u672c\u8d28\u662f\u5927\u91cf\u795e\u7ecf\u5143\u7a81\u89e6\u540e\u7535\u4f4d\uff08EPSP\/IPSP\uff09\u5728\u5934\u76ae\u4e0a\u7684\u53e0\u52a0\u7535\u4f4d\u5dee\uff0c\u5176\u5f62\u6210\u8fc7\u7a0b\u53ef\u6982\u62ec\u4e3a\uff1a<\/p>\n

\u795e\u7ecf\u5143\u6d3b\u52a8\uff1a\u5355\u4e2a\u795e\u7ecf\u5143\u7684\u7a81\u89e6\u540e\u7535\u4f4d\u5e45\u5ea6\u4ec5\u4e3a\u5fae\u4f0f\u7ea7\uff0c\u65e0\u6cd5\u88ab\u5934\u76ae\u7535\u6781\u68c0\u6d4b\uff1b\u4f46\u5f53\u5927\u8111\u76ae\u5c42\u4e2d\u6570\u5343\u81f3\u6570\u767e\u4e07\u4e2a\u9525\u4f53\u7ec6\u80de\u5728\u7a7a\u95f4\u6392\u5217\u4e00\u81f4\u4e14\u65f6\u95f4\u4e0a\u540c\u6b65\u6d3b\u52a8\u65f6\uff0c\u5176\u7a81\u89e6\u540e\u7535\u6d41\u7684\u53e0\u52a0\u6548\u5e94\u624d\u80fd\u5728\u5934\u76ae\u8868\u9762\u5f62\u6210\u53ef\u6d4b\u91cf\u7684\u7535\u4f4d\u5dee\uff08\u7ea6 1\u2013100 \u03bcV\uff09\u3002
\n\u8111\u7535\u6ce2\u7684\u5206\u7c7b\uff1a\u6839\u636e\u9891\u7387\u548c\u751f\u7406\u610f\u4e49\uff0cEEG \u4fe1\u53f7\u53ef\u5206\u4e3a\u4e0d\u540c\u9891\u6bb5\uff0c\u5404\u9891\u6bb5\u4e0e\u7279\u5b9a\u7684\u5927\u8111\u72b6\u6001\u5f3a\u76f8\u5173\uff1a
\n\u03b4 \u6ce2\uff080.5\u20134 Hz\uff09\uff1a\u6df1\u5ea6\u7761\u7720\u3001\u5a74\u513f\u671f\uff0c\u5f02\u5e38\u6d3b\u8dc3\u53ef\u80fd\u4e0e\u8111\u90e8\u75be\u75c5\u76f8\u5173
\n\u03b8 \u6ce2\uff084\u20138 Hz\uff09\uff1a\u56f0\u5026\u3001\u653e\u677e\u3001\u521b\u9020\u6027\u601d\u7ef4\uff0c\u6ce8\u610f\u529b\u5206\u6563\u65f6\u6613\u589e\u5f3a
\n\u03b1 \u6ce2\uff088\u201313 Hz\uff09\uff1a\u6e05\u9192\u4f46\u95ed\u773c\u653e\u677e\u72b6\u6001\uff08\u7741\u773c\u6216\u6ce8\u610f\u529b\u96c6\u4e2d\u65f6\u51fa\u73b0 \u03b1 \u6291\u5236\uff09\uff0c\u9876\u6795\u533a\u6700\u4e3a\u663e\u8457
\n\u03b2 \u6ce2\uff0813\u201330 Hz\uff09\uff1a\u8b66\u89c9\u3001\u4e13\u6ce8\u3001\u8fd0\u52a8\u51c6\u5907\uff0c\u4e2d\u592e\u524d\u56de\uff08\u8fd0\u52a8\u533a\uff09\u5728\u52a8\u4f5c\u6267\u884c\u65f6\u6d3b\u8dc3
\n\u03b3 \u6ce2\uff08>30 Hz\uff09\uff1a\u9ad8\u7ea7\u8ba4\u77e5\u529f\u80fd\uff08\u5982\u6ce8\u610f\u529b\u3001\u8bb0\u5fc6\u3001\u610f\u8bc6\u6574\u5408\uff09\uff0c\u524d\u989d\u53f6\u4e0e\u9876\u53f6\u534f\u540c\u6d3b\u52a8\u65f6\u589e\u5f3a
\n\u8fd9\u4e9b\u9891\u6bb5\u7684\u7ec4\u5408\u4e0e\u52a8\u6001\u53d8\u5316\uff0c\u6784\u6210\u4e86\u5927\u8111\u6d3b\u52a8\u7684 \u201c\u7535\u6307\u7eb9\u201d\uff0c\u662f BCI \u89e3\u7801\u7528\u6237\u610f\u56fe\u7684\u6838\u5fc3\u4f9d\u636e\u3002<\/p>\n

\u4e8c\u3001EEG \u4fe1\u53f7\u7684\u91c7\u96c6\u539f\u7406\uff1a\u4ece\u5934\u76ae\u5230\u6570\u636e\u7684\u8f6c\u5316
\nEEG \u91c7\u96c6\u7684\u6838\u5fc3\u662f\u901a\u8fc7\u7535\u6781\u6355\u6349\u5934\u76ae\u4e0a\u7684\u7535\u4f4d\u5dee\uff0c\u5e76\u5c06\u5176\u8f6c\u6362\u4e3a\u7a33\u5b9a\u53ef\u5206\u6790\u7684\u6570\u5b57\u4fe1\u53f7\uff0c\u5176\u5173\u952e\u539f\u7406\u5305\u62ec\uff1a<\/p>\n

1. \u7535\u6781\u4e0e\u5934\u76ae\u7684\u63a5\u89e6
\n\u7535\u6781\u6750\u8d28\uff1a\u5e38\u7528\u94f6 \/ \u6c2f\u5316\u94f6\uff08Ag\/AgCl\uff09\u7535\u6781\uff0c\u5177\u6709\u7a33\u5b9a\u7684\u7535\u5316\u5b66\u7279\u6027\u548c\u4f4e\u534a\u7535\u6c60\u7535\u4f4d\uff0c\u80fd\u51cf\u5c11\u76ae\u80a4\u4e0e\u7535\u6781\u95f4\u7684\u6781\u5316\u6548\u5e94\uff0c\u964d\u4f4e\u63a5\u89e6\u963b\u6297\u3002
\n\u5bfc\u7535\u4ecb\u8d28\uff1a\u6e7f\u7535\u6781\u9700\u4f7f\u7528\u5bfc\u7535\u818f \/ \u76d0\u6c34\uff0c\u964d\u4f4e\u76ae\u80a4\u89d2\u8d28\u5c42\u7684\u963b\u6297\uff08\u7406\u60f3\u63a5\u89e6\u963b\u6297\u5e94 < 5 k\u03a9\uff09\uff1b\u5e72\u7535\u6781\u65e0\u9700\u5bfc\u7535\u4ecb\u8d28\uff0c\u4f46\u672a\u6e05\u6d01\u76ae\u80a4\u65f6\u963b\u6297\u53ef\u9ad8\u8fbe\u6570 M\u03a9\uff0c\u9700\u4f9d\u8d56\u9ad8\u8f93\u5165\u963b\u6297\u653e\u5927\u5668\u8865\u507f\u3002
\n\u7535\u6781\u4f4d\u7f6e\uff1a\u9075\u5faa\u56fd\u9645\u6807\u51c6 10\u201320 \u7cfb\u7edf\uff08\u5982 Fp1\u3001Fp2\u3001C3\u3001C4\u3001O1\u3001O2 \u7b49\uff09\u53ca\u5176\u6269\u5c55\u7248 10-10 \u7cfb\u7edf\uff0c\u901a\u8fc7 \u201c10%\/20% \u9885\u9aa8\u5468\u957f\u95f4\u9694\u201d \u89c4\u5219\u5b9a\u4f4d\uff0c\u786e\u4fdd\u91c7\u96c6\u4f4d\u7f6e\u7684\u89c4\u8303\u6027\u548c\u8de8\u5b9e\u9a8c\u53ef\u6bd4\u6027\u3002
\n2. \u53c2\u8003\u7535\u6781\u4e0e\u63a5\u5730\u7535\u6781
\n\u53c2\u8003\u7535\u6781\uff1a\u4e3a\u6240\u6709\u6d4b\u91cf\u7535\u6781\u63d0\u4f9b\u7edf\u4e00\u7684\u7535\u4f4d\u6bd4\u8f83\u57fa\u51c6\u3002\u5e38\u7528\u4f4d\u7f6e\u5305\u62ec\u8033\u5782\uff08A1\u3001A2\uff09\u3001\u4e73\u7a81\u6216\u5934\u9876\uff08Cz\uff09\uff0c\u9700\u6ce8\u610f\uff1a\u6ca1\u6709\u7edd\u5bf9 \u201c\u96f6\u7535\u4f4d\u201d \u7684\u53c2\u8003\u70b9\uff0c\u53c2\u8003\u7535\u6781\u672c\u8eab\u4f1a\u8bb0\u5f55\u751f\u7406\u6d3b\u52a8\uff0c\u5176\u9009\u62e9\u76f4\u63a5\u5f71\u54cd\u4fe1\u53f7\u5f62\u6001\uff0c\u56e0\u6b64\u6570\u636e\u5206\u6790\u4e2d\u5e38\u8fdb\u884c \u201c\u518d\u53c2\u8003\u201d\uff08\u5982\u5e73\u5747\u53c2\u8003\u3001\u53cc\u4fa7\u4e73\u7a81\u53c2\u8003\uff09\u4f18\u5316\u3002
\n\u63a5\u5730\u7535\u6781\uff1a\u8fde\u63a5\u7cfb\u7edf\u5730\uff0c\u4e3a\u653e\u5927\u7535\u8def\u63d0\u4f9b\u516c\u5171\u53c2\u8003\u70b9\uff0c\u6291\u5236\u73af\u5883\u566a\u58f0\uff08\u5982\u5de5\u9891\u5e72\u6270\uff09\u7684\u5171\u6a21\u4f20\u5bfc\uff0c\u901a\u5e38\u653e\u7f6e\u5728\u989d\u5934\uff08Fpz \u9644\u8fd1\uff09\u6216\u9888\u90e8\uff0c\u9700\u8fdc\u79bb\u6d4b\u91cf\u7535\u6781\u907f\u514d\u4fe1\u53f7\u5e72\u6270\u3002
\n3. \u4fe1\u53f7\u653e\u5927\u4e0e\u6ee4\u6ce2
\n\u653e\u5927\uff1aEEG \u4fe1\u53f7\u5fae\u5f31\uff08\u03bcV \u7ea7\uff09\uff0c\u9700\u901a\u8fc7\u9ad8\u8f93\u5165\u963b\u6297\uff08\u226510\u00b9\u00b2 \u03a9\uff09\u3001\u9ad8\u5171\u6a21\u6291\u5236\u6bd4\uff08CMRR \u2265 100 dB\uff09 \u7684\u5dee\u5206\u653e\u5927\u5668\u653e\u5927 10\u2074\u201310\u2076 \u500d\uff0c\u4ec5\u653e\u5927\u6d4b\u91cf\u7535\u6781\u4e0e\u53c2\u8003\u7535\u6781\u7684\u5dee\u6a21\u4fe1\u53f7\uff08\u76ee\u6807 EEG\uff09\uff0c\u6291\u5236\u5171\u6a21\u566a\u58f0\uff08\u5982 50 Hz \u5de5\u9891\u3001\u7535\u78c1\u8f90\u5c04\uff09\u3002
\n\u6ee4\u6ce2\uff1a
\n\u9ad8\u901a\u6ee4\u6ce2\uff080.5\u20131 Hz\uff09\uff1a\u53bb\u9664\u4f4e\u9891\u6f02\u79fb\uff08\u5982\u547c\u5438\u3001\u51fa\u6c57\u3001\u7535\u6781\u6781\u5316\u5bfc\u81f4\u7684\u57fa\u7ebf\u6f02\u79fb\uff09\uff0c\u907f\u514d\u4fe1\u53f7\u57fa\u7ebf\u504f\u79fb\u5f71\u54cd\u540e\u7eed\u5206\u6790\u3002
\n\u4f4e\u901a\u6ee4\u6ce2\uff1a\u622a\u6b62\u9891\u7387\u901a\u5e38\u8bbe\u4e3a\u91c7\u6837\u7387\u7684 1\/3~1\/2\uff08\u6297\u6df7\u53e0\u8981\u6c42\uff09\uff0c\u82e5\u7814\u7a76\u5305\u542b \u03b3 \u6ce2\u6bb5\uff08>30 Hz\uff09\uff0c\u9700\u63d0\u9ad8\u81f3 100 Hz \u4ee5\u4e0a\uff0c\u907f\u514d\u9ad8\u9891\u4fe1\u53f7\u5931\u771f\u3002
\n\u9677\u6ce2\u6ee4\u6ce2\uff0850 Hz \u6216 60 Hz\uff0c\u542b\u8c10\u6ce2\uff09\uff1a\u9488\u5bf9\u6027\u53bb\u9664\u5de5\u9891\u5e72\u6270\uff0c\u9700\u6ce8\u610f\u7a84\u5e26\u9677\u6ce2\u53ef\u80fd\u5f71\u54cd\u90bb\u8fd1\u9891\u6bb5\uff08\u5982 48\u201352 Hz\uff09\u4fe1\u53f7\u3002
\n4. \u6a21\u6570\u8f6c\u6362\uff08ADC\uff09
\n\u91c7\u6837\u7387\uff1a\u9075\u5faa\u5948\u594e\u65af\u7279\u51c6\u5219\uff08\u91c7\u6837\u7387 \u2265 2 \u500d\u4fe1\u53f7\u6700\u9ad8\u9891\u7387\uff09\uff0cEEG \u91c7\u96c6\u5e38\u7528 250 Hz\u20131000 Hz\uff1b\u82e5\u5206\u6790 \u03b3 \u6ce2\uff08>30 Hz\uff09\uff0c\u5efa\u8bae\u91c7\u6837\u7387 \u2265 250 Hz\uff1b\u9ad8\u5bc6\u5ea6 EEG \u6216\u8111\u7535\u4e0e\u5176\u4ed6\u6a21\u6001\u540c\u6b65\u65f6\uff0c\u53ef\u63d0\u5347\u81f3 1000\u20132000 Hz\u3002
\n\u5206\u8fa8\u7387\uff1aADC \u4f4d\u6570\uff0816 \u4f4d \/ 24 \u4f4d\uff09\u51b3\u5b9a\u91cf\u5316\u7cbe\u5ea6\u3002\u4f8b\u5982\uff1a16 \u4f4d ADC \u8f93\u51fa\u8303\u56f4 \u00b132768\uff0c\u82e5\u653e\u5927\u5668\u6ee1\u91cf\u7a0b \u00b12.5 V\uff0c\u91cf\u5316\u6b65\u957f \u0394V \u2248 76.3 nV\uff1b\u5b9e\u9645\u539f\u59cb EEG \u4fe1\u53f7\u4e3a \u03bcV \u7ea7\uff0c\u9700\u901a\u8fc7 \u201c\u7cfb\u7edf\u589e\u76ca\u53cd\u5411\u8ba1\u7b97\u201d \u8fd8\u539f\u771f\u5b9e\u7535\u4f4d\uff08\u5982\u589e\u76ca 10\u2075 \u65f6\uff0cADC \u8bfb\u6570 1 LSB \u5bf9\u5e94\u539f\u59cb\u4fe1\u53f7 0.763 \u03bcV\uff09\u300224 \u4f4d ADC \u91cf\u5316\u6b65\u957f\u66f4\u5c0f\uff08\u540c\u6ee1\u91cf\u7a0b\u4e0b\u7ea6 0.06 \u03bcV\uff09\uff0c\u66f4\u9002\u5408\u5fae\u5f31\u4fe1\u53f7\u91c7\u96c6\u3002
\n\u4e09\u3001EEG \u91c7\u96c6\u7cfb\u7edf\u7684\u786c\u4ef6\u7ec4\u6210
\n\u4e00\u4e2a\u5b8c\u6574\u7684 EEG \u91c7\u96c6\u7cfb\u7edf\u901a\u5e38\u5305\u62ec\u4ee5\u4e0b\u6838\u5fc3\u6a21\u5757\uff0c\u5404\u6a21\u5757\u7684\u6027\u80fd\u76f4\u63a5\u51b3\u5b9a\u6570\u636e\u8d28\u91cf\uff1a<\/p>\n

1. \u7535\u6781\u5e3d \/ \u9635\u5217
\n\u5305\u542b\u6d4b\u91cf\u7535\u6781\u3001\u53c2\u8003\u7535\u6781\u548c\u63a5\u5730\u7535\u6781\uff0c\u6309\u5bfc\u7535\u65b9\u5f0f\u5206\u4e3a\uff1a<\/p>\n

\u6e7f\u7535\u6781\u5e3d\uff08\u9700\u5bfc\u7535\u818f\uff09\uff1a\u4fe1\u53f7\u8d28\u91cf\u9ad8\u3001\u963b\u6297\u7a33\u5b9a\uff0c\u9002\u5408\u79d1\u7814\u573a\u666f\uff08\u5982 Neuroscan SynAmps\u3001Biosemi ActiveTwo\uff09\uff0c\u4f46\u51c6\u5907\u65f6\u95f4\u957f\uff0815\u201330 \u5206\u949f\uff09\u3001\u7528\u6237\u8212\u9002\u5ea6\u4f4e\u3002
\n\u5e72\u7535\u6781\u5e3d\uff08\u65e0\u9700\u5bfc\u7535\u818f\uff09\uff1a\u4fbf\u643a\u6027\u597d\u3001\u51c6\u5907\u5feb\uff08<5 \u5206\u949f\uff09\uff0c\u9002\u5408\u6d88\u8d39\u7ea7 \/ \u79fb\u52a8\u573a\u666f\uff0c\u4f46\u963b\u6297\u9ad8\u3001\u4fe1\u53f7\u4fe1\u566a\u6bd4\u7565\u4f4e\uff0c\u901a\u9053\u6570\u901a\u5e38\u8f83\u5c11\uff084\u201316 \u901a\u9053\uff09\uff0c\u4e0d\u9002\u5408\u9ad8\u9891 \u03b3 \u5206\u6790\u6216\u9ad8\u7cbe\u5ea6\u7814\u7a76\u3002
\n2. \u4fe1\u53f7\u8c03\u7406\u6a21\u5757
\n\u5dee\u5206\u653e\u5927\u5668\uff1a\u6838\u5fc3\u90e8\u4ef6\uff0c\u8d1f\u8d23\u4fe1\u53f7\u653e\u5927\u4e0e\u5171\u6a21\u566a\u58f0\u6291\u5236\uff0c\u8f93\u5165\u963b\u6297\u548c CMRR \u662f\u5173\u952e\u6307\u6807\u3002
\n\u786c\u4ef6\u6ee4\u6ce2\uff1aRC \u6ee4\u6ce2\u6216\u6709\u6e90\u6ee4\u6ce2\uff0c\u5b9e\u73b0\u521d\u6b65\u6297\u6df7\u53e0\u548c\u4f4e\u9891\u6f02\u79fb\u53bb\u9664\uff0c\u51cf\u5c11\u540e\u7eed\u6570\u5b57\u6ee4\u6ce2\u538b\u529b\u3002
\n\u589e\u76ca\u63a7\u5236\uff1a\u53ef\u8c03\u8282\u653e\u5927\u500d\u6570\uff08\u5982 10\u2074\u201310\u2076 \u500d\uff09\uff0c\u9002\u914d\u4e0d\u540c\u7528\u6237\u7684\u4fe1\u53f7\u5f3a\u5ea6\uff08\u5982\u5934\u76ae\u963b\u6297\u5dee\u5f02\u5bfc\u81f4\u7684\u4fe1\u53f7\u5e45\u5ea6\u53d8\u5316\uff09\u3002
\n3. \u6570\u636e\u91c7\u96c6\u5361\uff08DAQ\uff09
\nADC \u82af\u7247\uff1a\u5c06\u6a21\u62df\u4fe1\u53f7\u8f6c\u6362\u4e3a\u6570\u5b57\u4fe1\u53f7\uff0c\u5206\u8fa8\u7387\uff0816\/24 \u4f4d\uff09\u548c\u91c7\u6837\u7387\u51b3\u5b9a\u6570\u636e\u91cf\u5316\u7cbe\u5ea6\u548c\u65f6\u95f4\u5206\u8fa8\u7387\u3002
\n\u4f20\u8f93\u63a5\u53e3\uff1a\u79d1\u7814\u7ea7\u8bbe\u5907\u5e38\u7528 USB 3.0 \u6216\u5149\u7ea4\uff08\u4f4e\u5ef6\u8fdf\u3001\u6297\u5e72\u6270\uff09\uff0c\u6d88\u8d39\u7ea7\u8bbe\u5907\u5e38\u7528\u84dd\u7259 5.0\/WiFi\uff08\u65e0\u7ebf\u4fbf\u643a\uff09\uff0c\u9700\u6ce8\u610f\u540c\u6b65\u6027\uff08\u591a\u6a21\u6001\u91c7\u96c6\u65f6\u5efa\u8bae\u7528\u786c\u4ef6\u89e6\u53d1\uff09\u3002
\n4. \u8f6f\u4ef6\u7cfb\u7edf
\n\u6570\u636e\u8bb0\u5f55\uff1a\u5b58\u50a8\u539f\u59cb EEG \u6570\u636e\uff0c\u5e38\u7528\u683c\u5f0f\u4e3a EDF\uff08\u4e34\u5e8a\uff09\u3001BDF\uff08\u6269\u5c55\u7248 EDF\uff09\u3001FIF\uff08MNE \u4e13\u7528\uff0c\u652f\u6301\u5143\u6570\u636e\u5b58\u50a8\uff09\u3002
\n\u5b9e\u65f6\u76d1\u63a7\uff1a\u663e\u793a\u4fe1\u53f7\u6ce2\u5f62\u3001\u7535\u6781\u963b\u6297\u3001\u566a\u58f0\u6c34\u5e73\uff0c\u4fbf\u4e8e\u91c7\u96c6\u8fc7\u7a0b\u4e2d\u6392\u67e5\u63a5\u89e6\u4e0d\u826f\u7b49\u95ee\u9898\u3002
\n\u53c2\u6570\u914d\u7f6e\uff1a\u652f\u6301\u91c7\u6837\u7387\u3001\u6ee4\u6ce2\u53c2\u6570\u3001\u7535\u6781\u4f4d\u7f6e\u6620\u5c04\u3001\u89e6\u53d1\u4fe1\u53f7\u8bbe\u7f6e\u7b49\uff0c\u9002\u914d\u4e0d\u540c\u5b9e\u9a8c\u8303\u5f0f\u3002
\n\u56db\u3001EEG \u6570\u636e\u7684\u7279\u6027\uff1a\u6311\u6218\u4e0e\u673a\u9047
\nEEG \u6570\u636e\u7684\u72ec\u7279\u7279\u6027\u65e2\u662f BCI \u89e3\u8bfb\u5927\u8111\u610f\u56fe\u7684\u57fa\u7840\uff0c\u4e5f\u662f\u6570\u636e\u5904\u7406\u7684\u6838\u5fc3\u6311\u6218\uff1a<\/p>\n

1. \u4f4e\u4fe1\u566a\u6bd4\uff08SNR\uff09
\nEEG \u4fe1\u53f7\u5fae\u5f31\uff081\u2013100 \u03bcV\uff09\uff0c\u6613\u53d7\u73af\u5883\u566a\u58f0\uff0850 Hz \u5de5\u9891\u3001\u7535\u78c1\u5e72\u6270\u3001\u7535\u6e90\u7ebf\u8f90\u5c04\uff09\u548c\u751f\u7406\u566a\u58f0\uff08\u773c\u7535 EOG\u3001\u808c\u7535 EMG\u3001\u5fc3\u7535 ECG\u3001\u547c\u5438\u5e72\u6270\uff09\u5f71\u54cd\uff0c SNR \u901a\u5e38 < 10 dB\uff0c\u4e14\u968f\u4efb\u52a1\u7c7b\u578b\u3001\u9891\u6bb5\u3001\u7535\u6781\u4f4d\u7f6e\u53d8\u5316\u663e\u8457\uff08\u5982 \u03b3 \u6ce2\u6bb5 SNR \u4f4e\u4e8e \u03b1 \u6ce2\u6bb5\uff09\u3002<\/p>\n

2. \u975e\u5e73\u7a33\u6027
\nEEG \u4fe1\u53f7\u7684\u7edf\u8ba1\u7279\u6027\uff08\u5747\u503c\u3001\u65b9\u5dee\u3001\u529f\u7387\u8c31\uff09\u548c\u62d3\u6251\u5206\u5e03\u968f\u65f6\u95f4\u52a8\u6001\u53d8\u5316\uff1a\u5982\u4ece\u653e\u677e\u5230\u4e13\u6ce8\u65f6\uff0c\u03b1 \u6ce2\u529f\u7387\u4e0b\u964d\u3001\u03b2 \u6ce2\u529f\u7387\u4e0a\u5347\uff1b\u6ce8\u610f\u529b\u8f6c\u79fb\u65f6\uff0c\u03b1 \u6ce2\u4f18\u52bf\u533a\u4ece\u9876\u6795\u533a\u8f6c\u5411\u989d\u533a\uff0c\u96be\u4ee5\u7528\u56fa\u5b9a\u6a21\u578b\u63cf\u8ff0\uff0c\u9700\u7528\u81ea\u9002\u5e94\u5904\u7406\u65b9\u6cd5\uff08\u5982\u65f6\u53d8\u6ee4\u6ce2\u3001\u6ed1\u52a8\u7a97\u53e3\u5206\u6790\uff09\u3002<\/p>\n

3. \u4e2a\u4f53\u5dee\u5f02
\n\u4e0d\u540c\u4eba\u7684\u8111\u7535\u6ce2\u6a21\u5f0f\u5b58\u5728\u663e\u8457\u5dee\u5f02\uff1a\u5982 \u03b1 \u6ce2\u5cf0\u503c\u9891\u7387\u5728 8\u201313 Hz \u8303\u56f4\u5185\u56e0\u4eba\u800c\u5f02\uff0c\u8fd0\u52a8\u60f3\u8c61\u76f8\u5173\u7684 \u03b2 \u6ce2\u589e\u5f3a\u7a0b\u5ea6\u4e5f\u5b58\u5728\u4e2a\u4f53\u5dee\u5f02\uff0c\u5bfc\u81f4 BCI \u6a21\u578b\u6cdb\u5316\u80fd\u529b\u53d7\u9650\uff0c\u901a\u5e38\u9700\u8981\u4e2a\u6027\u5316\u6821\u51c6\u3002<\/p>\n

4. \u9ad8\u7ef4\u5ea6\u6027
\n\u9ad8\u5bc6\u5ea6 EEG \u7cfb\u7edf\uff0864\/128\/256 \u901a\u9053\uff09\u4ea7\u751f\u6d77\u91cf\u6570\u636e\uff1a\u5982 128 \u901a\u9053 \u00d7 500 Hz \u91c7\u6837\u7387 = 64000 \u6570\u636e\u70b9 \/ \u79d2\uff0c1 \u5c0f\u65f6\u6570\u636e\u91cf\u7ea6 230 MB\uff0816 \u4f4d\uff09\uff0c\u5bf9\u5b58\u50a8\u548c\u8ba1\u7b97\u8d44\u6e90\u8981\u6c42\u8f83\u9ad8\uff0c\u9700\u901a\u8fc7\u7279\u5f81\u63d0\u53d6\u6216\u964d\u7ef4\uff08\u5982 CSP\u3001PCA\uff09\u51cf\u5c11\u6570\u636e\u7ef4\u5ea6\u3002<\/p>\n

5. \u751f\u7406\u4f2a\u8ff9\uff08Artifacts\uff09
\n\u4f2a\u8ff9\u662f EEG \u6570\u636e\u5904\u7406\u7684\u4e3b\u8981\u6311\u6218\uff0c\u9700\u91cd\u70b9\u8bc6\u522b\u4e0e\u53bb\u9664\uff1a<\/p>\n

\u773c\u7535\u4f2a\u8ff9\uff08EOG\uff09\uff1a\u7728\u773c\u6216\u773c\u7403\u8fd0\u52a8\u5bfc\u81f4\uff0c\u5e45\u5ea6\u53ef\u8fbe 100\u2013500 \u03bcV\uff0c\u662f\u6700\u4e3b\u8981\u7684\u4f2a\u8ff9\u6765\u6e90\uff0c\u5728\u524d\u989d\u7535\u6781\uff08Fp1\u3001Fp2\uff09\u6700\u4e3a\u663e\u8457\u3002
\n\u808c\u7535\u4f2a\u8ff9\uff08EMG\uff09\uff1a\u5934\u90e8\u808c\u8089\u6536\u7f29\uff08\u5480\u56bc\u3001\u76b1\u7709\u3001\u9888\u90e8\u808c\u8089\u7d27\u5f20\uff09\u4ea7\u751f\u7684\u9ad8\u9891\u566a\u58f0\uff08>30 Hz\uff09\uff0c\u5728\u989e\u533a\u3001\u989d\u533a\u7535\u6781\u6613\u51fa\u73b0\u3002
\n\u5fc3\u7535\u4f2a\u8ff9\uff08ECG\uff09\uff1a\u5fc3\u810f\u8df3\u52a8\u4ea7\u751f\u7684\u5468\u671f\u6027\u4fe1\u53f7\uff08\u7ea6 1 Hz\uff09\uff0c\u5728\u9760\u8fd1\u9888\u90e8\u7684\u7535\u6781\uff08\u5982 A1\u3001A2\uff09\u6216\u4f4e\u9891\u6bb5\uff08\u03b4\/\u03b8 \u6ce2\uff09\u660e\u663e\u3002
\n\u7535\u6781\u4f2a\u8ff9\uff1a\u7535\u6781\u79fb\u52a8\u3001\u63a5\u89e6\u4e0d\u826f\u5bfc\u81f4\u7684\u7a81\u53d1\u5c16\u5cf0\u566a\u58f0\uff0c\u5e45\u5ea6\u5927\u3001\u65e0\u89c4\u5f8b\uff0c\u6613\u6c61\u67d3\u5c40\u90e8\u901a\u9053\u6570\u636e\u3002
\n\u4e94\u3001\u4ee3\u7801\u793a\u4f8b\uff1aEEG \u539f\u59cb\u6570\u636e\u7684\u5904\u7406\u4e0e\u7279\u6027\u5206\u6790
\n\u4ee5\u4e0b\u901a\u8fc7 Python \u4ee3\u7801\u793a\u4f8b\uff0c\u5c55\u793a EEG \u539f\u59cb\u6570\u636e\u7684\u52a0\u8f7d\u3001\u53ef\u89c6\u5316\u3001\u529f\u7387\u8c31\u5206\u6790\u3001\u4f2a\u8ff9\u5904\u7406\u53ca ERP \u5206\u6790\uff0c\u4f7f\u7528 MNE-Python \u5e93\uff08\u79d1\u7814\u7ea7 EEG \u5904\u7406\u5de5\u5177\uff09\u548c\u5185\u7f6e\u516c\u5f00\u6570\u636e\u96c6\uff0c\u786e\u4fdd\u4ee3\u7801\u53ef\u76f4\u63a5\u590d\u73b0\u3002<\/p>\n

\u73af\u5883\u51c6\u5907
\n\u5148\u5b89\u88c5\u5fc5\u8981\u4f9d\u8d56\u5e93\uff1a<\/p>\n

bash<\/p>\n

pip install mne numpy scipy matplotlib
\n\u4e00\u952e\u83b7\u53d6\u5b8c\u6574\u9879\u76ee\u4ee3\u7801
\n1. \u52a0\u8f7d EEG \u793a\u4f8b\u6570\u636e
\n\u4f7f\u7528 MNE \u5185\u7f6e\u7684 EEG Motor Movement\/Imagery \u516c\u5f00\u6570\u636e\u96c6\uff08BCI Competition IV \u6570\u636e\u96c6\u5b50\u96c6\uff09\uff0c\u5305\u542b\u8fd0\u52a8\u60f3\u8c61\u4efb\u52a1\u6570\u636e\uff0c\u65e0\u9700\u989d\u5916\u4e0b\u8f7d\u3002<\/p>\n

python<\/p>\n

import mne
\nimport numpy as np
\nimport matplotlib.pyplot as plt
\nfrom mne.datasets import eegbci
\nfrom mne import events_from_annotations<\/p>\n

# \u52a0\u8f7d\u53d7\u8bd5\u80051\u7684Run 3\u6570\u636e\uff08\u5305\u542b\u5de6\u624b\/\u53f3\u624b\u8fd0\u52a8\u60f3\u8c61\u4efb\u52a1\uff09
\nraw_fnames = eegbci.load_data(subject=1, runs=[3])
\nraw = mne.io.read_raw_edf(raw_fnames[0], preload=True) # preload=True \u52a0\u8f7d\u6570\u636e\u5230\u5185\u5b58<\/p>\n

# \u6807\u51c6\u5316\u8bbe\u7f6e\uff1a\u7535\u6781\u4f4d\u7f6e\u6620\u5c04\u3001\u53c2\u8003\u7535\u6781\u914d\u7f6e
\nmne.datasets.eegbci.standardize(raw) # \u6807\u51c6\u5316\u901a\u9053\u540d\u79f0\u548c\u7c7b\u578b
\nraw.set_montage(‘standard_1005’) # \u52a0\u8f7d10-20\u7cfb\u7edf\u6269\u5c55\u7248\u7535\u6781\u4f4d\u7f6e<\/p>\n

# \u67e5\u770b\u6570\u636e\u6838\u5fc3\u4fe1\u606f
\nprint(“\u6570\u636e\u5f62\u72b6\uff08\u901a\u9053\u6570\u00d7\u65f6\u95f4\u70b9\uff09:”, raw.get_data().shape)
\nprint(“\u91c7\u6837\u7387:”, raw.info[‘sfreq’], “Hz”)
\nprint(“\u603b\u901a\u9053\u6570:”, len(raw.info[‘ch_names’]))
\nprint(“\u524d10\u4e2a\u901a\u9053\u540d\u79f0:”, raw.info[‘ch_names’][:10])
\nprint(“\u6570\u636e\u603b\u65f6\u957f:”, round(raw.times[-1], 2), “\u79d2”)
\n\u4e00\u952e\u83b7\u53d6\u5b8c\u6574\u9879\u76ee\u4ee3\u7801<\/p>\n

2. \u539f\u59cb EEG \u4fe1\u53f7\u7684\u53ef\u89c6\u5316
\n\u76f4\u89c2\u89c2\u5bdf\u539f\u59cb\u4fe1\u53f7\u7684\u6ce2\u5f62\u3001\u5e45\u5ea6\u53ca\u566a\u58f0\u7279\u6027\uff1a<\/p>\n

python<\/p>\n

# \u9009\u62e9\u4e2d\u592e\u533a\u5173\u952e\u7535\u6781\uff08\u8fd0\u52a8\u60f3\u8c61\u76f8\u5173\uff09\u8fdb\u884c\u53ef\u89c6\u5316
\npicks = [‘C3’, ‘C4’, ‘Cz’, ‘P3’, ‘P4′] # C3\/C4\u5bf9\u5e94\u5de6\u53f3\u8fd0\u52a8\u533a\uff0cCz\u4e3a\u5934\u9876\u4e2d\u592e
\nraw.plot(
\nduration=5, # \u663e\u793a\u524d5\u79d2
\nn_channels=len(picks),
\nscalings=dict(eeg=50e-6), # EEG\u4fe1\u53f7\u7f29\u653e\u4e3a50\u03bcV\uff08\u9002\u914d\u539f\u59cb\u4fe1\u53f7\u5e45\u5ea6\uff09
\ntitle=’\u539f\u59cbEEG\u4fe1\u53f7\uff08\u4e2d\u592e\u533a\u7535\u6781\uff0c\u524d5\u79d2\uff09’,
\nshow=True,
\nblock=True # \u963b\u585e\u7a0b\u5e8f\u76f4\u5230\u5173\u95ed\u56fe\u50cf\u7a97\u53e3
\n)<\/p>\n

# \u63d0\u53d6C3\u7535\u6781\u6570\u636e\uff0c\u7ed8\u5236\u524d20\u79d2\u8be6\u7ec6\u6ce2\u5f62
\nc3_data, c3_times = raw[‘C3’] # \u8fd4\u56de\uff08\u6570\u636e\u6570\u7ec4\uff0c\u65f6\u95f4\u6570\u7ec4\uff09
\nsample_num = int(20 * raw.info[‘sfreq’]) # 20\u79d2\u5bf9\u5e94\u7684\u91c7\u6837\u70b9\u6570<\/p>\n

plt.figure(figsize=(12, 4))
\nplt.plot(c3_times[:sample_num], c3_data[0, :sample_num], color=’#2E86AB’)
\nplt.title(‘C3\u7535\u6781\u539f\u59cbEEG\u4fe1\u53f7\uff08\u524d20\u79d2\uff09’, fontsize=12)
\nplt.xlabel(‘\u65f6\u95f4\uff08\u79d2\uff09’, fontsize=10)
\nplt.ylabel(‘\u7535\u538b\uff08\u03bcV\uff09’, fontsize=10)
\nplt.grid(True, alpha=0.3)
\nplt.tight_layout()
\nplt.show()
\n\u4e00\u952e\u83b7\u53d6\u5b8c\u6574\u9879\u76ee\u4ee3\u7801<\/p>\n

3. EEG \u4fe1\u53f7\u7684\u529f\u7387\u8c31\u5206\u6790
\n\u901a\u8fc7\u529f\u7387\u8c31\u5bc6\u5ea6\uff08PSD\uff09\u5206\u6790\u4e0d\u540c\u9891\u6bb5\u7684\u80fd\u91cf\u5206\u5e03\uff0c\u9a8c\u8bc1\u8111\u7535\u6ce2\u7279\u6027\uff1a<\/p>\n

python<\/p>\n

from mne.time_frequency import psd_array_welch<\/p>\n

# \u8ba1\u7b97\u5168\u901a\u9053\u529f\u7387\u8c31\u5bc6\u5ea6\uff08Welch\u65b9\u6cd5\uff0c\u5e38\u7528EEG\u9891\u57df\u5206\u6790\uff09
\npsds, freqs = psd_array_welch(
\nraw.get_data(), # \u539f\u59cbEEG\u6570\u636e\uff08\u901a\u9053\u6570\u00d7\u65f6\u95f4\u70b9\uff09
\nsfreq=raw.info[‘sfreq’], # \u91c7\u6837\u7387
\nfmin=0.5, fmax=45, # \u5206\u67900.5-45Hz\uff08\u8986\u76d6\u03b4-\u03b3\u6ce2\u6bb5\uff09
\nn_fft=1024, # FFT\u70b9\u6570\uff08\u5e73\u8861\u9891\u7387\u5206\u8fa8\u7387\u548c\u8ba1\u7b97\u901f\u5ea6\uff09
\naverage=’mean’ # \u65f6\u95f4\u5206\u6bb5\u5e73\u5747\uff0c\u964d\u4f4e\u566a\u58f0
\n)<\/p>\n

# \u9009\u62e9C3\uff08\u5de6\u624b\u8fd0\u52a8\u533a\uff09\u548cC4\uff08\u53f3\u624b\u8fd0\u52a8\u533a\uff09\u7535\u6781\u7ed8\u5236PSD
\nc3_idx = raw.ch_names.index(‘C3’)
\nc4_idx = raw.ch_names.index(‘C4′)<\/p>\n

plt.figure(figsize=(10, 5))
\n# \u5bf9\u6570\u5750\u6807\u663e\u793a\uff08\u66f4\u6e05\u6670\u5448\u73b0\u4f4e\u9891\u6bb5\u529f\u7387\u5dee\u5f02\uff09
\nplt.semilogy(freqs, psds[c3_idx], label=’C3\uff08\u5de6\u624b\u8fd0\u52a8\u533a\uff09’, color=’#A23B72′, alpha=0.8, linewidth=2)
\nplt.semilogy(freqs, psds[c4_idx], label=’C4\uff08\u53f3\u624b\u8fd0\u52a8\u533a\uff09’, color=’#F18F01′, alpha=0.8, linewidth=2)<\/p>\n

# \u6807\u6ce8\u5404\u9891\u6bb5\u533a\u57df
\nbands = [(0.5, 4, ‘\u03b4\u6ce2’, ‘red’), (4, 8, ‘\u03b8\u6ce2’, ‘orange’),
\n(8, 13, ‘\u03b1\u6ce2’, ‘yellow’), (13, 30, ‘\u03b2\u6ce2’, ‘green’),
\n(30, 45, ‘\u03b3\u6ce2’, ‘purple’)]
\nfor f1, f2, lab, col in bands:
\nplt.axvspan(f1, f2, alpha=0.1, color=col, label=lab)<\/p>\n

plt.title(‘C3\/C4\u7535\u6781\u529f\u7387\u8c31\u5bc6\u5ea6\uff08PSD\uff09’, fontsize=12)
\nplt.xlabel(‘\u9891\u7387\uff08Hz\uff09’, fontsize=10)
\nplt.ylabel(‘\u529f\u7387\u8c31\u5bc6\u5ea6\uff08\u03bcV\u00b2\/Hz\uff09’, fontsize=10)
\nplt.legend(loc=’upper right’, fontsize=9)
\nplt.grid(True, which=’both’, alpha=0.3)
\nplt.tight_layout()
\nplt.show()
\n\u4e00\u952e\u83b7\u53d6\u5b8c\u6574\u9879\u76ee\u4ee3\u7801<\/p>\n

4. \u4f2a\u8ff9\u8bc6\u522b\u4e0e\u5904\u7406\uff08ICA \u53bb\u9664\u773c\u7535\u4f2a\u8ff9\uff09
\n\u72ec\u7acb\u6210\u5206\u5206\u6790\uff08ICA\uff09\u662f EEG \u4f2a\u8ff9\u53bb\u9664\u7684\u5e38\u7528\u65b9\u6cd5\uff0c\u91cd\u70b9\u6f14\u793a\u773c\u7535\u4f2a\u8ff9\u7684\u8bc6\u522b\u4e0e\u53bb\u9664\uff1a<\/p>\n

python<\/p>\n

from mne.preprocessing import ICA<\/p>\n

# \u9884\u5904\u7406\uff1a\u6ee4\u6ce2\uff081Hz\u9ad8\u901a\uff0c\u53bb\u9664\u4f4e\u9891\u6f02\u79fb\uff0c\u63d0\u5347ICA\u6536\u655b\u6027\uff09
\nraw_filtered = raw.copy().filter(l_freq=1.0, h_freq=None)<\/p>\n

# \u521d\u59cb\u5316\u5e76\u62df\u5408ICA\u6a21\u578b
\nica = ICA(
\nn_components=15, # \u63d0\u53d615\u4e2a\u72ec\u7acb\u6210\u5206\uff08\u901a\u9053\u657032\uff0c\u53d6\u4e00\u534a\u5de6\u53f3\u907f\u514d\u8fc7\u62df\u5408\uff09
\nrandom_state=97, # \u56fa\u5b9a\u968f\u673a\u79cd\u5b50\uff0c\u786e\u4fdd\u7ed3\u679c\u53ef\u590d\u73b0
\nmax_iter=’auto’ # \u81ea\u52a8\u8c03\u6574\u8fed\u4ee3\u6b21\u6570\uff0c\u4fdd\u8bc1\u6536\u655b
\n)
\nica.fit(raw_filtered) # \u57fa\u4e8e\u6ee4\u6ce2\u540e\u7684\u6570\u636e\u8bad\u7ec3ICA<\/p>\n

# \u81ea\u52a8\u8bc6\u522b\u773c\u7535\u4f2a\u8ff9\u6210\u5206\uff08\u901a\u8fc7Fp1\/Fp2\u7535\u6781\u7684\u76f8\u5173\u6027\uff09
\neog_indices, eog_scores = ica.find_bads_eog(raw, ch_name=[‘Fp1’, ‘Fp2′])
\nprint(“\u81ea\u52a8\u8bc6\u522b\u7684\u773c\u7535\u4f2a\u8ff9\u6210\u5206\u7d22\u5f15:”, eog_indices)<\/p>\n

# \u53ef\u89c6\u5316\u773c\u7535\u6210\u5206\uff08\u62d3\u6251\u56fe+\u65f6\u95f4\u5e8f\u5217\uff09
\nif eog_indices:
\n# \u7ed8\u5236\u4f2a\u8ff9\u6210\u5206\u7684\u7a7a\u95f4\u62d3\u6251\u56fe
\nica.plot_components(
\npicks=eog_indices,
\ntitle=’\u773c\u7535\u4f2a\u8ff9\u5bf9\u5e94\u7684ICA\u6210\u5206\uff08\u7a7a\u95f4\u62d3\u6251\uff09’,
\nshow=True
\n)<\/p>\n

# \u7ed8\u5236\u4f2a\u8ff9\u6210\u5206\u7684\u65f6\u95f4\u5e8f\u5217\uff08\u89c2\u5bdf\u4e0e\u773c\u52a8\u7684\u540c\u6b65\u6027\uff09
\nica.plot_sources(
\nraw,
\npicks=eog_indices,
\ntitle=’\u773c\u7535\u4f2a\u8ff9\u6210\u5206\u7684\u65f6\u95f4\u5e8f\u5217’,
\nshow=True,
\nblock=True
\n)<\/p>\n

# \u5e94\u7528ICA\u53bb\u9664\u4f2a\u8ff9\uff08\u6392\u9664\u773c\u7535\u76f8\u5173\u6210\u5206\uff09
\nraw_cleaned = raw.copy()
\nica.apply(raw_cleaned, exclude=eog_indices)<\/p>\n

# \u5bf9\u6bd4\u5904\u7406\u524d\u540e\u7684Fp1\u7535\u6781\u4fe1\u53f7\uff08\u773c\u7535\u4f2a\u8ff9\u6700\u660e\u663e\uff09
\nfp1_idx = raw.ch_names.index(‘Fp1′)
\nfig, axes = plt.subplots(2, 1, figsize=(12, 6), sharex=True)<\/p>\n

# \u5904\u7406\u524d\u4fe1\u53f7\uff08\u542b\u773c\u7535\u4f2a\u8ff9\uff09
\naxes[0].plot(
\nraw.times[:3000],
\nraw.get_data()[fp1_idx, :3000],
\ncolor=’#E94B3C’,
\nalpha=0.8
\n)
\naxes[0].set_title(‘Fp1\u7535\u6781\u4fe1\u53f7\uff08ICA\u5904\u7406\u524d\uff0c\u542b\u773c\u7535\u4f2a\u8ff9\uff09’, fontsize=12)
\naxes[0].set_ylabel(‘\u7535\u538b\uff08\u03bcV\uff09’, fontsize=10)
\naxes[0].grid(True, alpha=0.3)<\/p>\n

# \u5904\u7406\u540e\u4fe1\u53f7\uff08\u53bb\u9664\u773c\u7535\u4f2a\u8ff9\uff09
\naxes[1].plot(
\nraw_cleaned.times[:3000],
\nraw_cleaned.get_data()[fp1_idx, :3000],
\ncolor=’#2ECC71′,
\nalpha=0.8
\n)
\naxes[1].set_title(‘Fp1\u7535\u6781\u4fe1\u53f7\uff08ICA\u53bb\u9664\u773c\u7535\u540e\uff09’, fontsize=12)
\naxes[1].set_xlabel(‘\u65f6\u95f4\uff08\u79d2\uff09’, fontsize=10)
\naxes[1].set_ylabel(‘\u7535\u538b\uff08\u03bcV\uff09’, fontsize=10)
\naxes[1].grid(True, alpha=0.3)<\/p>\n

plt.tight_layout()
\nplt.show()
\nelse:
\nprint(“\u672a\u81ea\u52a8\u8bc6\u522b\u5230\u660e\u786e\u773c\u7535\u6210\u5206\uff0c\u5efa\u8bae\u624b\u52a8\u7b5b\u9009\uff1a”)
\nica.plot_components(title=’ICA\u6210\u5206\u62d3\u6251\u56fe\uff08\u624b\u52a8\u9009\u62e9\u4f2a\u8ff9\u6210\u5206\uff09’, show=True)
\n# \u624b\u52a8\u9009\u62e9\u540e\uff0c\u901a\u8fc7 ica.apply(raw, exclude=[\u624b\u52a8\u7d22\u5f15]) \u53bb\u9664\u4f2a\u8ff9
\n\u4e00\u952e\u83b7\u53d6\u5b8c\u6574\u9879\u76ee\u4ee3\u7801<\/p>\n

5. \u8fd0\u52a8\u60f3\u8c61\u4e8b\u4ef6\u76f8\u5173\u7535\u4f4d\uff08ERP\uff09\u5206\u6790
\nERP \u662f\u5927\u8111\u5bf9\u7279\u5b9a\u4e8b\u4ef6\u7684\u540c\u6b65\u7535\u53cd\u5e94\uff0c\u901a\u8fc7\u5206\u6790\u8fd0\u52a8\u60f3\u8c61\u4efb\u52a1\u7684 ERP\uff0c\u9a8c\u8bc1\u8111\u7535\u4fe1\u53f7\u7684\u4efb\u52a1\u76f8\u5173\u6027\uff1a<\/p>\n

python<\/p>\n

# \u6b65\u9aa41\uff1a\u63d0\u53d6\u4e8b\u4ef6\u6807\u8bb0\uff08\u6570\u636e\u96c6\u6807\u6ce8\u7684\u4efb\u52a1\u89e6\u53d1\u4fe1\u53f7\uff09
\nevents, event_id = events_from_annotations(raw)
\nprint(“\u6570\u636e\u96c6\u4e2d\u7684\u4e8b\u4ef6\u7c7b\u578b:”, event_id)<\/p>\n

# \u7b5b\u9009\u8fd0\u52a8\u60f3\u8c61\u76f8\u5173\u4e8b\u4ef6\uff08T1=\u5de6\u624b\u60f3\u8c61\uff0cT2=\u53f3\u624b\u60f3\u8c61\uff0c\u67e5\u9605\u6570\u636e\u96c6\u6587\u6863\u786e\u8ba4\uff09
\nused_event_id = {‘\u5de6\u624b\u60f3\u8c61\uff08T1\uff09’: event_id[‘T1’], ‘\u53f3\u624b\u60f3\u8c61\uff08T2\uff09’: event_id[‘T2’]}<\/p>\n

# \u6b65\u9aa42\uff1a\u521b\u5efaEpochs\uff08\u6309\u4e8b\u4ef6\u5206\u6bb5\uff0c\u63d0\u53d6\u4e8b\u4ef6\u524d\u540e\u7684\u4fe1\u53f7\uff09
\nepochs = mne.Epochs(
\nraw_cleaned, # \u4f7f\u7528\u53bb\u4f2a\u8ff9\u540e\u7684\u5e72\u51c0\u6570\u636e
\nevents,
\nevent_id=used_event_id,
\ntmin=-1.0, tmax=4.0, # \u4e8b\u4ef6\u524d1\u79d2\uff08\u57fa\u7ebf\uff09\u5230\u4e8b\u4ef6\u540e4\u79d2\uff08\u60f3\u8c61\u671f\uff09
\nbaseline=(-1.0, 0), # \u7528\u4e8b\u4ef6\u524d1\u79d2\u505a\u57fa\u7ebf\u6821\u6b63\uff08\u6d88\u9664\u57fa\u7ebf\u6f02\u79fb\uff09
\npreload=True, # \u52a0\u8f7d\u6570\u636e\u5230\u5185\u5b58
\nreject_by_annotation=False # \u4e0d\u81ea\u52a8\u5254\u9664\u574f\u6bb5\uff08\u624b\u52a8\u540e\u7eed\u5904\u7406\uff09
\n)<\/p>\n

print(f”\\nEpochs\u5f62\u72b6\uff08\u8bd5\u6b21\u6570\u00d7\u901a\u9053\u6570\u00d7\u65f6\u95f4\u70b9\uff09:”, epochs.get_data().shape)
\nprint(f”\u5de6\u624b\u60f3\u8c61\u8bd5\u6b21\u6570:”, len(epochs[‘\u5de6\u624b\u60f3\u8c61\uff08T1\uff09’]))
\nprint(f”\u53f3\u624b\u60f3\u8c61\u8bd5\u6b21\u6570:”, len(epochs[‘\u53f3\u624b\u60f3\u8c61\uff08T2\uff09’]))<\/p>\n

# \u6b65\u9aa43\uff1a\u8ba1\u7b97\u5e73\u5747ERP\uff08\u5404\u6761\u4ef6\u4e0b\u7684\u540c\u6b65\u7535\u53cd\u5e94\uff09
\nleft_erp = epochs[‘\u5de6\u624b\u60f3\u8c61\uff08T1\uff09’].average() # \u5de6\u624b\u60f3\u8c61ERP
\nright_erp = epochs[‘\u53f3\u624b\u60f3\u8c61\uff08T2\uff09’].average() # \u53f3\u624b\u60f3\u8c61ERP<\/p>\n

# \u6b65\u9aa44\uff1a\u53ef\u89c6\u5316ERP\u5bf9\u6bd4\uff08C3\/C4\u7535\u6781\uff0c\u8fd0\u52a8\u60f3\u8c61\u6838\u5fc3\u901a\u9053\uff09
\nc3_idx = epochs.ch_names.index(‘C3’)
\nc4_idx = epochs.ch_names.index(‘C4′)<\/p>\n

fig, axes = plt.subplots(1, 2, figsize=(14, 5), sharey=True)<\/p>\n

# \u5de6\u56fe\uff1aC3\u7535\u6781\uff08\u5bf9\u5e94\u5de6\u624b\u8fd0\u52a8\u533a\uff0c\u5bf9\u4fa7\u6fc0\u6d3b\uff09
\naxes[0].plot(
\nleft_erp.times, left_erp.data[c3_idx],
\nlabel=’\u5de6\u624b\u60f3\u8c61’, color=’#3498DB’, linewidth=2
\n)
\naxes[0].plot(
\nright_erp.times, right_erp.data[c3_idx],
\nlabel=’\u53f3\u624b\u60f3\u8c61’, color=’#E74C3C’, linewidth=2
\n)
\naxes[0].axvline(x=0, color=’black’, linestyle=’–‘, alpha=0.5, label=’\u60f3\u8c61\u5f00\u59cb’)
\naxes[0].set_title(‘C3\u7535\u6781 – \u8fd0\u52a8\u60f3\u8c61ERP\u5bf9\u6bd4’, fontsize=12)
\naxes[0].set_xlabel(‘\u65f6\u95f4\uff08\u79d2\uff09’, fontsize=10)
\naxes[0].set_ylabel(‘\u7535\u538b\uff08\u03bcV\uff09’, fontsize=10)
\naxes[0].legend(fontsize=9)
\naxes[0].grid(True, alpha=0.3)<\/p>\n

# \u53f3\u56fe\uff1aC4\u7535\u6781\uff08\u5bf9\u5e94\u53f3\u624b\u8fd0\u52a8\u533a\uff0c\u5bf9\u4fa7\u6fc0\u6d3b\uff09
\naxes[1].plot(
\nleft_erp.times, left_erp.data[c4_idx],
\nlabel=’\u5de6\u624b\u60f3\u8c61’, color=’#3498DB’, linewidth=2
\n)
\naxes[1].plot(
\nright_erp.times, right_erp.data[c4_idx],
\nlabel=’\u53f3\u624b\u60f3\u8c61’, color=’#E74C3C’, linewidth=2
\n)
\naxes[1].axvline(x=0, color=’black’, linestyle=’–‘, alpha=0.5, label=’\u60f3\u8c61\u5f00\u59cb’)
\naxes[1].set_title(‘C4\u7535\u6781 – \u8fd0\u52a8\u60f3\u8c61ERP\u5bf9\u6bd4’, fontsize=12)
\naxes[1].set_xlabel(‘\u65f6\u95f4\uff08\u79d2\uff09’, fontsize=10)
\naxes[1].legend(fontsize=9)
\naxes[1].grid(True, alpha=0.3)<\/p>\n

plt.suptitle(‘\u8fd0\u52a8\u60f3\u8c61ERP\u7684\u5bf9\u4fa7\u6fc0\u6d3b\u6548\u5e94\uff08C3\u2192\u5de6\u624b\uff0cC4\u2192\u53f3\u624b\uff09’, fontsize=14)
\nplt.tight_layout()
\nplt.show()<\/p>\n

# \u6b65\u9aa45\uff1aERP\u5dee\u5f02\u6ce2\u62d3\u6251\u56fe\uff08\u76f4\u89c2\u89c2\u5bdf\u5934\u76ae\u6fc0\u6d3b\u5dee\u5f02\uff09
\n# \u8ba1\u7b97\u5dee\u5f02\u6ce2\uff1a\u5de6\u624b\u60f3\u8c61 – \u53f3\u624b\u60f3\u8c61
\ndiff_erp = mne.combine_evoked([left_erp, right_erp], weights=[1, -1])
\ndiff_erp.plot_joint(
\ntitle=’ERP\u5dee\u5f02\u6ce2\uff08\u5de6\u624b-\u53f3\u624b\uff09\uff1a\u5934\u76ae\u6fc0\u6d3b\u62d3\u6251’,
\ntimes=[0.5, 1.0, 1.5, 2.0], # \u5173\u952e\u65f6\u95f4\u70b9\uff08\u60f3\u8c61\u6267\u884c\u671f\uff09
\nshow=True
\n)
\n\u4e00\u952e\u83b7\u53d6\u5b8c\u6574\u9879\u76ee\u4ee3\u7801<\/p>\n

\u516d\u3001\u603b\u7ed3\u4e0e\u5c55\u671b
\nEEG \u4fe1\u53f7\u7684\u91c7\u96c6\u662f BCI \u7cfb\u7edf\u7684 \u201c\u6e90\u5934\u5de5\u7a0b\u201d\uff0c\u5176\u8d28\u91cf\u76f4\u63a5\u51b3\u5b9a\u540e\u7eed\u7279\u5f81\u63d0\u53d6\u3001\u610f\u56fe\u89e3\u7801\u7684\u6548\u679c\u3002\u672c\u6587\u4ece\u4ea7\u751f\u673a\u5236\uff08\u795e\u7ecf\u57fa\u7840\uff09\u2192 \u91c7\u96c6\u539f\u7406\uff08\u4fe1\u53f7\u8f6c\u5316\uff09\u2192 \u786c\u4ef6\u7ec4\u6210\uff08\u8bbe\u5907\u652f\u6491\uff09\u2192 \u6570\u636e\u7279\u6027\uff08\u6838\u5fc3\u6311\u6218\uff09 \u56db\u4e2a\u7ef4\u5ea6\uff0c\u7cfb\u7edf\u89e3\u6790\u4e86 EEG \u6570\u636e\u7684\u6e90\u5934\u903b\u8f91\uff0c\u5e76\u901a\u8fc7\u53ef\u590d\u73b0\u7684 Python \u4ee3\u7801\uff0c\u5c55\u793a\u4e86\u539f\u59cb\u6570\u636e\u7684\u5904\u7406\u4e0e\u7279\u6027\u9a8c\u8bc1\u65b9\u6cd5\u3002<\/p>\n

\u6838\u5fc3\u7ed3\u8bba
\nEEG \u4fe1\u53f7\u7684\u672c\u8d28\u662f\u5927\u91cf\u795e\u7ecf\u5143\u540c\u6b65\u6d3b\u52a8\u7684\u53e0\u52a0\u7535\u4f4d\u5dee\uff0c\u5176\u9891\u6bb5\u5206\u5e03\u4e0e\u5927\u8111\u72b6\u6001\u5f3a\u76f8\u5173\uff0c\u662f BCI \u89e3\u7801\u7684\u6838\u5fc3\u4f9d\u636e\uff1b
\n\u91c7\u96c6\u73af\u8282\u7684\u5173\u952e\u6307\u6807\uff08\u7535\u6781\u963b\u6297\u3001\u653e\u5927\u5668 CMRR\u3001ADC \u5206\u8fa8\u7387\uff09\u76f4\u63a5\u5f71\u54cd\u6570\u636e\u8d28\u91cf\uff0c\u9700\u4f18\u5148\u4fdd\u969c\u786c\u4ef6\u6027\u80fd\uff1b
\n\u4f4e SNR\u3001\u975e\u5e73\u7a33\u6027\u3001\u4e2a\u4f53\u5dee\u5f02\u3001\u4f2a\u8ff9\u5e72\u6270\u662f EEG \u6570\u636e\u7684\u6838\u5fc3\u6311\u6218\uff0c\u9700\u901a\u8fc7\u9884\u5904\u7406\uff08\u6ee4\u6ce2\u3001ICA\uff09\u548c\u4e2a\u6027\u5316\u6821\u51c6\u7f13\u89e3\u3002
\n\u672a\u6765\u53d1\u5c55\u65b9\u5411
\n\u5e72\u7535\u6781\u6280\u672f\u7a81\u7834\uff1a\u7ed3\u5408\u4e3b\u52a8\u5c4f\u853d\u3001\u963b\u6297\u81ea\u9002\u5e94\u7b97\u6cd5\uff0c\u63a8\u52a8\u5e72\u7535\u6781\u6027\u80fd\u63a5\u8fd1\u6e7f\u7535\u6781\uff0c\u540c\u65f6\u964d\u4f4e\u6210\u672c\u548c\u4f7f\u7528\u95e8\u69db\uff1b
\n\u9ad8\u5bc6\u5ea6 EEG \u7cfb\u7edf\uff1a256\/512 \u901a\u9053\u7cfb\u7edf\u5c06\u7a7a\u95f4\u5206\u8fa8\u7387\u63d0\u5347\u81f3\u6beb\u7c73\u7ea7\uff0c\u914d\u5408\u6e90\u5b9a\u4f4d\u7b97\u6cd5\uff0c\u5b9e\u73b0\u66f4\u7cbe\u51c6\u7684\u5927\u8111\u6d3b\u52a8\u5b9a\u4f4d\uff1b
\n\u4fbf\u643a\u53ef\u7a7f\u6234\u5316\uff1a\u4f4e\u529f\u8017 ASIC \u82af\u7247\u3001\u65e0\u7ebf\u4f20\u8f93\u6280\u672f\uff08BLE 5.0\/UWB\uff09\u652f\u6301\u65e5\u5e38\u573a\u666f\u957f\u671f\u76d1\u6d4b\uff0c\u62d3\u5c55 BCI \u5e94\u7528\u8fb9\u754c\uff1b
\n\u591a\u6a21\u6001\u878d\u5408\uff1aEEG \u4e0e\u529f\u80fd\u6027\u8fd1\u7ea2\u5916\u5149\u8c31\uff08fNIRS\uff09\u3001\u773c\u52a8\uff08EOG\uff09\u3001\u808c\u7535\uff08EMG\uff09\u878d\u5408\uff0c\u4e92\u8865\u65f6\u7a7a\u5206\u8fa8\u7387\uff0c\u63d0\u5347\u7cfb\u7edf\u9c81\u68d2\u6027\u3002<\/p>\n

\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014
\n\u7248\u6743\u58f0\u660e\uff1a\u672c\u6587\u4e3aCSDN\u535a\u4e3b\u300cBrduino\u8111\u673a\u63a5\u53e3\u6280\u672f\u7b54\u7591\u300d\u7684\u539f\u521b\u6587\u7ae0\uff0c\u9075\u5faaCC 4.0 BY-SA\u7248\u6743\u534f\u8bae\uff0c\u8f6c\u8f7d\u8bf7\u9644\u4e0a\u539f\u6587\u51fa\u5904\u94fe\u63a5\u53ca\u672c\u58f0\u660e\u3002
\n\u539f\u6587\u94fe\u63a5\uff1ahttps:\/\/blog.csdn.net\/m0_63827302\/article\/details\/155277321<\/p>\n

<\/div>\n

<\/path><\/svg><\/i> \"Loading\"<\/p>\n

<\/div>\n","protected":false},"excerpt":{"rendered":"

\u8111\u673a\u63a5\u53e3\uff08BCI\uff09\u7cfb\u7edf\u7684\u6838\u5fc3\u662f\u5bf9\u5927\u8111\u6d3b\u52a8\u4fe1\u53f7\u7684\u6709\u6548\u91c7\u96c6\u4e0e\u89e3\u8bfb\uff0c\u800c\u8111\u7535\u56fe\uff08EEG\uff09\u4f5c\u4e3a\u6700\u5e38\u7528\u7684\u975e\u4fb5\u5165\u5f0f\u8111\u4fe1\u53f7\u91c7\u96c6 […]<\/p>\n

<\/div>\n

<\/path><\/svg><\/i> \"Loading\"<\/p>\n

<\/div>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11,21,14],"tags":[],"class_list":["post-92","post","type-post","status-publish","format-standard","hentry","category-bci","category-bci_apple","category-eeg"],"_links":{"self":[{"href":"https:\/\/qdszxyy.com\/index.php?rest_route=\/wp\/v2\/posts\/92","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/qdszxyy.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/qdszxyy.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/qdszxyy.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/qdszxyy.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=92"}],"version-history":[{"count":1,"href":"https:\/\/qdszxyy.com\/index.php?rest_route=\/wp\/v2\/posts\/92\/revisions"}],"predecessor-version":[{"id":93,"href":"https:\/\/qdszxyy.com\/index.php?rest_route=\/wp\/v2\/posts\/92\/revisions\/93"}],"wp:attachment":[{"href":"https:\/\/qdszxyy.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=92"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/qdszxyy.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=92"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/qdszxyy.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=92"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}