.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "auto_examples/stride_segmentation/barth_dtw_stride_segmentation_roi.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        Click :ref:`here <sphx_glr_download_auto_examples_stride_segmentation_barth_dtw_stride_segmentation_roi.py>`
        to download the full example code

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_auto_examples_stride_segmentation_barth_dtw_stride_segmentation_roi.py:


.. _roi_stride_segmentation:

Stride segmentation with Regions of Interest
============================================

In long datasets it might be desirable to predefine regions of interest in which strides should be segmented.
This can massively speed up the computation or focus the analysis on a specific part of the data.

Most of the time such regions of interest would be defined using some sort of gait detection algorithms (e.g.
:mod:`~gaitmap.gait_detection.UllrichGaitSequenceDetection`).
In this example we will create an example output of such an algorithms and will use it to limit the search region for
the `BarthDtw` stride segmentation algorithms.
However, all other stride segmentation algorithms should this functionality as well.

Note, that for smaller datasets, the algorithms might actually be slower when using multiple regions of interest,
compared to analysing the entire dataset.
This is, because the method relies on a slow Python loop to process the individual regions of interest.

.. GENERATED FROM PYTHON SOURCE LINES 20-26

.. code-block:: default


    import matplotlib.pyplot as plt
    import numpy

    numpy.random.seed(0)








.. GENERATED FROM PYTHON SOURCE LINES 27-35

Getting some example data
--------------------------

For this we take some example data that contains the regular walking movement during a 2x20m walk test of a healthy
subject. The IMU signals are already rotated so that they align with the gaitmap SF coordinate system.
The data contains information from two sensors - one from the right and one from the left foot.
We will further convert the data into the body frame, as this is required by the `BarthDTW` method to perform the
stride segmentation.

.. GENERATED FROM PYTHON SOURCE LINES 35-45

.. code-block:: default

    from gaitmap.example_data import get_healthy_example_imu_data
    from gaitmap.utils.coordinate_conversion import convert_to_fbf

    data = get_healthy_example_imu_data()
    sampling_rate_hz = 204.8
    bf_data = convert_to_fbf(data, left_like="left_", right_like="right_")


    data.sort_index(axis=1).head(1)






.. raw:: html

    <div class="output_subarea output_html rendered_html output_result">
    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead tr th {
            text-align: left;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr>
          <th>sensor</th>
          <th colspan="6" halign="left">left_sensor</th>
          <th colspan="6" halign="left">right_sensor</th>
        </tr>
        <tr>
          <th>axis</th>
          <th>acc_x</th>
          <th>acc_y</th>
          <th>acc_z</th>
          <th>gyr_x</th>
          <th>gyr_y</th>
          <th>gyr_z</th>
          <th>acc_x</th>
          <th>acc_y</th>
          <th>acc_z</th>
          <th>gyr_x</th>
          <th>gyr_y</th>
          <th>gyr_z</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0.0</th>
          <td>0.880811</td>
          <td>2.762208</td>
          <td>9.40865</td>
          <td>-0.112402</td>
          <td>-0.032157</td>
          <td>-0.062261</td>
          <td>0.311553</td>
          <td>-2.398646</td>
          <td>9.513275</td>
          <td>-0.323037</td>
          <td>0.084604</td>
          <td>-0.025288</td>
        </tr>
      </tbody>
    </table>
    </div>
    </div>
    <br />
    <br />

.. GENERATED FROM PYTHON SOURCE LINES 46-51

Defining regions of interest
----------------------------

We will define two regions of interest (one at the beginning and one at the end of the dataset).
We expect the algorithm to only find strides in these regions.

.. GENERATED FROM PYTHON SOURCE LINES 51-57

.. code-block:: default

    import pandas as pd

    roi = pd.DataFrame([[0, 2000], [5000, 7500]], columns=["start", "end"])
    roi.index.name = "roi_id"
    roi






.. raw:: html

    <div class="output_subarea output_html rendered_html output_result">
    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>start</th>
          <th>end</th>
        </tr>
        <tr>
          <th>roi_id</th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>0</td>
          <td>2000</td>
        </tr>
        <tr>
          <th>1</th>
          <td>5000</td>
          <td>7500</td>
        </tr>
      </tbody>
    </table>
    </div>
    </div>
    <br />
    <br />

.. GENERATED FROM PYTHON SOURCE LINES 58-61

Setting up the Stride Segmentation
----------------------------------
We will use BarthDTW with the default settings.

.. GENERATED FROM PYTHON SOURCE LINES 61-65

.. code-block:: default

    from gaitmap.stride_segmentation import BarthDtw

    dtw = BarthDtw()








.. GENERATED FROM PYTHON SOURCE LINES 66-69

Setting up the ROI wrapper
--------------------------
To apply our method to multiple regions of interest wie use the `RoiStrideSegmentation` wrapper.

.. GENERATED FROM PYTHON SOURCE LINES 69-73

.. code-block:: default

    from gaitmap.stride_segmentation import RoiStrideSegmentation

    roi_seg = RoiStrideSegmentation(segmentation_algorithm=dtw)








.. GENERATED FROM PYTHON SOURCE LINES 74-77

Applying the segmentation
-------------------------
Instead of our original dtw object we now use the wrapper with the same inputs to segment.

.. GENERATED FROM PYTHON SOURCE LINES 77-79

.. code-block:: default

    roi_seg = roi_seg.segment(data=bf_data, sampling_rate_hz=sampling_rate_hz, regions_of_interest=roi)








.. GENERATED FROM PYTHON SOURCE LINES 80-84

Inspecting the results
----------------------
The wrapper will automatically combine all stride lists from all ROIs into one.
The additional "roi_id" column indicates in which ROI a stride was identified in.

.. GENERATED FROM PYTHON SOURCE LINES 84-89

.. code-block:: default


    stride_list_left = roi_seg.stride_list_["left_sensor"]
    print(f"{len(stride_list_left)} strides were detected.")
    stride_list_left





.. rst-class:: sphx-glr-script-out

 .. code-block:: none

    16 strides were detected.


.. raw:: html

    <div class="output_subarea output_html rendered_html output_result">
    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>roi_id</th>
          <th>start</th>
          <th>end</th>
        </tr>
        <tr>
          <th>s_id</th>
          <th></th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>0</td>
          <td>364</td>
          <td>584</td>
        </tr>
        <tr>
          <th>1</th>
          <td>0</td>
          <td>584</td>
          <td>802</td>
        </tr>
        <tr>
          <th>2</th>
          <td>0</td>
          <td>802</td>
          <td>1023</td>
        </tr>
        <tr>
          <th>3</th>
          <td>0</td>
          <td>1023</td>
          <td>1242</td>
        </tr>
        <tr>
          <th>4</th>
          <td>0</td>
          <td>1242</td>
          <td>1458</td>
        </tr>
        <tr>
          <th>5</th>
          <td>0</td>
          <td>1458</td>
          <td>1672</td>
        </tr>
        <tr>
          <th>6</th>
          <td>0</td>
          <td>1672</td>
          <td>1887</td>
        </tr>
        <tr>
          <th>7</th>
          <td>1</td>
          <td>5043</td>
          <td>5267</td>
        </tr>
        <tr>
          <th>8</th>
          <td>1</td>
          <td>5267</td>
          <td>5489</td>
        </tr>
        <tr>
          <th>9</th>
          <td>1</td>
          <td>5489</td>
          <td>5713</td>
        </tr>
        <tr>
          <th>10</th>
          <td>1</td>
          <td>5713</td>
          <td>5936</td>
        </tr>
        <tr>
          <th>11</th>
          <td>1</td>
          <td>5936</td>
          <td>6167</td>
        </tr>
        <tr>
          <th>12</th>
          <td>1</td>
          <td>6167</td>
          <td>6395</td>
        </tr>
        <tr>
          <th>13</th>
          <td>1</td>
          <td>6395</td>
          <td>6628</td>
        </tr>
        <tr>
          <th>14</th>
          <td>1</td>
          <td>6628</td>
          <td>6858</td>
        </tr>
        <tr>
          <th>15</th>
          <td>1</td>
          <td>6858</td>
          <td>7091</td>
        </tr>
      </tbody>
    </table>
    </div>
    </div>
    <br />
    <br />

.. GENERATED FROM PYTHON SOURCE LINES 90-97

All other outputs of our stride segmentation method are not accumulated, as they differ from method to method.
However, we can inspect the individual instances stored in the `instances_per_roi_` attribute to get this information.

In the following we will plot the cost matrices that are created for ROI.
We will see that we now have multiple cost matrices and cost functions per sensor.
This is simply because the DTW was applied separately to the two regions.
To plot the outputs together, we need to loop all dtw instances.

.. GENERATED FROM PYTHON SOURCE LINES 97-127

.. code-block:: default

    import numpy as np

    # Create combined outputs
    sensor = "left_sensor"
    cost_funcs = {roi: dtw.cost_function_[sensor] for roi, dtw in roi_seg.instances_per_roi_.items()}
    cost_mat = {roi: dtw.acc_cost_mat_[sensor] for roi, dtw in roi_seg.instances_per_roi_.items()}
    full_cost_matrix = np.full((len(roi_seg.segmentation_algorithm.template.get_data()), len(roi_seg.data)), np.nan)

    fig, axs = plt.subplots(nrows=3, sharex=True, figsize=(10, 5))
    roi_seg.data[sensor]["gyr_ml"].reset_index(drop=True).plot(ax=axs[0])
    axs[0].set_ylabel("gyro [deg/s]")
    for roi, (start, end) in roi_seg.regions_of_interest[["start", "end"]].iterrows():
        axs[1].plot(np.arange(start, end), cost_funcs[roi], c="C0")
        full_cost_matrix[:, start:end] = cost_mat[roi]
    axs[1].set_ylabel("dtw cost [a.u.]")
    axs[1].axhline(roi_seg.segmentation_algorithm.max_cost, color="k", linestyle="--")
    axs[2].imshow(full_cost_matrix, aspect="auto")
    axs[2].set_ylabel("template position [#]")
    for roi_id, dtw_instance in roi_seg.instances_per_roi_.items():
        roi_start = roi_seg.regions_of_interest.loc[roi_id]["start"]
        for p in dtw_instance.paths_[sensor]:
            axs[2].plot(p.T[1] + roi_start, p.T[0])
        for start, end in dtw_instance.matches_start_end_original_[sensor]:
            axs[1].axvspan(start + roi_start, end + roi_start, alpha=0.3, color="g")
    for _, s in roi_seg.stride_list_[sensor][["start", "end"]].iterrows():
        axs[0].axvspan(*s, alpha=0.3, color="g")
    axs[0].set_xlabel("time [#]")
    fig.tight_layout()
    fig.show()




.. image-sg:: /auto_examples/stride_segmentation/images/sphx_glr_barth_dtw_stride_segmentation_roi_001.png
   :alt: barth dtw stride segmentation roi
   :srcset: /auto_examples/stride_segmentation/images/sphx_glr_barth_dtw_stride_segmentation_roi_001.png
   :class: sphx-glr-single-img






.. rst-class:: sphx-glr-timing

   **Total running time of the script:** ( 0 minutes  2.298 seconds)

**Estimated memory usage:**  59 MB


.. _sphx_glr_download_auto_examples_stride_segmentation_barth_dtw_stride_segmentation_roi.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example


    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: barth_dtw_stride_segmentation_roi.py <barth_dtw_stride_segmentation_roi.py>`

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: barth_dtw_stride_segmentation_roi.ipynb <barth_dtw_stride_segmentation_roi.ipynb>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_