support adding coding systems later on

ycq091044 · Jul 7, 2022 · e44890e · e44890e
1 parent 7889b70
commit e44890e
Show file tree

Hide file tree

Showing 16 changed files with 396 additions and 269 deletions.
diff --git a/MedCode.egg-info/PKG-INFO b/MedCode.egg-info/PKG-INFO
diff --git a/MedCode.egg-info/SOURCES.txt b/MedCode.egg-info/SOURCES.txt
@@ -1,3 +1,4 @@
+LICENSE
 README.md
 setup.py
 MedCode/__init__.py

diff --git a/MedCode.egg-info/entry_points.txt b/MedCode.egg-info/entry_points.txt
@@ -1,3 +1,2 @@
 [console_scripts]
 MedCode = MedCode:MedCode
-
diff --git a/MedCode/__init__.py b/MedCode/__init__.py
@@ -5,6 +5,8 @@
 from urllib import request
 import sys
 
+__version__ = "0.1.1"
+
 # README
 README = """
 ---------------------------- Help ---------------------------
@@ -153,4 +155,27 @@ def map_combine(self, key, mid, value):
                 if m_item in Dict2:
                     temp += Dict2[m_item]
             if len(temp) > 1:
-                exec("self.{}_to_{}[k] = list(set(temp))".format(key, value))
+                exec("self.{}_to_{}[k] = list(set(temp))".format(key, value))
+
+
+    def add_code(self, new_code):
+        code_pairs = [(new_code, code) for code in self.Codes if code != new_code]
+        for code1, code2 in code_pairs:
+            if code2 == code1: continue
+            if code1 != 'SMILES':
+                if not hasattr(self, "{}_to_{}".format(code1, code2)):
+                    if hasattr(self, "{}_to_{}".format(code2, code1)):
+                        exec("self.{0}_to_{1} = self.mapping_reverse_dict(self.{1}_to_{0})".format(code1, code2))
+                    else:
+                        shortest_path = nx.shortest_path(self.G, source=code1, target=code2)
+                        key = shortest_path[0]
+                        for i in range(len(shortest_path) - 2):
+                            mid, value = shortest_path[i+1: i+3]
+                            self.map_combine(key, mid, value)
+                            self.G.add_edge(key, value)
+                print ("mapping finished: {} -> {}".format(code1, code2))
+
+            if code2 != 'SMILES':
+                if not hasattr(self, "{}_to_{}".format(code2, code1)) and (code2 != 'SMILES'):
+                    exec("self.{0}_to_{1} = self.mapping_reverse_dict(self.{1}_to_{0})".format(code2, code1))
+                print ("mapping finished: {} -> {}".format(code2, code1))
diff --git a/MedCode/__pycache__/__init__.cpython-38.pyc b/MedCode/__pycache__/__init__.cpython-38.pyc
diff --git a/MedCode/__pycache__/__init__.cpython-39.pyc b/MedCode/__pycache__/__init__.cpython-39.pyc
diff --git a/README.md b/README.md
@@ -24,13 +24,14 @@ $ pip install MedCode
 ```bash
 # local installation
 $ cd ~/MedCode
-$ pip3 install dist/MedCode-0.1-py3-none-any.whl
+$ pip3 install dist/MedCode-[VERSION-ID]-py3-none-any.whl
 ```
 To look up for help, directly type "MedCode" in the cmd and the help message will pop up.
 ```bash
 $ MedCode
 ```
 ## 2. Quick Usage
+### Load all mappings during initialization
 ```python
 from MedCode import CodeMapping
 # initialize with a list of supported coding systems
@@ -42,6 +43,14 @@ tool.RXCUI_to_SMILES['312055']
 tool.NDC_to_Name['76413-153-06']
 ...
 ```
+### Want to add more coding systems later?
+```python
+# add additional coding system
+tool.add_code("RxNorm")
+
+# we are good to go, e.g.,
+tool.NDC_to_RxNorm['76413-153-06']
+```
 - check ```test/Example.ipynb``` for examples.
 - **Implementation Features --- Minimal Computation.** The tool will require minimum computation cost for generating all necessary combinations within the listed coding systems, and other mapping functions will not be generated. For example, in this demo, ```NDC_to_RxNorm``` will not be accessible since ```RxNorm``` is not listed in the user input. Our minimal cost computation relies on that we maintain a graph structure of the listed coding systems and use shortest path method to find the missing mapping.
 

diff --git a/build/lib/MedCode/__init__.py b/build/lib/MedCode/__init__.py
@@ -5,6 +5,8 @@
 from urllib import request
 import sys
 
+__version__ = "0.1.1"
+
 # README
 README = """
 ---------------------------- Help ---------------------------
@@ -153,4 +155,27 @@ def map_combine(self, key, mid, value):
                 if m_item in Dict2:
                     temp += Dict2[m_item]
             if len(temp) > 1:
-                exec("self.{}_to_{}[k] = list(set(temp))".format(key, value))
+                exec("self.{}_to_{}[k] = list(set(temp))".format(key, value))
+
+
+    def add_code(self, new_code):
+        code_pairs = [(new_code, code) for code in self.Codes if code != new_code]
+        for code1, code2 in code_pairs:
+            if code2 == code1: continue
+            if code1 != 'SMILES':
+                if not hasattr(self, "{}_to_{}".format(code1, code2)):
+                    if hasattr(self, "{}_to_{}".format(code2, code1)):
+                        exec("self.{0}_to_{1} = self.mapping_reverse_dict(self.{1}_to_{0})".format(code1, code2))
+                    else:
+                        shortest_path = nx.shortest_path(self.G, source=code1, target=code2)
+                        key = shortest_path[0]
+                        for i in range(len(shortest_path) - 2):
+                            mid, value = shortest_path[i+1: i+3]
+                            self.map_combine(key, mid, value)
+                            self.G.add_edge(key, value)
+                print ("mapping finished: {} -> {}".format(code1, code2))
+
+            if code2 != 'SMILES':
+                if not hasattr(self, "{}_to_{}".format(code2, code1)) and (code2 != 'SMILES'):
+                    exec("self.{0}_to_{1} = self.mapping_reverse_dict(self.{1}_to_{0})".format(code2, code1))
+                print ("mapping finished: {} -> {}".format(code2, code1))
diff --git a/dist/MedCode-0.1-py3-none-any.whl b/dist/MedCode-0.1-py3-none-any.whl
diff --git a/dist/MedCode-0.1.1-py2.py3-none-any.whl b/dist/MedCode-0.1.1-py2.py3-none-any.whl
diff --git a/dist/MedCode-0.1.tar.gz b/dist/MedCode-0.1.tar.gz
diff --git a/dist/MedCodeMap-0.1-py3-none-any.whl b/dist/MedCodeMap-0.1-py3-none-any.whl
diff --git a/dist/MedCodeMap-0.1.tar.gz b/dist/MedCodeMap-0.1.tar.gz
diff --git a/setup.py b/setup.py
@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="MedCode",
-    version="0.1",
+    version="0.1.1",
     author="Chaoqi Yang",
     author_email="chaoqiy2@illinois.edu",
     description="A package for medical code mapping",

diff --git a/tests/.ipynb_checkpoints/Examples-checkpoint.ipynb b/tests/.ipynb_checkpoints/Examples-checkpoint.ipynb
@@ -0,0 +1,185 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "a4da1d02",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# import\n",
+    "from MedCode import CodeMapping"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "6d13de5e",
+   "metadata": {},
+   "source": [
+    "# Usage 1: load the all mappings during initialization"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "35f979d5",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "----- preparing code mappings -----\n",
+      "source 1 loaded from https://drive.google.com/uc?id=1I2G6fsBDXDiAK95qFWwtnl3Ib2MaLeCx\n",
+      "source 2 loaded from https://drive.google.com/uc?id=1d2HzsByXrPadvjaKDOEaOt78OkAZOrjC\n",
+      "source 3 loaded from https://drive.google.com/uc?id=199i8mP2gMQNhwUe-2ZNmIr5fhiBbzVlK\n",
+      "source 4 loaded from https://drive.google.com/uc?id=1Z11J4st1sI44jPborls9jIxzcpF-GpGt\n",
+      "mapping finished: NDC -> RXCUI\n",
+      "mapping finished: RXCUI -> NDC\n",
+      "mapping finished: NDC -> Name\n",
+      "mapping finished: Name -> NDC\n",
+      "mapping finished: NDC -> SMILES\n",
+      "mapping finished: RXCUI -> Name\n",
+      "mapping finished: Name -> RXCUI\n",
+      "mapping finished: RXCUI -> SMILES\n",
+      "mapping finished: Name -> SMILES\n",
+      "load time: 8.500431776046753s\n",
+      "-----------------------------------------\n"
+     ]
+    }
+   ],
+   "source": [
+    "# initialize with a list of supported coding systems\n",
+    "tool = CodeMapping('NDC', 'RXCUI', 'Name', 'SMILES')\n",
+    "tool.load_mapping()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "39a94b85",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['C[C@H]1OC(OC2CC(O)C(C(O)CC(=O)CC(O)C(O)CCC(O)CC(O)CC(O)CC(=O)OC(C)C(C)C(O)C(C)\\\\C=C\\\\C=C\\\\CC\\\\C=C\\\\C=C\\\\C=C\\\\C=C\\\\2)C(O)=O)[C@@H](O)[C@@H](N)[C@@H]1O',\n",
+       " 'CNC(C)C1CCC(N)C(OC2C(N)CC(N)C(OC3OCC(C)(O)C(NC)C3O)C2O)O1']"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# usage 1\n",
+    "tool.RXCUI_to_SMILES['312055']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "94141162",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['Azithromycin']"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# usage 2\n",
+    "tool.NDC_to_Name['76413-153-06']"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ad333a1c",
+   "metadata": {},
+   "source": [
+    "# Usage 2: load additional mappings later on"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "ef4f013b",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "mapping finished: RxNorm -> NDC\n",
+      "mapping finished: NDC -> RxNorm\n",
+      "mapping finished: RxNorm -> RXCUI\n",
+      "mapping finished: RXCUI -> RxNorm\n",
+      "mapping finished: RxNorm -> Name\n",
+      "mapping finished: Name -> RxNorm\n",
+      "mapping finished: RxNorm -> SMILES\n"
+     ]
+    }
+   ],
+   "source": [
+    "tool.add_code(\"RxNorm\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "9fb8b3e2",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['68084027801', '59762306003']"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "tool.NDC_to_RxNorm['76413-153-06']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3166f96e",
+   "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": "3.9.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}