-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.qmd
More file actions
165 lines (111 loc) · 5.33 KB
/
index.qmd
File metadata and controls
165 lines (111 loc) · 5.33 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
---
format:
revealjs:
theme: ["styles.scss"]
footer: "<https://github.com/pythonhealthdatascience/stars_wp4_presentation>"
height: 800
width: 1250
preload-iframes: true
---
# Update on WP4: Training materials
<hr>
:::: {.columns}
::: {.column width="75%"}
<br>
<h3>Amy Heather <a href="https://orcid.org/0000-0002-6596-3479" aria-label="View ORCID record"><img src="images/orcid.png" alt="ORCID iD" style="width:32px; margin-left:15px; margin-right:6px; vertical-align:middle;"></a>0000-0002-6596-3479</h3>
<br>
<br>
Part of the "STARS" project:
:::
::: {.column width="25%"}
:::
::::
## Stages for developing the training materials {auto-animate=true auto-animate-unmatched=false}
:::: {.r-vstack}
::: {.stage-box .evidence}
<div data-id="green" class="stage-icon green"></div>
Gather evidence on requirements for reproducible DES
:::
::: {.stage-box .example}
From theory to practice: build working examples
:::
::: {.stage-box .stroke}
Testing with a real-world stroke model
:::
::: {.stage-box .book}
Knowledge mobilisation: online book
:::
::::
{{< include html/evidence.md >}}
* **Empirical research** - Barriers and facilitators identified from reproducing eight published healthcare DES models (Python/R).
:::: {.columns}
::: {.column width='50%'}
:::
::: {.column width='50%'}
:::
::::
{{< include html/evidence.md >}}
* **Empirical research** - Barriers and facilitators identified from reproducing eight published healthcare DES models (Python/R).
* **NHS RAP guidelines** - Established framework for developing reproducible analytical pipelines (RAP) in the NHS.
{fig-align="center"}
{{< include html/evidence.md >}}
* **Empirical research** - Barriers and facilitators identified from reproducing eight published healthcare DES models (Python/R).
* **NHS RAP guidelines** - Established framework for developing reproducible analytical pipelines (RAP) in the NHS.
These serve different purposes. Distinction between **reproducibility** and a RAP:
{fig-align="center"}
{{< include html/evidence.md >}}
* **Empirical research** - Barriers and facilitators identified from reproducing eight published healthcare DES models (Python/R).
* **NHS RAP guidelines** - Established framework for developing reproducible analytical pipelines (RAP) in the NHS.
These serve different purposes. Distinction between reproducibility and **a RAP**:
{fig-align="center"}
{{< include html/evidence.md >}}
* **Empirical research** - Barriers and facilitators identified from reproducing eight published healthcare DES models (Python/R).
* **NHS RAP guidelines** - Established framework for developing reproducible analytical pipelines (RAP) in the NHS.
* **Established DES modeling practices** - All the necessary stages of creation of a DES model were identified from domain expertise and textbook recommendations.
* **Why?** The RAP should include all the analysis stages of a typical DES modelling process.
{{< include html/example.md >}}
Developed simple M/M/s examples as templates in Python and R that meet all requirements we identified.
::: {.scroll-container style="overflow-y: scroll; height: 700px;"}
<p align="center">
<img src="images/rap_template_r.png"/>
</p>
:::
{{< include html/example.md >}}
Developed simple M/M/s examples as templates in Python (**SimPy**) and R (**simmer**) that meet all requirements we identified.
* Meets all requirements (from Heather et al. (2025) and NHS Levels of RAP) - for example:
* Structured as a package.
* Unit tests, back tests and functional tests.
. . .
* Complementary structure between Python and R models.
. . .
* Python mostly classes, R mostly functions (as is typical for each language).
. . .
* Provides code for simulation with scenarios and sensitivity analysis as well as choice of model parameters (input modelling, warm-up length, number of replications).
{{< include html/stroke.md >}}
Applied the templates to a stroke pathway model described in a published paper. This provided a **real**, more **complex** example and helped **refine** the template.
::: {.scroll-container style="overflow-y: scroll; height: 700px;"}
<p align="center">
<img src="images/rap_stroke_python.png"/>
</p>
:::
{{< include html/book.md >}}
The M/M/s and stroke models demonstrate what reproducible DES looks like, they don't show modellers *how* to get there.
I am developing an online book which breaks down these models, showing how they met each requirement. The book:
* Provides **actionable steps** which guide modellers to create a model which meets all the reproducibility and RAP requirements.
* Allows people to easily choose particular pages of interest (e.g. already have a model, but want to find out more about citation).
{{< include html/book.md >}}
<iframe class="stretch" data-src="https://pythonhealthdatascience.github.io/des_rap_book/"></iframe>
{{< include html/future.md >}}
<br>
**Review:** At a good stage for review of some of these materials.
* Tom has reviewed the M/M/s models.
* Alison - review of R (mentioning before - timing better now?)
* Nav - review of book, if happy to?
* Andy & Chris - at a later stage?
<br>
**Plan:** Continuing to work on these materials. Will describe in a publication.
<br>