Merge pull request #49 from AI2MS/45-absolute-creatinine-criterion-do…

…es-not-get-calculated-correctly 45 absolute creatinine criterion does not get calculated correctly
aidh-ms · Sep 25, 2023 · 26251d2 · 26251d2
2 parents 2c4a9e0 + 2fd5662
commit 26251d2
Show file tree

Hide file tree

Showing 21 changed files with 10,119 additions and 4 deletions.
diff --git a/generate_tests.ipynb b/generate_tests.ipynb
diff --git a/mimic_experiment.ipynb b/mimic_experiment.ipynb
diff --git a/prepare_tests_for_physicians.ipynb b/prepare_tests_for_physicians.ipynb
@@ -0,0 +1,339 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "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></th>\n",
+       "      <th>stay_id</th>\n",
+       "      <th>charttime</th>\n",
+       "      <th>urineoutput</th>\n",
+       "      <th>urineoutput_stage</th>\n",
+       "      <th>creat</th>\n",
+       "      <th>abs_crea_stage</th>\n",
+       "      <th>rel_crea_stage</th>\n",
+       "      <th>weight</th>\n",
+       "      <th>rrt_status</th>\n",
+       "      <th>rrt_stage</th>\n",
+       "      <th>stage</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>30057454.0</td>\n",
+       "      <td>2171-11-07 21:00:00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>87.2</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2171-11-07 22:00:00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>87.2</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2171-11-07 23:00:00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>87.2</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2171-11-08 00:00:00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>87.2</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2171-11-08 01:00:00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>87.2</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17939</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2148-01-08 12:00:00</td>\n",
+       "      <td>100.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>120.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17940</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2148-01-08 13:00:00</td>\n",
+       "      <td>270.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>120.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17941</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2148-01-08 14:00:00</td>\n",
+       "      <td>120.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>120.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17942</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2148-01-08 15:00:00</td>\n",
+       "      <td>125.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>120.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17943</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>2148-01-08 16:00:00</td>\n",
+       "      <td>250.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>120.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>17944 rows × 11 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "          stay_id           charttime  urineoutput  urineoutput_stage  creat  \\\n",
+       "0      30057454.0 2171-11-07 21:00:00          NaN                NaN    2.0   \n",
+       "1             NaN 2171-11-07 22:00:00          NaN                NaN    2.0   \n",
+       "2             NaN 2171-11-07 23:00:00          NaN                NaN    2.0   \n",
+       "3             NaN 2171-11-08 00:00:00          NaN                NaN    2.0   \n",
+       "4             NaN 2171-11-08 01:00:00          NaN                NaN    2.0   \n",
+       "...           ...                 ...          ...                ...    ...   \n",
+       "17939         NaN 2148-01-08 12:00:00        100.0                NaN    NaN   \n",
+       "17940         NaN 2148-01-08 13:00:00        270.0                NaN    NaN   \n",
+       "17941         NaN 2148-01-08 14:00:00        120.0                NaN    NaN   \n",
+       "17942         NaN 2148-01-08 15:00:00        125.0                NaN    NaN   \n",
+       "17943         NaN 2148-01-08 16:00:00        250.0                NaN    NaN   \n",
+       "\n",
+       "       abs_crea_stage  rel_crea_stage  weight  rrt_status  rrt_stage  stage  \n",
+       "0                 NaN             NaN    87.2         NaN        NaN    NaN  \n",
+       "1                 NaN             NaN    87.2         NaN        NaN    NaN  \n",
+       "2                 NaN             NaN    87.2         NaN        NaN    NaN  \n",
+       "3                 NaN             NaN    87.2         NaN        NaN    NaN  \n",
+       "4                 NaN             NaN    87.2         NaN        NaN    NaN  \n",
+       "...               ...             ...     ...         ...        ...    ...  \n",
+       "17939             NaN             NaN   120.0         NaN        NaN    NaN  \n",
+       "17940             NaN             NaN   120.0         NaN        NaN    NaN  \n",
+       "17941             NaN             NaN   120.0         NaN        NaN    NaN  \n",
+       "17942             NaN             NaN   120.0         NaN        NaN    NaN  \n",
+       "17943             NaN             NaN   120.0         NaN        NaN    NaN  \n",
+       "\n",
+       "[17944 rows x 11 columns]"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data = pd.read_excel('tests/experiment/data/test_human_aki.xlsx')\n",
+    "data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[37267577.0,\n",
+       " 39268883.0,\n",
+       " 38540883.0,\n",
+       " 34592300.0,\n",
+       " 35009126.0,\n",
+       " 32314488.0,\n",
+       " 34617352.0,\n",
+       " 39635619.0,\n",
+       " 32391858.0,\n",
+       " 32119961.0,\n",
+       " 38383343.0,\n",
+       " 35514836.0,\n",
+       " 30849778.0,\n",
+       " 33281088.0,\n",
+       " 35258379.0]"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import random\n",
+    "stay_ids = data.stay_id.unique()\n",
+    "# select 3 sets of 5 random stays\n",
+    "random.seed(301)\n",
+    "random_stay_ids = random.sample(list(stay_ids), 15)\n",
+    "random_stay_ids"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "([37267577.0, 39268883.0, 38540883.0, 34592300.0, 35009126.0],\n",
+       " [32314488.0, 34617352.0, 39635619.0, 32391858.0, 32119961.0],\n",
+       " [38383343.0, 35514836.0, 30849778.0, 33281088.0, 35258379.0])"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "set_1 = random_stay_ids[:5]\n",
+    "set_2 = random_stay_ids[5:10]\n",
+    "set_3 = random_stay_ids[10:15]\n",
+    "set_1, set_2, set_3"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "pyakienv",
+   "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": "3.11.4"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/pyAKI/preprocessors.py b/pyAKI/preprocessors.py
@@ -80,7 +80,7 @@ def __init__(
         stay_identifier: str = "stay_id",
         time_identifier: str = "charttime",
         interpolate: bool = True,
-        threshold: Optional[int] = None,
+        threshold: Optional[int] = 8,
     ) -> None:
         """
         Initialize a new instance of the UrineOutputPreProcessor class.
@@ -140,7 +140,7 @@ def __init__(
         stay_identifier: str = "stay_id",
         time_identifier: str = "charttime",
         ffill: bool = True,
-        threshold: Optional[int] = None,
+        threshold: Optional[int] = 72,
     ) -> None:
         """
         Initialize a new instance of the CreatininePreProcessor class.

diff --git a/pyAKI/probes.py b/pyAKI/probes.py
@@ -305,7 +305,7 @@ def probe(self, df: pd.DataFrame = None, **kwargs) -> pd.DataFrame:
         df = df.copy()
         df[self.RESNAME] = 0
         df.loc[(df[self._column] - baseline_values) >= 0.3, self.RESNAME] = 1
-        df.loc[(df[self._column] - baseline_values) >= 4, self.RESNAME] = 3
+        df.loc[df[self._column] >= 4, self.RESNAME] = 3
 
         df.loc[df[self._column] == 0, self.RESNAME] = None
         df[self.RESNAME] = df[self.RESNAME].ffill().fillna(0)

diff --git a/setup.cfg b/setup.cfg
@@ -4,7 +4,7 @@ description_file = README.md
 long_description_content_type = text/markdown
 summary = pyAKI allows calculation of Acute Kidney Injury from urine output and creatinine based on KDIGO criteria.
 author = 'Jan Ernsting, Christian Porschen'
-author_email = '[email protected], '
+author_email = '[email protected], [email protected]'
 url = https://github.com/AI2MS/pyAKI
 project_urls=
     Source Code = https://github.com/AI2MS/pyAKI

diff --git a/test_machine.py b/test_machine.py
@@ -0,0 +1,20 @@
+import pyAKI.kdigo
+import pandas as pd
+
+if __name__ == "__main__":
+    # Load the test data
+    urine_output = pd.read_csv("tests/data/test_machine_urineoutput.csv")
+    creatinine = pd.read_csv("tests/data/test_machine_creatinine.csv")
+    crrt = pd.read_csv("tests/data/test_machine_crrt.csv")
+    patient_data = pd.read_csv("tests/data/test_machine_weights.csv")
+
+    ana = pyAKI.kdigo.Analyser(
+        [
+            pyAKI.utils.Dataset(pyAKI.utils.DatasetType.URINEOUTPUT, urine_output),
+            pyAKI.utils.Dataset(pyAKI.utils.DatasetType.CREATININE, creatinine),
+            pyAKI.utils.Dataset(pyAKI.utils.DatasetType.DEMOGRAPHICS, patient_data),
+            pyAKI.utils.Dataset(pyAKI.utils.DatasetType.CRRT, crrt),
+        ]
+    )
+    ana.process_stays().to_csv("tests/data/test_machine_aki.csv")
+    ana.process_stays().to_excel("tests/data/test_machine_aki.xlsx")