Merge pull request #2 from nhsx/directed

Directed and Successor Pages
nhsx · May 30, 2023 · dedcb06 · dedcb06
2 parents a4533ea + 618d25e
commit dedcb06
Show file tree

Hide file tree

Showing 21 changed files with 2,291 additions and 4,550 deletions.
diff --git a/.vscode/settings.json b/.vscode/settings.json
@@ -0,0 +1,6 @@
+{
+    "[python]": {
+        "editor.defaultFormatter": "ms-python.black-formatter"
+    },
+    "python.formatting.provider": "none"
+}
diff --git a/markdown_text/graphs.txt b/markdown_text/graphs.txt
@@ -3,8 +3,6 @@ Using hypergraphs enables more complexities and relationships from the data to b
 _Why?_ Let's start by describing __undirected graphs__, __directed graphs__, __undirected hypergraphs__ and __directed 
 hypergraphs__. 
 
-### Graphs
-
 Graphs are very good at structuring interactivity and networks, which makes them particularly useful in healthcare where healthcare 
 data can be sparse.
 

diff --git a/markdown_text/mm_description.txt b/markdown_text/mm_description.txt
@@ -1,5 +1,3 @@
-## Multimorbidity
-
 Multimorbidity is defined as having 2 or more long term chronic health conditions simultaneously. Multimorbidity is  associated with increased health service utilisation (Cassell et al., 2018).
 
 People are living longer and so the population is ageing. This is due to 

diff --git a/markdown_text/overview.txt b/markdown_text/overview.txt
@@ -1,10 +1,17 @@
 ### Overview
 
+This work is a continuation of the hypergraphs work carried out by Rafferty et al. 2021:
+
+[Multimorbidity Hypergraphs GitHub](https://github.com/SwanseaUniversityMedical/multimorbidity_hypergraphs), 
+[Original Paper](https://www.sciencedirect.com/science/article/pii/S1532046421002458).
+
 We wish to demonstrate how hypergraphs can be used to capture relationships between multiple diseases 
 rather than relationships between pairs of diseases. For example, looking at multimorbidity 
 patterns instead of just comorbidities, which standard graphs tend to do. More people are suffering 
 from multiple chronic diseases at the same time, so it's important to think about how we can find 
 patterns in the uptake of multimoribidities, to see if prevention techniques or treatments could 
 be put in place sooner to improve the quality of life in patients. 
 
-
+On this page you can find out the [aims of this project](#project-aims), find out a bit more about 
+what [multimorbidity](#multimorbidity) is and what types of [graphs](#graphs) we talk about within this project  
+(including standard graphs, undirected hypergraphs and directed hypergraphs).
diff --git a/markdown_text/prototype.txt b/markdown_text/prototype.txt
@@ -5,9 +5,4 @@ and any content here should __not__ be regarded as official output in any form.
 For more information about NHS England please visit our official 
 [website](https://www.england.nhs.uk/)._
 
-This applet randomly generates a set of fictious 'patients' and their disease 
-pathways to demonstrate the use of hypergraphs in understanding multimorbidity. The 
-sidebar on the left of this page can be used to change the number of 'patients' to 
-generate and the maximum number of diseases to include in their pathways.
-In changing the number of patients and diseases, the outputs will change 
-and this is purposeful to enable you to observe how population alterations result in different outcomes.
+
diff --git a/markdown_text/purpose.txt b/markdown_text/purpose.txt
@@ -2,8 +2,12 @@
 
 The model shown in this applet builds upon the work carried out in Rafferty et al., 2021.
 This Streamlit applet can be used to learn about how the hypergraphs multimorbidity model works with fictitious patient data. 
-It provides test cases to allow users to understand the model and apply the `hypergraphs_mm` package to their own data.
+It provides test cases to allow users to understand the model and apply the `hypergraphs_mm` package to their own data 
+(the `hypergraph-mm` package is currently held within a private NHSX GitHub repository but will soon become available publically and as a PyPI package).
 It aims to explain graphs and hypergraphs in a health setting. It can be used to demonstrate the complexity of different types of 
 graphs (left sidebar 'Show Maximum Number of Edges').
+This website gives a interactive walkthrough of how to calulcate hyperedge weights, hyperarc weights, undirected hypergraph centrality, 
+transition matrices and PageRank. The user can navigate between the pages using the tabs `Undirected Hypergraph`, `Directed Hypergraph` 
+and `Successors` at the top of each page.
 It demonstrates a method to draw B-hypergraphs using the NetworkX package, where psuedonodes are used for hyperarc tails to 
 transition to hyperarc heads.
diff --git a/markdown_text/summary.txt b/markdown_text/summary.txt
@@ -0,0 +1,6 @@
+This applet randomly generates a set of fictious 'patients' and their disease 
+pathways to demonstrate the use of hypergraphs in understanding multimorbidity. The 
+sidebar on the left of this page can be used to change the number of 'patients' to 
+generate and the maximum number of diseases to include in their pathways.
+In changing the number of patients and diseases, the outputs will change 
+and this is purposeful to enable you to observe how population alterations result in different outcomes.
diff --git a/requirements.txt b/requirements.txt
@@ -9,7 +9,6 @@ networkx
 hypernetx
 matplotlib
 numpy
-numba
 pandas
 seaborn
 streamlit