playground_WM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e58e5d10-a56c-44be-91c8-f0263cb93cc2",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "# Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "22f15b72-d26f-4e33-a01e-6535d2a20537",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Logging and warnings\n",
    "import logging\n",
    "\n",
    "# File R/W\n",
    "import os\n",
    "\n",
    "# Additional Inbuilt Utilities\n",
    "import itertools\n",
    "import functools\n",
    "\n",
    "# Regex\n",
    "import re\n",
    "\n",
    "# Array\n",
    "import numpy as np\n",
    "\n",
    "# Table\n",
    "import pandas as pd\n",
    "\n",
    "# Advanced Computations\n",
    "import scipy as sc\n",
    "\n",
    "# Statistics\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.stats as st\n",
    "\n",
    "# Neurophysiological Data Analysis\n",
    "import mne\n",
    "import mne_nirs # esp. for fNIRS\n",
    "\n",
    "# Neuroimaging Statistical Tools\n",
    "import nilearn\n",
    "import nilearn.plotting\n",
    "\n",
    "# Plotting\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Vistual Display\n",
    "from pyvirtualdisplay import Display"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8b8c124c-8401-4761-8854-854099ba2352",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Autoreload Custom Modules\n",
    "%load_ext autoreload\n",
    "%autoreload 1\n",
    "\n",
    "# Custom Class\n",
    "%aimport nirs\n",
    "NIRS = nirs.NIRS\n",
    "\n",
    "# Custom Settings\n",
    "%aimport constants\n",
    "\n",
    "# Custom Functions\n",
    "%aimport utils\n",
    "\n",
    "# Modified MBLL\n",
    "%aimport mbll"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fee37a86-6ff6-4f58-abcf-98002ad5cc6c",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## Config"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "83055f20-573e-46f6-9747-d5011a4f6a65",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set logging level for custom modules\n",
    "logging.getLogger().setLevel(logging.INFO)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "635e9279-25b2-4f38-9129-e83aa38f4ed0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set logging level for MNE\n",
    "mne.set_log_level('WARNING')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "625b59f2-363a-457f-9264-8d357a1426b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set plotting backend for MNE\n",
    "mne.viz.set_browser_backend('matplotlib') # qt, matplotlib\n",
    "mne.viz.set_3d_backend('notebook') # pyvistaqt, notebook"
   ]
  },
  {
   "cell_type": "raw",
   "id": "44bfc7c6-f20a-48df-9807-aa812417c60a",
   "metadata": {},
   "source": [
    "# # Set recommended (by MNE) plotting backend\n",
    "# %matplotlib qt"
   ]
  },
  {
   "cell_type": "raw",
   "id": "474ba645-f9c6-44d4-b295-5431724c4128",
   "metadata": {},
   "source": [
    "# # Setup and start xvfb virtual display\n",
    "# disp = Display()\n",
    "# disp.start()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f746fc3-9fcb-4f87-a09a-85bd028c5d6e",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "# MAIN"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "583dfc5c-b818-4b14-aa28-f09febbd6f1f",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "##### `Pre-Processing`"
   ]
  },
  {
   "cell_type": "raw",
   "id": "9f29fc48-a21b-4972-8163-bba007022c57",
   "metadata": {},
   "source": [
    "# Instantiate NIRS() and read data\n",
    "nirs = NIRS(project='Working-Memory') # Working-Memory # Multi-object Tracking\n",
    "nirs.read(1, 1, 1, augment=False, transform=False, correction_factor=0.779);"
   ]
  },
  {
   "cell_type": "raw",
   "id": "14ce3844-821e-4785-9253-d7f21017f104",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Pre-process the data using the default pipeline.\n",
    "# Output is a dictionary of intermediate mne.Raw instances.\n",
    "# Stages: CW     (raw signal)\n",
    "#         CWx    (backlight removed raw signal)\n",
    "#         OD     (optical density)\n",
    "#         TDDR   (motion artifact removal)\n",
    "#         AP     (autopick channels)\n",
    "#         SSR    (short-channel regression)\n",
    "#         HB     (chromophore/haemoglobin)\n",
    "#         LS     (pick long channels, after saving short channels)\n",
    "#         FL     (bandpass filtering)\n",
    "#         NCE    (negative correlation improvement)\n",
    "savepoints = nirs.default_pipeline(remove_backlight=True,\n",
    "                                   tddr=True,\n",
    "                                   scalp_coupling_index=True,\n",
    "                                   autopick_channels=True,\n",
    "                                   short_channel_regression=True,\n",
    "                                   bandpass=True,\n",
    "                                   negative_correlation_enhancement=False,\n",
    "                                   threshold_heart_rate=0.5,\n",
    "                                   show_failed=False,\n",
    "                                   l_trans_bandwidth=0.007,\n",
    "                                   h_trans_bandwidth=0.007)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "179bfb7f-4546-4270-90df-56518336a597",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/pulkitgoyal56/fNIRS-attentional-load/nirs.py:531: RuntimeWarning: Omitted 16 annotation(s) that were outside data range.\n",
      "  self.raw.set_annotations(mne.Annotations(\n",
      "/home/pulkitgoyal56/fNIRS-attentional-load/nirs.py:531: RuntimeWarning: Limited 1 annotation(s) that were expanding outside the data range.\n",
      "  self.raw.set_annotations(mne.Annotations(\n"
     ]
    }
   ],
   "source": [
    "nirs_1 = NIRS(project='Working-Memory').read(1, 1, 1, augment=True, transform=False, correction_factor=0.779)\n",
    "nirs_2 = NIRS(project='Working-Memory').read(1, 1, 2, augment=True, transform=False, correction_factor=0.779)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d0cb4fde-31be-4373-aa0f-c46c8a794751",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/pulkitgoyal56/fNIRS-attentional-load/nirs.py:621: RuntimeWarning: Negative intensities encountered. Setting to abs(x)\n",
      "  return func(self.raw, *args, **kwargs)\n",
      "WARNING:root:Could not fit Gaussian curve for channel S40_D33 774. Discarding...\n",
      "/home/pulkitgoyal56/fNIRS-attentional-load/nirs.py:621: RuntimeWarning: Negative intensities encountered. Setting to abs(x)\n",
      "  return func(self.raw, *args, **kwargs)\n",
      "WARNING:root:Could not fit Gaussian curve for channel S33_D41 774. Discarding...\n",
      "WARNING:root:Could not fit Gaussian curve for channel S34_D39 774. Discarding...\n",
      "WARNING:root:Could not fit Gaussian curve for channel S40_D33 865. Discarding...\n"
     ]
    }
   ],
   "source": [
    "savepoints_1 = nirs_1.default_pipeline(remove_backlight=True, tddr=True, autopick_channels=True, short_channel_regression=True, bandpass=True, negative_correlation_enhancement=False, l_trans_bandwidth=0.007, h_trans_bandwidth=0.007, threshold_heart_rate=0.5)\n",
    "savepoints_2 = nirs_2.default_pipeline(remove_backlight=True, tddr=True, autopick_channels=True, short_channel_regression=True, bandpass=True, negative_correlation_enhancement=False, l_trans_bandwidth=0.007, h_trans_bandwidth=0.007, threshold_heart_rate=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1df298b4-792c-43fd-a85e-47cd911e822b",
   "metadata": {},
   "outputs": [],
   "source": [
    "nirs_1.get_epochs(tmin=-5, tmax=15, baseline=(-0.1, 0))\n",
    "nirs_2.get_epochs(tmin=-5, tmax=15, baseline=(-0.1, 0));\n",
    "\n",
    "# Drop epochs where response was not correct\n",
    "# nirs_1.epochs.drop(nirs_1.mat['id_correct'] != 1)\n",
    "# nirs_2.epochs.drop(nirs_2.mat['id_correct'] != 1)\n",
    "\n",
    "nirs_1.block_average();\n",
    "nirs_2.block_average();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ba5a83d6-8d22-40e4-878a-1e41549f2c1a",
   "metadata": {},
   "outputs": [],
   "source": [
    "nirs = nirs_1 + nirs_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0970450f-cd38-4297-979b-6431b89ef0ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "raw_haemo = nirs.raw.copy().pick(nirs.good_ch_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1d5ecaf7-2ffb-4bbb-9171-3d41e4095b58",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# nirs.get_epochs(tmin=-5, # -(nirs.DUR['wait'])\n",
    "#                 tmax=15, # (nirs.DUR['motion'] + nirs.DUR['probe'] + nirs.DUR['feedb'])\n",
    "#                 baseline=(-0.1, 0));\n",
    "\n",
    "# Drop epochs where response was not correct\n",
    "# nirs.epochs.drop(pd.concat(nirs.mat)['id_correct'] != 1);\n",
    "\n",
    "nirs.block_average();"
   ]
  },
  {
   "cell_type": "raw",
   "id": "c38728d2-8b77-4ae8-96a7-00087fe4ba3f",
   "metadata": {},
   "source": [
    "nirs.raw.plot_sensors(ch_type='hbo', show_names=utils.filter_channels('S(\\d+)_D(\\d+) (.+)', lambda s, d, t: s > d and t == 'hbo', nirs.raw.ch_names)[0]);"
   ]
  },
  {
   "cell_type": "raw",
   "id": "b67d8c18-4c44-4e4e-8e62-d5a5376f3dbc",
   "metadata": {},
   "source": [
    "mne.viz.plot_epochs_image(nirs.epochs, picks=nirs.goof_ch_names, sigma=.5, vmin=-10, vmax=10, show=True)"
   ]
  },
  {
   "cell_type": "raw",
   "id": "2c2ea2db-2d9a-43bb-a24a-323e88473f04",
   "metadata": {},
   "source": [
    "mne.viz.plot_compare_evokeds(nirs.evoked_dict, picks=nirs.good_ch_names[-1], combine='mean', ci=0.95, axes=plt.figure(figsize=(16, 4)).add_subplot(111),\n",
    "                             colors={'0': 'k', '2': 'r', '3':'g', '4':'b', '5':'y'},\n",
    "                             linestyles={'hbo': '-',\n",
    "                                         'hbr': ':'},\n",
    "                             ylim={'hbo': [-5, 5]},\n",
    "                             title='Signal Averages Across Trials and Channels for Number of Targets');"
   ]
  },
  {
   "cell_type": "raw",
   "id": "27b6ab5b-bcf4-4646-8235-27ea13337965",
   "metadata": {},
   "source": [
    "nirs.epochs['0'].average(picks='hbo').plot_joint(times=np.arange(nirs.T_EPOCH_START, nirs.T_EPOCH_END, 3), topomap_args={'extrapolate': 'local'});"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d44ad10-da17-4322-9518-b890037d261f",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "##### `GLM Analysis`\n",
    "\n",
    "... GLM Analysis using inbuilt MNE functions ...\n",
    "> Find significant channels and their 𝛽 values for the different load conditions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "32db1336-1692-4cc8-85a0-495d50361322",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "design_matrix = mne_nirs.experimental_design.make_first_level_design_matrix(raw_haemo,\n",
    "                                                                            hrf_model= 'spm', # 'spm + derivative + dispersion',\n",
    "                                                                            drift_model=None, # 'cosine', 'polynomial'\n",
    "                                                                            # drift_order=3,\n",
    "                                                                            # high_pass=1/2/(nirs.DUR['trial']),\n",
    "                                                                            stim_dur=nirs.DUR['motion'])\n",
    "\n",
    "# design_matrix['mean_ss_hbo'] = np.mean(nirs.raw_ss.copy().pick('hbo').get_data(), axis=0)\n",
    "# design_matrix['mean_ss_hbr'] = np.mean(nirs.raw_ss.copy().pick('hbr').get_data(), axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0ce1ae9b-1ca2-4f9c-9a90-f8d5d05607a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "glm_est = mne_nirs.statistics.run_glm(raw_haemo, design_matrix)\n",
    "glm_results = glm_est.to_dataframe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a8900b59-90c7-4f11-84c7-c8c11cabba10",
   "metadata": {},
   "outputs": [],
   "source": [
    "glm_results['Significant'] = st.multitest.multipletests(glm_results['p_value'])[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d8143c5f-6b86-4af5-ae64-5b68fa62e753",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>variable</th>\n",
       "      <th>Condition</th>\n",
       "      <th>df</th>\n",
       "      <th>mse</th>\n",
       "      <th>p_value</th>\n",
       "      <th>se</th>\n",
       "      <th>t</th>\n",
       "      <th>theta</th>\n",
       "      <th>Source</th>\n",
       "      <th>Detector</th>\n",
       "      <th>Chroma</th>\n",
       "      <th>Significant</th>\n",
       "      <th>ch_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.888109e-14</td>\n",
       "      <td>0.000144</td>\n",
       "      <td>9.661449e-08</td>\n",
       "      <td>-24.807393</td>\n",
       "      <td>-0.000002</td>\n",
       "      <td>33</td>\n",
       "      <td>40</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S33_D40 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>9.030607e-15</td>\n",
       "      <td>0.000308</td>\n",
       "      <td>6.681703e-08</td>\n",
       "      <td>-19.201676</td>\n",
       "      <td>-0.000001</td>\n",
       "      <td>33</td>\n",
       "      <td>43</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S33_D43 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.417735e-14</td>\n",
       "      <td>0.000177</td>\n",
       "      <td>1.477844e-07</td>\n",
       "      <td>-23.109937</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>34</td>\n",
       "      <td>41</td>\n",
       "      <td>hbr</td>\n",
       "      <td>True</td>\n",
       "      <td>S34_D41 hbr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.328401e-14</td>\n",
       "      <td>0.000145</td>\n",
       "      <td>1.072896e-07</td>\n",
       "      <td>-24.715390</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>34</td>\n",
       "      <td>43</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S34_D43 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.010884e-14</td>\n",
       "      <td>0.000286</td>\n",
       "      <td>7.069347e-08</td>\n",
       "      <td>19.694390</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>34</td>\n",
       "      <td>43</td>\n",
       "      <td>hbr</td>\n",
       "      <td>True</td>\n",
       "      <td>S34_D43 hbr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.642304e-14</td>\n",
       "      <td>0.000235</td>\n",
       "      <td>9.010635e-08</td>\n",
       "      <td>-21.038401</td>\n",
       "      <td>-0.000002</td>\n",
       "      <td>39</td>\n",
       "      <td>40</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S39_D40 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.534897e-14</td>\n",
       "      <td>0.000295</td>\n",
       "      <td>8.711006e-08</td>\n",
       "      <td>-19.500200</td>\n",
       "      <td>-0.000002</td>\n",
       "      <td>39</td>\n",
       "      <td>40</td>\n",
       "      <td>hbr</td>\n",
       "      <td>True</td>\n",
       "      <td>S39_D40 hbr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84</th>\n",
       "      <td>HIGH</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.138816e-14</td>\n",
       "      <td>0.000288</td>\n",
       "      <td>1.193212e-07</td>\n",
       "      <td>-19.652521</td>\n",
       "      <td>-0.000002</td>\n",
       "      <td>39</td>\n",
       "      <td>43</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S39_D43 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.138816e-14</td>\n",
       "      <td>0.000093</td>\n",
       "      <td>1.028289e-07</td>\n",
       "      <td>-28.650962</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>39</td>\n",
       "      <td>43</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S39_D43 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.947960e-14</td>\n",
       "      <td>0.000104</td>\n",
       "      <td>9.813381e-08</td>\n",
       "      <td>-27.618208</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>40</td>\n",
       "      <td>33</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S40_D33 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>112</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.539234e-14</td>\n",
       "      <td>0.000243</td>\n",
       "      <td>8.723304e-08</td>\n",
       "      <td>-20.806006</td>\n",
       "      <td>-0.000002</td>\n",
       "      <td>40</td>\n",
       "      <td>39</td>\n",
       "      <td>hbr</td>\n",
       "      <td>True</td>\n",
       "      <td>S40_D39 hbr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.067800e-14</td>\n",
       "      <td>0.000121</td>\n",
       "      <td>1.011074e-07</td>\n",
       "      <td>-26.291800</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>40</td>\n",
       "      <td>44</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S40_D44 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.273030e-14</td>\n",
       "      <td>0.000255</td>\n",
       "      <td>1.272048e-07</td>\n",
       "      <td>-20.454707</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>41</td>\n",
       "      <td>33</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S41_D33 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>HIGH</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.204616e-14</td>\n",
       "      <td>0.000296</td>\n",
       "      <td>8.954787e-08</td>\n",
       "      <td>-19.475999</td>\n",
       "      <td>-0.000002</td>\n",
       "      <td>41</td>\n",
       "      <td>33</td>\n",
       "      <td>hbr</td>\n",
       "      <td>True</td>\n",
       "      <td>S41_D33 hbr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>8.826122e-15</td>\n",
       "      <td>0.000203</td>\n",
       "      <td>6.605621e-08</td>\n",
       "      <td>-22.108951</td>\n",
       "      <td>-0.000001</td>\n",
       "      <td>43</td>\n",
       "      <td>33</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S43_D33 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>HIGH</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.758715e-14</td>\n",
       "      <td>0.000232</td>\n",
       "      <td>1.355140e-07</td>\n",
       "      <td>-21.124936</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>43</td>\n",
       "      <td>39</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S43_D39 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.758715e-14</td>\n",
       "      <td>0.000194</td>\n",
       "      <td>1.167837e-07</td>\n",
       "      <td>-22.437544</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>43</td>\n",
       "      <td>39</td>\n",
       "      <td>hbo</td>\n",
       "      <td>True</td>\n",
       "      <td>S43_D39 hbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>184</th>\n",
       "      <td>LOW</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.013413e-14</td>\n",
       "      <td>0.000099</td>\n",
       "      <td>7.078185e-08</td>\n",
       "      <td>-28.099343</td>\n",
       "      <td>-0.000002</td>\n",
       "      <td>44</td>\n",
       "      <td>33</td>\n",
       "      <td>hbr</td>\n",
       "      <td>True</td>\n",
       "      <td>S44_D33 hbr</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "variable Condition   df           mse   p_value            se          t  \\\n",
       "7              LOW  3.0  1.888109e-14  0.000144  9.661449e-08 -24.807393   \n",
       "19             LOW  3.0  9.030607e-15  0.000308  6.681703e-08 -19.201676   \n",
       "52             LOW  3.0  4.417735e-14  0.000177  1.477844e-07 -23.109937   \n",
       "55             LOW  3.0  2.328401e-14  0.000145  1.072896e-07 -24.715390   \n",
       "58             LOW  3.0  1.010884e-14  0.000286  7.069347e-08  19.694390   \n",
       "79             LOW  3.0  1.642304e-14  0.000235  9.010635e-08 -21.038401   \n",
       "82             LOW  3.0  1.534897e-14  0.000295  8.711006e-08 -19.500200   \n",
       "84            HIGH  3.0  2.138816e-14  0.000288  1.193212e-07 -19.652521   \n",
       "85             LOW  3.0  2.138816e-14  0.000093  1.028289e-07 -28.650962   \n",
       "97             LOW  3.0  1.947960e-14  0.000104  9.813381e-08 -27.618208   \n",
       "112            LOW  3.0  1.539234e-14  0.000243  8.723304e-08 -20.806006   \n",
       "127            LOW  3.0  2.067800e-14  0.000121  1.011074e-07 -26.291800   \n",
       "133            LOW  3.0  3.273030e-14  0.000255  1.272048e-07 -20.454707   \n",
       "135           HIGH  3.0  1.204616e-14  0.000296  8.954787e-08 -19.475999   \n",
       "157            LOW  3.0  8.826122e-15  0.000203  6.605621e-08 -22.108951   \n",
       "168           HIGH  3.0  2.758715e-14  0.000232  1.355140e-07 -21.124936   \n",
       "169            LOW  3.0  2.758715e-14  0.000194  1.167837e-07 -22.437544   \n",
       "184            LOW  3.0  1.013413e-14  0.000099  7.078185e-08 -28.099343   \n",
       "\n",
       "variable     theta Source Detector Chroma  Significant      ch_name  \n",
       "7        -0.000002     33       40    hbo         True  S33_D40 hbo  \n",
       "19       -0.000001     33       43    hbo         True  S33_D43 hbo  \n",
       "52       -0.000003     34       41    hbr         True  S34_D41 hbr  \n",
       "55       -0.000003     34       43    hbo         True  S34_D43 hbo  \n",
       "58        0.000001     34       43    hbr         True  S34_D43 hbr  \n",
       "79       -0.000002     39       40    hbo         True  S39_D40 hbo  \n",
       "82       -0.000002     39       40    hbr         True  S39_D40 hbr  \n",
       "84       -0.000002     39       43    hbo         True  S39_D43 hbo  \n",
       "85       -0.000003     39       43    hbo         True  S39_D43 hbo  \n",
       "97       -0.000003     40       33    hbo         True  S40_D33 hbo  \n",
       "112      -0.000002     40       39    hbr         True  S40_D39 hbr  \n",
       "127      -0.000003     40       44    hbo         True  S40_D44 hbo  \n",
       "133      -0.000003     41       33    hbo         True  S41_D33 hbo  \n",
       "135      -0.000002     41       33    hbr         True  S41_D33 hbr  \n",
       "157      -0.000001     43       33    hbo         True  S43_D33 hbo  \n",
       "168      -0.000003     43       39    hbo         True  S43_D39 hbo  \n",
       "169      -0.000003     43       39    hbo         True  S43_D39 hbo  \n",
       "184      -0.000002     44       33    hbr         True  S44_D33 hbr  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "glm_results_significant = glm_results.loc[glm_results['Significant']]\n",
    "glm_results_significant[glm_results_significant['Condition'].isin(nirs.cases)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4ea95f97-3bec-417e-bb52-3bb7f17a18cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "significant_channels = list(glm_results_significant[\n",
    "    glm_results_significant['Condition'].isin(nirs.cases)\n",
    "    # & (((glm_results_significant['Chroma'] == 'hbo') & (glm_results_significant['theta'] < 0))\n",
    "    #  | ((glm_results_significant['Chroma'] == 'hbr') & (glm_results_significant['theta'] > 0)))\n",
    "]['ch_name'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "aa66aebc-8313-4d9c-a12d-175dfddf69a0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x600 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nirs.plot_average_heatmap(picks=significant_channels + utils.find_ch_pairs(nirs.raw.ch_names, significant_channels),\n",
    "                          clim={'hbo': [-1, 1], 'hbr': [-1, 1]}).tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "6f2714eb-4140-4d8a-a212-c8b44736f1a7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='Condition', ylabel='Theta'>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# fig, axs = plt.subplots(int(len(significant_channels)/5), 5, figsize=(20, 40))\n",
    "\n",
    "# for ch, ax in zip(significant_channels, axs.ravel()):\n",
    "#     glm_est.copy().pick(ch).scatter(axes=ax)\n",
    "#     ax.set_title(ch)\n",
    "\n",
    "# plt.tight_layout()\n",
    "\n",
    "glm_est.copy().pick(significant_channels).scatter()"
   ]
  },
  {
   "cell_type": "raw",
   "id": "eda207d7-26d8-40ca-b3cc-2dd13a067a27",
   "metadata": {},
   "source": [
    "nirs.plot_average_waveform(picks=significant_channels, cases=nirs.cases, title_format_hex=False, separate_cases=True, sharey=True).tight_layout()"
   ]
  },
  {
   "cell_type": "raw",
   "id": "07e4162b-ba26-457e-af66-a7693072284f",
   "metadata": {
    "tags": []
   },
   "source": [
    "mne.viz.plot_epochs_image(nirs.epochs, picks=significant_channels, sigma=.5, vmin=-5000, vmax=5000, show=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d12490e2-f23c-4a33-899d-076e5b89e982",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "##### `Contrast Analysis`\n",
    "\n",
    "> Compute and visualize constrasts between cases"
   ]
  },
  {
   "cell_type": "raw",
   "id": "9000123d-1067-4dc1-902e-ff973562a788",
   "metadata": {},
   "source": [
    "contrast_matrix = np.eye(design_matrix.shape[1])\n",
    "basic_conts = dict([(column, contrast_matrix[i]) for i, column in enumerate(design_matrix.columns)])"
   ]
  },
  {
   "cell_type": "raw",
   "id": "034bac9f-702f-41f6-bd11-4ebc87dbc626",
   "metadata": {},
   "source": [
    "contrast_est = glm_est.compute_contrast(basic_conts[nirs.case[1]] - basic_conts[nirs.case[0]])\n",
    "contrast_results = contrast_est.to_dataframe()"
   ]
  },
  {
   "cell_type": "raw",
   "id": "31b78f5d-47c6-40ef-b2de-41c0970f3de5",
   "metadata": {},
   "source": [
    "contrast_results['Significant'] = st.multitest.multipletests(contrast_results['p_value'])[0]"
   ]
  },
  {
   "cell_type": "raw",
   "id": "94e8ce1b-1229-4aeb-bece-2715769b11d3",
   "metadata": {},
   "source": [
    "contrast_results_significant = contrast_results[contrast_results['Significant']]\n",
    "contrast_results_significant[contrast_results_significant['Condition'].isin(nirs.cases)]"
   ]
  },
  {
   "cell_type": "raw",
   "id": "25576d0c-32ca-4a38-90d2-6f2ee9ba24db",
   "metadata": {},
   "source": [
    "significant_channels_contrast = list(contrast_results_significant['ch_name'].unique())"
   ]
  },
  {
   "cell_type": "raw",
   "id": "cecd1eea-36b2-4852-866a-c49ad89e3fb0",
   "metadata": {},
   "source": [
    "contrast.scatter()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c0d2d4d5-844f-43e0-a59b-3b3d3851e808",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "##### `Plotting for GLM Analysis`"
   ]
  },
  {
   "cell_type": "raw",
   "id": "2cdbcef1-c8e5-42e3-9688-e2ef1d01aa6f",
   "metadata": {},
   "source": [
    "# Plot design matrix\n",
    "\n",
    "nilearn.plotting.plot_design_matrix(design_matrix)"
   ]
  },
  {
   "cell_type": "raw",
   "id": "73333113-a1e1-46fb-86ae-6a61e5c417f1",
   "metadata": {},
   "source": [
    "# Visualize regression coefficients for all channels\n",
    "\n",
    "glm_est.scatter();"
   ]
  },
  {
   "cell_type": "raw",
   "id": "f0be2d36-e70f-4d5b-965b-b4e20ddf9c98",
   "metadata": {},
   "source": [
    "# View topographic distibution of activities\n",
    "\n",
    "glm_est.plot_topo(conditions=nirs.cases);"
   ]
  },
  {
   "cell_type": "raw",
   "id": "b0da82dd-e2c5-418d-b9d1-5486fa58323c",
   "metadata": {},
   "source": [
    "# Compute and visualize constrasts between cases\n",
    "\n",
    "contrast_matrix = np.eye(design_matrix.shape[1])\n",
    "basic_conts = dict([(column, contrast_matrix[i]) for i, column in enumerate(design_matrix.columns)])\n",
    "contrast_LvR = basic_conts['2'] - basic_conts['0']\n",
    "\n",
    "contrast = glm_est.compute_contrast(contrast_LvR)\n",
    "contrast.plot_topo();"
   ]
  },
  {
   "cell_type": "raw",
   "id": "83afe472-eb99-4771-885a-2f5f0b7a0be4",
   "metadata": {},
   "source": [
    "# Summarize data (compute TP and FP rates)\n",
    "\n",
    "(glm_est.to_dataframe()\n",
    " .query(\"Condition in nirs.cases\")\n",
    " .drop(['df', 'mse', 'p_value', 't'], axis=1)\n",
    " .groupby(['Condition', 'Chroma', 'ch_name'])\n",
    " .agg(['mean']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "464f86f1-0b0b-4f08-9d93-83e50a9d69d5",
   "metadata": {
    "tags": []
   },
   "source": [
    "##### `Clustering`\n",
    "\n",
    "> Relevant features  \n",
    ">> For intervals of 1s in the epochs.  \n",
    ">> - Area under the curve  \n",
    ">> - Average slope  \n",
    ">> - Maximum amplitude"
   ]
  },
  {
   "cell_type": "raw",
   "id": "06737bb7-1c86-4654-9585-e68411eb9929",
   "metadata": {},
   "source": [
    "clusters = sc.cluster.hierarchy.linkage(nirs.epochs.get_data()[:, 0, :], method='ward', metric=\"euclidean\")"
   ]
  },
  {
   "cell_type": "raw",
   "id": "b308f2ed-0178-43e7-b5e7-5d26731a0c44",
   "metadata": {},
   "source": [
    "sc.cluster.hierarchy.dendrogram(Z=clusters)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fd78eae2-754d-44b0-843c-881a0e941fe6",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.8"
  },
  "vscode": {
   "interpreter": {
    "hash": "0f589f7137c7f25103dd6a6e330ec0c0772eb7f22b3e4ba8c4980947ae7bbe74"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}