.. DO NOT EDIT. .. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY. .. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE: .. "auto_revalidation/stride_length/_01_sl_analysis.py" .. LINE NUMBERS ARE GIVEN BELOW. .. only:: html .. note:: :class: sphx-glr-download-link-note :ref:`Go to the end ` to download the full example code. .. rst-class:: sphx-glr-example-title .. _sphx_glr_auto_revalidation_stride_length__01_sl_analysis.py: .. _sl_val_results: Performance of the stride length algorithms on the TVS dataset ============================================================== The following provides an analysis and comparison of the stride length algorithms on the TVS dataset (lab and free-living). We look into the actual performance of the algorithms compared to the reference data and compare the results to the previous results generated by the matlab pipeline. .. note:: If you are interested in how these results are calculated, head over to the :ref:`processing page `. .. GENERATED FROM PYTHON SOURCE LINES 18-22 Below are the list of algorithms that we will compare. Note, that we use the prefix "MobGap" to refer to the reimplemented python algorithms. For the zjils algorithm, we compare both potential threshold values that were determined as part of the pre-validation analysis on the MsProject dataset. .. GENERATED FROM PYTHON SOURCE LINES 22-33 .. code-block:: Python algorithms = { "SlZjilstra__MS_ALL": ("SlZjilstra - MS-all", "MobGap"), "SlZjilstra__MS_MS": ("SlZjilstra - MS-MS", "MobGap"), "matlab_zjilsV3__MS_ALL": ( "SlZjilstra - MS-all", "Original Implementation", ), "matlab_zjilsV3__MS_MS": ("SlZjilstra - MS-MS", "Original Implementation"), } .. GENERATED FROM PYTHON SOURCE LINES 34-41 The code below loads the data and prepares it for the analysis. By default, the data will be downloaded from an online repository (and cached locally). If you want to use a local copy of the data, you can set the `MOBGAP_VALIDATION_DATA_PATH` environment variable. and the MOBGAP_VALIDATION_USE_LOCA_DATA to `1`. The file download will print a couple log information, which can usually be ignored. You can also change the `version` parameter to load a different version of the data. .. GENERATED FROM PYTHON SOURCE LINES 41-119 .. code-block:: Python from pathlib import Path import pandas as pd from mobgap.data.validation_results import ValidationResultLoader from mobgap.utils.misc import get_env_var def format_loaded_results( values: dict[tuple[str, str], pd.DataFrame], index_cols: list[str], convert_rel_error: bool = False, ) -> pd.DataFrame: formatted = ( pd.concat(values, names=["algo", "version", *index_cols]) .reset_index() .assign( algo_with_version=lambda df: ( df["algo"] + " (" + df["version"] + ")" ), _combined="combined", ) ) if not convert_rel_error: return formatted rel_cols = [c for c in formatted.columns if "rel_error" in c] formatted[rel_cols] = formatted[rel_cols] * 100 return formatted local_data_path = ( Path(get_env_var("MOBGAP_VALIDATION_DATA_PATH")) / "results" if int(get_env_var("MOBGAP_VALIDATION_USE_LOCAL_DATA", 0)) else None ) __RESULT_VERSION = "v1.0.0" loader = ValidationResultLoader( "sl", result_path=local_data_path, version=__RESULT_VERSION ) free_living_index_cols = [ "cohort", "participant_id", "time_measure", "recording", "recording_name", "recording_name_pretty", ] free_living_results = format_loaded_results( { v: loader.load_single_results(k, "free_living") for k, v in algorithms.items() }, free_living_index_cols, convert_rel_error=True, ) lab_index_cols = [ "cohort", "participant_id", "time_measure", "test", "trial", "test_name", "test_name_pretty", ] lab_results = format_loaded_results( { v: loader.load_single_results(k, "laboratory") for k, v in algorithms.items() }, lab_index_cols, convert_rel_error=True, ) cohort_order = ["HA", "CHF", "COPD", "MS", "PD", "PFF"] .. rst-class:: sphx-glr-script-out .. code-block:: none 0%| | 0.00/12.0k [00:00 pd.DataFrame: return ( df.pipe(apply_transformations, format_transforms) .rename(columns=final_names) .loc[:, list(final_names.values())] ) .. GENERATED FROM PYTHON SOURCE LINES 277-285 Free-Living Comparison ---------------------- We focus on the free-living data for the comparison as this is the expected use case for the algorithms. All results across all cohorts ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The results below represent the average performance across all participants independent of the cohort. .. GENERATED FROM PYTHON SOURCE LINES 285-315 .. code-block:: Python import matplotlib.pyplot as plt import seaborn as sns fig, ax = plt.subplots() sns.boxplot( data=free_living_results, x="algo_with_version", y="wb__abs_error", ax=ax ) plt.xticks(rotation=45, ha="right") fig.tight_layout() fig.show() perf_metrics_all = free_living_results.pipe( multilevel_groupby_apply_merge, [ ( ["algo", "version"], partial(apply_aggregations, aggregations=custom_aggs), ), ( ["algo"], partial(apply_transformations, transformations=stats_transform), ), ], ).pipe(format_tables) perf_metrics_all.style.pipe( revalidation_table_styles, validation_thresholds, ["algo"], ) .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_001.png :alt: 01 sl analysis :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_001.png :class: sphx-glr-single-img .. raw:: html
    # participants WD mean and CI [m] INDIP mean and CI [m] Bias and LoA [m] Abs. Error [m] Rel. Error [%] Abs. Rel. Error [%] ICC # Failed WBs
algo version                  
SlZjilstra - MS-MS MobGap 101 0.89 [0.86, 0.92] 0.79 [0.76, 0.83] 0.10 [-0.10, 0.29] 0.15 [0.14, 0.16]** 19.51 [16.12, 22.89] 24.71 [22.03, 27.38]* 0.70 [0.16, 0.87] 1
Original Implementation 101 0.93 [0.90, 0.96] 0.79 [0.76, 0.83] 0.13 [-0.08, 0.34] 0.17 [0.16, 0.19] 26.03 [21.99, 30.08] 30.15 [26.67, 33.62] 0.59 [-0.04, 0.83] 1
SlZjilstra - MS-all MobGap 101 0.92 [0.89, 0.95] 0.79 [0.76, 0.83] 0.13 [-0.07, 0.32] 0.16 [0.15, 0.17]** 23.47 [19.97, 26.97] 27.03 [24.07, 29.99]* 0.64 [-0.03, 0.86] 1
Original Implementation 101 0.96 [0.93, 0.99] 0.79 [0.76, 0.83] 0.16 [-0.05, 0.37] 0.19 [0.17, 0.21] 30.20 [26.02, 34.38] 32.89 [29.13, 36.66] 0.53 [-0.09, 0.81] 1


.. GENERATED FROM PYTHON SOURCE LINES 316-319 Per Cohort ~~~~~~~~~~ The results below represent the average performance across all participants within a cohort. .. GENERATED FROM PYTHON SOURCE LINES 319-352 .. code-block:: Python fig, ax = plt.subplots() sns.boxplot( data=free_living_results, x="cohort", y="wb__abs_error", hue="algo_with_version", order=cohort_order, ax=ax, ) fig.show() perf_metrics_cohort = ( free_living_results.pipe( multilevel_groupby_apply_merge, [ ( ["cohort", "algo", "version"], partial(apply_aggregations, aggregations=custom_aggs), ), ( ["cohort", "algo"], partial(apply_transformations, transformations=stats_transform), ), ], ) .pipe(format_tables) .loc[cohort_order] ) perf_metrics_cohort.style.pipe( revalidation_table_styles, validation_thresholds, ["cohort", "algo"], ) .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_002.png :alt: 01 sl analysis :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_002.png :class: sphx-glr-single-img .. raw:: html
      # participants WD mean and CI [m] INDIP mean and CI [m] Bias and LoA [m] Abs. Error [m] Rel. Error [%] Abs. Rel. Error [%] ICC # Failed WBs
cohort algo version                  
HA SlZjilstra - MS-MS MobGap 20 0.90 [0.85, 0.95] 0.82 [0.76, 0.87] 0.09 [-0.05, 0.22] 0.13 [0.11, 0.14]* 16.67 [11.83, 21.52] 20.99 [17.81, 24.18] 0.69 [-0.03, 0.90] 0
Original Implementation 20 0.93 [0.88, 0.99] 0.82 [0.76, 0.87] 0.11 [-0.04, 0.27] 0.15 [0.13, 0.17] 21.98 [16.32, 27.65] 25.62 [21.41, 29.83] 0.55 [-0.10, 0.85] 0
SlZjilstra - MS-all MobGap 20 0.93 [0.88, 0.99] 0.82 [0.76, 0.87] 0.12 [-0.02, 0.25] 0.14 [0.12, 0.16] 20.54 [15.53, 25.55] 23.33 [19.57, 27.09] 0.60 [-0.09, 0.88] 0
Original Implementation 20 0.96 [0.91, 1.02] 0.82 [0.76, 0.87] 0.15 [-0.01, 0.30] 0.17 [0.15, 0.19] 26.01 [20.16, 31.87] 28.31 [23.52, 33.11] 0.47 [-0.10, 0.81] 0
CHF SlZjilstra - MS-MS MobGap 10 0.91 [0.82, 1.00] 0.86 [0.75, 0.97] 0.06 [-0.10, 0.21] 0.13 [0.10, 0.17] 14.97 [5.64, 24.31] 22.01 [14.20, 29.83] 0.85 [0.43, 0.96] 0
Original Implementation 10 0.94 [0.85, 1.03] 0.86 [0.75, 0.97] 0.08 [-0.07, 0.23] 0.15 [0.12, 0.19] 20.31 [10.82, 29.79] 26.73 [18.56, 34.89] 0.80 [0.09, 0.96] 0
SlZjilstra - MS-all MobGap 10 0.94 [0.85, 1.04] 0.86 [0.75, 0.97] 0.09 [-0.07, 0.24] 0.14 [0.10, 0.17] 18.78 [9.14, 28.43] 23.51 [14.93, 32.09] 0.80 [0.07, 0.95] 0
Original Implementation 10 0.97 [0.88, 1.06] 0.86 [0.75, 0.97] 0.11 [-0.04, 0.26] 0.16 [0.12, 0.20] 24.29 [14.49, 34.08] 28.38 [19.39, 37.37] 0.74 [-0.07, 0.94] 0
COPD SlZjilstra - MS-MS MobGap 17 0.94 [0.87, 1.00] 0.81 [0.75, 0.88] 0.12 [-0.03, 0.27] 0.15 [0.13, 0.17] 19.76 [14.57, 24.94] 22.96 [19.24, 26.69] 0.59 [-0.10, 0.87] 0
Original Implementation 17 0.97 [0.91, 1.03] 0.81 [0.75, 0.88] 0.16 [-0.00, 0.31] 0.18 [0.15, 0.20] 25.88 [19.96, 31.81] 27.94 [23.15, 32.72] 0.46 [-0.09, 0.82] 0
SlZjilstra - MS-all MobGap 17 0.97 [0.90, 1.03] 0.81 [0.75, 0.88] 0.15 [0.00, 0.30] 0.17 [0.15, 0.20] 23.72 [18.36, 29.08] 25.76 [21.44, 30.08] 0.51 [-0.09, 0.84] 0
Original Implementation 17 1.01 [0.94, 1.07] 0.81 [0.75, 0.88] 0.19 [0.03, 0.35] 0.20 [0.17, 0.23] 30.05 [23.92, 36.17] 31.25 [25.91, 36.60] 0.39 [-0.07, 0.78] 0
MS SlZjilstra - MS-MS MobGap 18 0.98 [0.91, 1.05] 0.84 [0.76, 0.92] 0.13 [-0.08, 0.35] 0.18 [0.14, 0.21] 24.22 [15.50, 32.94] 27.95 [20.55, 35.35] 0.57 [-0.07, 0.85] 0
Original Implementation 18 1.03 [0.96, 1.09] 0.84 [0.76, 0.92] 0.18 [-0.05, 0.41] 0.21 [0.17, 0.25] 31.40 [21.12, 41.67] 34.03 [24.71, 43.35] 0.44 [-0.11, 0.79] 0
SlZjilstra - MS-all MobGap 18 1.01 [0.94, 1.08] 0.84 [0.76, 0.92] 0.17 [-0.05, 0.39] 0.20 [0.16, 0.24] 28.33 [19.32, 37.35] 30.88 [22.81, 38.96] 0.51 [-0.10, 0.83] 0
Original Implementation 18 1.06 [0.99, 1.13] 0.84 [0.76, 0.92] 0.21 [-0.02, 0.44] 0.23 [0.18, 0.28] 35.74 [25.12, 46.36] 37.28 [27.25, 47.31] 0.39 [-0.10, 0.76] 0
PD SlZjilstra - MS-MS MobGap 19 0.86 [0.78, 0.94] 0.79 [0.70, 0.87] 0.07 [-0.20, 0.35] 0.15 [0.12, 0.19] 17.93 [6.19, 29.67] 26.35 [17.07, 35.63] 0.65 [0.27, 0.85] 0
Original Implementation 19 0.90 [0.82, 0.97] 0.79 [0.70, 0.87] 0.11 [-0.19, 0.41] 0.18 [0.13, 0.22] 24.69 [9.99, 39.39] 31.55 [18.76, 44.33] 0.55 [0.09, 0.80] 0
SlZjilstra - MS-all MobGap 19 0.89 [0.81, 0.97] 0.79 [0.70, 0.87] 0.10 [-0.18, 0.38] 0.16 [0.12, 0.21] 21.84 [9.71, 33.97] 28.02 [17.85, 38.19] 0.61 [0.14, 0.84] 0
Original Implementation 19 0.93 [0.85, 1.01] 0.79 [0.70, 0.87] 0.14 [-0.16, 0.44] 0.19 [0.14, 0.24] 28.81 [13.62, 44.00] 33.69 [20.00, 47.39] 0.50 [-0.01, 0.79] 0
PFF SlZjilstra - MS-MS MobGap 17 0.74 [0.67, 0.80] 0.65 [0.56, 0.73] 0.09 [-0.08, 0.25] 0.14 [0.11, 0.16] 22.19 [14.14, 30.24] 27.28 [21.25, 33.30] 0.75 [0.08, 0.92] 1
Original Implementation 17 0.78 [0.72, 0.84] 0.65 [0.56, 0.73] 0.13 [-0.04, 0.30] 0.17 [0.14, 0.19] 30.38 [21.15, 39.61] 34.25 [26.60, 41.90] 0.62 [-0.09, 0.89] 1
SlZjilstra - MS-all MobGap 17 0.76 [0.69, 0.83] 0.65 [0.56, 0.73] 0.11 [-0.05, 0.28] 0.15 [0.12, 0.17] 26.24 [17.92, 34.56] 29.71 [22.83, 36.60] 0.70 [-0.06, 0.91] 1
Original Implementation 17 0.81 [0.74, 0.87] 0.65 [0.56, 0.73] 0.16 [-0.02, 0.33] 0.18 [0.15, 0.21] 34.69 [25.16, 44.22] 37.30 [28.91, 45.69] 0.56 [-0.09, 0.87] 1


.. GENERATED FROM PYTHON SOURCE LINES 353-358 Deep Dive Analysis of Main Algorithms ------------------------------------- Below, you can find detailed correlation and residual plots comparing the new and the old implementation of each algorithm. Each datapoint represents one participant. .. GENERATED FROM PYTHON SOURCE LINES 358-433 .. code-block:: Python from mobgap.plotting import ( calc_min_max_with_margin, make_square, move_legend_outside, plot_regline, residual_plot, ) def combo_residual_plot(data): fig, axs = plt.subplots( ncols=2, sharey=True, sharex=True, figsize=(15, 9), constrained_layout=True, ) fig.suptitle(data.name) for (version, subdata), ax in zip(data.groupby("version"), axs): residual_plot( subdata, "wb__reference", "wb__detected", "cohort", "m", ax=ax, legend=ax == axs[-1], ) ax.set_title(version) move_legend_outside(fig, axs[-1]) plt.show() def combo_scatter_plot(data): fig, axs = plt.subplots( ncols=2, sharey=True, sharex=True, figsize=(15, 8), constrained_layout=True, ) fig.suptitle(data.name) min_max = calc_min_max_with_margin( data["wb__reference"], data["wb__detected"] ) for (version, subdata), ax in zip(data.groupby("version"), axs): subdata = subdata[["wb__reference", "wb__detected", "cohort"]].dropna( how="any" ) sns.scatterplot( subdata, x="wb__reference", y="wb__detected", hue="cohort", ax=ax, legend=ax == axs[-1], ) plot_regline(subdata["wb__reference"], subdata["wb__detected"], ax=ax) make_square(ax, min_max, draw_diagonal=True) ax.set_title(version) ax.set_xlabel("Reference [m]") ax.set_ylabel("Detected [m]") move_legend_outside(fig, axs[-1]) plt.tight_layout() plt.show() free_living_results.groupby("algo").apply( combo_residual_plot, include_groups=False ) free_living_results.groupby("algo").apply( combo_scatter_plot, include_groups=False ) .. rst-class:: sphx-glr-horizontal * .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_003.png :alt: SlZjilstra - MS-MS, MobGap, Original Implementation :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_003.png :class: sphx-glr-single-img * .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_004.png :alt: SlZjilstra - MS-all, MobGap, Original Implementation :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_004.png :class: sphx-glr-single-img * .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_005.png :alt: SlZjilstra - MS-MS, MobGap, Original Implementation :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_005.png :class: sphx-glr-single-img * .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_006.png :alt: SlZjilstra - MS-all, MobGap, Original Implementation :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_006.png :class: sphx-glr-single-img .. rst-class:: sphx-glr-script-out .. code-block:: none /home/docs/checkouts/readthedocs.org/user_builds/mobgap/checkouts/v1.1.0/revalidation/stride_length/_01_sl_analysis.py:422: UserWarning: The figure layout has changed to tight plt.tight_layout() /home/docs/checkouts/readthedocs.org/user_builds/mobgap/checkouts/v1.1.0/revalidation/stride_length/_01_sl_analysis.py:422: UserWarning: The figure layout has changed to tight plt.tight_layout() .. raw:: html


.. GENERATED FROM PYTHON SOURCE LINES 434-436 Below, we show the direct correlation between the results from the old and the new implementation. Each datapoint represents one participant. .. GENERATED FROM PYTHON SOURCE LINES 436-477 .. code-block:: Python def compare_scatter_plot(data): fig, ax = plt.subplots(figsize=(9, 9), constrained_layout=True) reformated_data = ( data.pivot_table( values="wb__detected", index=("cohort", "participant_id"), columns="version", ) .reset_index() .dropna(how="any") ) min_max = calc_min_max_with_margin( reformated_data["Original Implementation"], reformated_data["MobGap"] ) sns.scatterplot( reformated_data, x="Original Implementation", y="MobGap", hue="cohort", ax=ax, ) plot_regline( reformated_data["Original Implementation"], reformated_data["MobGap"], ax=ax, ) make_square(ax, min_max, draw_diagonal=True) move_legend_outside(fig, ax) ax.set_title(data.name) ax.set_xlabel("Original Implementation [m]") ax.set_ylabel("MobGap [m]") plt.show() free_living_results.groupby("algo").apply( compare_scatter_plot, include_groups=False ) .. rst-class:: sphx-glr-horizontal * .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_007.png :alt: SlZjilstra - MS-MS :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_007.png :class: sphx-glr-single-img * .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_008.png :alt: SlZjilstra - MS-all :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_008.png :class: sphx-glr-single-img .. raw:: html


.. GENERATED FROM PYTHON SOURCE LINES 478-487 Speed dependency ~~~~~~~~~~~~~~~~ One important aspect of the algorithm performance is the dependency on the walking speed. Aka, how well do the algorithms perform at different walking speeds. For this we plot the absolute relative error against the walking speed of the reference data. For better granularity, we use the values per WB, instead of the aggregates per participant. The overlayed dots represent the trend-line calculated by taking the median of the absolute relative error within bins of 0.05 m/s. .. GENERATED FROM PYTHON SOURCE LINES 487-574 .. code-block:: Python import numpy as np wb_level_results = format_loaded_results( { v: loader.load_single_csv_file( k, "free_living", "raw_wb_level_values_with_errors.csv" ) for k, v in algorithms.items() }, free_living_index_cols, ) # For plotting all participants at the end combined = wb_level_results.copy() combined["cohort"] = "Combined" wb_level_results = pd.concat([wb_level_results, combined]).reset_index( drop=True ) algo_names = wb_level_results["algo_with_version"].unique() cohort_names = wb_level_results["cohort"].unique() wb_level_results["cohort"] = pd.Categorical( wb_level_results["cohort"], categories=cohort_names, ordered=True ) wb_level_results["algo_with_version"] = pd.Categorical( wb_level_results["algo_with_version"], categories=algo_names, ordered=True ) fig = plt.figure(constrained_layout=True, figsize=(18, 3 * len(algo_names))) subfigs = fig.subfigures(len(algo_names), 1, wspace=0.1, hspace=0.1) min_max_x = calc_min_max_with_margin(wb_level_results["reference_ws"]) min_max_y = calc_min_max_with_margin(wb_level_results["abs_rel_error"]) for subfig, (algo, data) in zip( subfigs, wb_level_results.groupby("algo_with_version", observed=True) ): subfig.suptitle(algo) subfig.supxlabel("Walking Speed (m/s)") subfig.supylabel("Absolute Relative Error") axs = subfig.subplots(1, len(cohort_names), sharex=True, sharey=True) for ax, (cohort, cohort_data) in zip( axs, data.groupby("cohort", observed=True) ): sns.scatterplot( data=cohort_data, x="reference_ws", y="abs_rel_error", ax=ax, alpha=0.3, ) bins = np.arange(0, cohort_data["reference_ws"].max() + 0.05, 0.05) cohort_data["speed_bin"] = pd.cut( cohort_data["reference_ws"], bins=bins ) # Calculate bin centers for plotting cohort_data["bin_center"] = cohort_data["speed_bin"].apply( lambda x: x.mid ) # Calculate medians per bin and cohort binned_data = ( cohort_data.groupby("bin_center", observed=True)["abs_rel_error"] .median() .reset_index() ) # Plot median lines sns.scatterplot( data=binned_data, x="bin_center", y="abs_rel_error", ax=ax, ) ax.set_title(cohort) ax.set_xlabel(None) ax.set_ylabel(None) ax.set_xlim(*min_max_x) ax.set_ylim(*min_max_y) fig.show() .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_009.png :alt: CHF, COPD, HA, MS, PD, PFF, Combined, CHF, COPD, HA, MS, PD, PFF, Combined, CHF, COPD, HA, MS, PD, PFF, Combined, CHF, COPD, HA, MS, PD, PFF, Combined :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_009.png :class: sphx-glr-single-img .. rst-class:: sphx-glr-script-out .. code-block:: none 0%| | 0.00/280k [00:00
    # participants WD mean and CI [m] INDIP mean and CI [m] Bias and LoA [m] Abs. Error [m] Rel. Error [%] Abs. Rel. Error [%] ICC # Failed WBs
algo version                  
SlZjilstra - MS-MS MobGap 1168 1.02 [1.01, 1.04] 1.02 [1.01, 1.03] 0.00 [-0.26, 0.27] 0.11 [0.10, 0.11]* 3.15 [2.10, 4.20] 12.61 [11.81, 13.40]* 0.84 [0.82, 0.85] 113
Original Implementation 1168 1.04 [1.03, 1.05] 1.02 [1.01, 1.03] 0.02 [-0.27, 0.31] 0.12 [0.11, 0.12] 5.58 [4.38, 6.79] 14.10 [13.12, 15.07] 0.80 [0.78, 0.82] 145
SlZjilstra - MS-all MobGap 1168 1.06 [1.04, 1.07] 1.02 [1.01, 1.03] 0.04 [-0.23, 0.31] 0.11 [0.11, 0.12]* 6.57 [5.48, 7.65] 13.37 [12.50, 14.23]* 0.83 [0.79, 0.86] 113
Original Implementation 1168 1.08 [1.06, 1.09] 1.02 [1.01, 1.03] 0.05 [-0.23, 0.34] 0.12 [0.12, 0.13] 9.07 [7.83, 10.32] 15.08 [14.03, 16.14] 0.79 [0.72, 0.84] 145


.. GENERATED FROM PYTHON SOURCE LINES 609-612 Per Cohort ~~~~~~~~~~ The results below represent the average performance across all trails of all participants within a cohort. .. GENERATED FROM PYTHON SOURCE LINES 612-646 .. code-block:: Python fig, ax = plt.subplots() sns.boxplot( data=lab_results, x="cohort", y="wb__abs_error", hue="algo_with_version", order=cohort_order, ax=ax, ) fig.show() perf_metrics_cohort = ( lab_results.pipe( multilevel_groupby_apply_merge, [ ( ["cohort", "algo", "version"], partial(apply_aggregations, aggregations=custom_aggs), ), ( ["cohort", "algo"], partial(apply_transformations, transformations=stats_transform), ), ], ) .pipe(format_tables) .loc[cohort_order] ) perf_metrics_cohort.style.pipe( revalidation_table_styles, validation_thresholds, ["cohort", "algo"], ) .. image-sg:: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_011.png :alt: 01 sl analysis :srcset: /auto_revalidation/stride_length/images/sphx_glr__01_sl_analysis_011.png :class: sphx-glr-single-img .. raw:: html
      # participants WD mean and CI [m] INDIP mean and CI [m] Bias and LoA [m] Abs. Error [m] Rel. Error [%] Abs. Rel. Error [%] ICC # Failed WBs
cohort algo version                  
HA SlZjilstra - MS-MS MobGap 227 1.07 [1.05, 1.10] 1.08 [1.05, 1.10] -0.01 [-0.26, 0.25] 0.11 [0.10, 0.12] 1.14 [-0.75, 3.03] 11.22 [9.92, 12.52] 0.77 [0.71, 0.82] 36
Original Implementation 227 1.08 [1.06, 1.10] 1.08 [1.05, 1.10] -0.00 [-0.27, 0.27] 0.11 [0.10, 0.12] 2.26 [0.04, 4.47] 12.21 [10.51, 13.91] 0.73 [0.66, 0.79] 36
SlZjilstra - MS-all MobGap 227 1.11 [1.08, 1.13] 1.08 [1.05, 1.10] 0.03 [-0.23, 0.29] 0.11 [0.10, 0.12] 4.49 [2.54, 6.45] 11.81 [10.41, 13.21] 0.76 [0.69, 0.82] 36
Original Implementation 227 1.11 [1.09, 1.14] 1.08 [1.05, 1.10] 0.04 [-0.24, 0.31] 0.12 [0.10, 0.13] 5.64 [3.35, 7.92] 12.78 [10.95, 14.61] 0.72 [0.63, 0.78] 36
CHF SlZjilstra - MS-MS MobGap 106 1.04 [1.00, 1.09] 1.10 [1.05, 1.15] -0.06 [-0.33, 0.22] 0.11 [0.09, 0.13] -3.81 [-6.26, -1.37] 10.15 [8.47, 11.83] 0.81 [0.69, 0.88] 9
Original Implementation 106 1.09 [1.05, 1.13] 1.10 [1.05, 1.15] -0.08 [-0.40, 0.25] 0.14 [0.12, 0.16] -4.63 [-7.50, -1.75] 12.22 [10.29, 14.14] 0.72 [0.53, 0.83] 38
SlZjilstra - MS-all MobGap 106 1.08 [1.04, 1.12] 1.10 [1.05, 1.15] -0.02 [-0.30, 0.25] 0.10 [0.08, 0.12] -0.63 [-3.15, 1.90] 9.79 [8.07, 11.51] 0.83 [0.75, 0.88] 9
Original Implementation 106 1.13 [1.09, 1.17] 1.10 [1.05, 1.15] -0.04 [-0.36, 0.28] 0.13 [0.11, 0.15] -1.47 [-4.44, 1.49] 11.89 [9.93, 13.85] 0.75 [0.62, 0.84] 38
COPD SlZjilstra - MS-MS MobGap 214 1.12 [1.10, 1.14] 1.10 [1.07, 1.12] 0.02 [-0.19, 0.23] 0.09 [0.08, 0.10] 3.29 [1.64, 4.94] 9.16 [7.93, 10.39] 0.79 [0.72, 0.84] 34
Original Implementation 214 1.13 [1.11, 1.15] 1.10 [1.07, 1.12] 0.03 [-0.19, 0.25] 0.09 [0.08, 0.10] 5.03 [2.92, 7.13] 9.93 [8.07, 11.78] 0.73 [0.64, 0.80] 34
SlZjilstra - MS-all MobGap 214 1.15 [1.13, 1.18] 1.10 [1.07, 1.12] 0.06 [-0.16, 0.28] 0.10 [0.09, 0.11] 6.72 [5.01, 8.42] 10.54 [9.22, 11.87] 0.75 [0.57, 0.84] 34
Original Implementation 214 1.17 [1.14, 1.19] 1.10 [1.07, 1.12] 0.07 [-0.16, 0.29] 0.10 [0.09, 0.11] 8.50 [6.33, 10.68] 11.32 [9.34, 13.30] 0.68 [0.44, 0.81] 34
MS SlZjilstra - MS-MS MobGap 228 1.06 [1.03, 1.09] 1.04 [1.01, 1.07] 0.02 [-0.28, 0.32] 0.12 [0.11, 0.13] 4.13 [1.80, 6.46] 13.42 [11.76, 15.07] 0.78 [0.72, 0.83] 6
Original Implementation 228 1.08 [1.05, 1.11] 1.04 [1.01, 1.07] 0.04 [-0.27, 0.36] 0.13 [0.12, 0.14] 7.14 [4.53, 9.76] 14.76 [12.74, 16.77] 0.73 [0.65, 0.79] 6
SlZjilstra - MS-all MobGap 228 1.10 [1.06, 1.13] 1.04 [1.01, 1.07] 0.06 [-0.25, 0.36] 0.13 [0.11, 0.14] 7.58 [5.17, 9.99] 14.13 [12.27, 15.98] 0.76 [0.66, 0.83] 6
Original Implementation 228 1.12 [1.09, 1.15] 1.04 [1.01, 1.07] 0.08 [-0.23, 0.40] 0.14 [0.12, 0.15] 10.68 [7.98, 13.39] 15.70 [13.44, 17.96] 0.70 [0.53, 0.80] 6
PD SlZjilstra - MS-MS MobGap 224 0.99 [0.96, 1.02] 1.00 [0.97, 1.03] -0.01 [-0.28, 0.26] 0.11 [0.10, 0.13] 1.57 [-0.60, 3.74] 12.95 [11.52, 14.38] 0.81 [0.75, 0.85] 28
Original Implementation 224 1.01 [0.99, 1.04] 1.00 [0.97, 1.03] 0.01 [-0.27, 0.29] 0.12 [0.11, 0.13] 4.44 [1.97, 6.91] 13.86 [12.01, 15.71] 0.78 [0.72, 0.83] 30
SlZjilstra - MS-all MobGap 224 1.02 [0.99, 1.05] 1.00 [0.97, 1.03] 0.02 [-0.25, 0.30] 0.12 [0.11, 0.13] 4.94 [2.70, 7.18] 13.44 [11.86, 15.02] 0.80 [0.75, 0.85] 28
Original Implementation 224 1.05 [1.02, 1.07] 1.00 [0.97, 1.03] 0.05 [-0.23, 0.33] 0.12 [0.11, 0.13] 7.89 [5.34, 10.44] 14.64 [12.61, 16.67] 0.77 [0.69, 0.83] 30
PFF SlZjilstra - MS-MS MobGap 169 0.85 [0.81, 0.88] 0.82 [0.78, 0.87] 0.02 [-0.25, 0.29] 0.11 [0.10, 0.12] 9.80 [5.58, 14.03] 17.79 [14.19, 21.38] 0.87 [0.83, 0.91] 0
Original Implementation 169 0.86 [0.83, 0.89] 0.82 [0.78, 0.87] 0.04 [-0.25, 0.33] 0.13 [0.11, 0.14] 13.36 [8.69, 18.02] 20.86 [16.85, 24.87] 0.84 [0.78, 0.88] 1
SlZjilstra - MS-all MobGap 169 0.87 [0.84, 0.91] 0.82 [0.78, 0.87] 0.05 [-0.21, 0.31] 0.11 [0.10, 0.13]* 13.44 [9.08, 17.81] 19.10 [15.24, 22.96] 0.87 [0.80, 0.91] 0
Original Implementation 169 0.89 [0.86, 0.92] 0.82 [0.78, 0.87] 0.07 [-0.22, 0.35] 0.13 [0.12, 0.15] 17.11 [12.28, 21.93] 22.73 [18.46, 27.00] 0.83 [0.72, 0.89] 1


.. rst-class:: sphx-glr-timing **Total running time of the script:** (0 minutes 13.337 seconds) **Estimated memory usage:** 96 MB .. _sphx_glr_download_auto_revalidation_stride_length__01_sl_analysis.py: .. only:: html .. container:: sphx-glr-footer sphx-glr-footer-example .. container:: sphx-glr-download sphx-glr-download-jupyter :download:`Download Jupyter notebook: _01_sl_analysis.ipynb <_01_sl_analysis.ipynb>` .. container:: sphx-glr-download sphx-glr-download-python :download:`Download Python source code: _01_sl_analysis.py <_01_sl_analysis.py>` .. container:: sphx-glr-download sphx-glr-download-zip :download:`Download zipped: _01_sl_analysis.zip <_01_sl_analysis.zip>` .. only:: html .. rst-class:: sphx-glr-signature `Gallery generated by Sphinx-Gallery `_