-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathwork.qmd
More file actions
121 lines (77 loc) · 7.31 KB
/
Copy pathwork.qmd
File metadata and controls
121 lines (77 loc) · 7.31 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
---
title: "Work"
format: html
---
From current workplace (unless specified otherwise): IT department at University of Oslo
# Courses and workshops held
[Introduction to R and RStudio](https://www.uio.no/tjenester/it/kurs/alle/dht/r/events/2024-03-rstudio-intro.html)
[Carpentry@UiO workshop: R programming - a crash course for beginners](https://www.ub.uio.no/english/courses-events/courses/carpentry/251020_r)
[Introduction to Machine learning in R: Classification ](https://www.uio.no/tjenester/it/kurs/alle/dht/ml/events/2024-04-ml-r-intro.html)
[Introduction to Machine learning in Python: Classification ](https://www.uio.no/tjenester/it/kurs/alle/dht/ml-python/events/2024-05-ml-python-intro.html)
[Getting started with statistics and programming using GPT UiO](https://www.uio.no/tjenester/it/kurs/alle/dht/gpt-stats-programmering/events/2024-05-gpt-statistics.html)
[Workshop: Getting up-to-speed with Python](https://www.uio.no/dscience/english/news-and-events/events/workshop-getting-up-to-speed-with-python-.html)
[Deep learning in R: Image classification with Torch](https://www.ub.uio.no/english/courses-events/events/dsc/2025/digital-scholarship-days/06-deep-learning-in-r)
# Selected work projects {#sec-work-proj}
[POICE Platform](https://www.uio.no/dscience/english/data-resources/projects/poice/): Main contributor to the POICE image processing platform (python) in collaboration with [dScience](https://www.uio.no/dscience/english/data-resources/)
[RASK](https://www.antibiotika.no/asp/riktigere-antibiotikabruk/rask/): Automated HTML reports with R and Quarto. Plots and tables summarizing the use of antibiotics Norway, aggregated by state and years.
[Digikost](https://www.med.uio.no/imb/forskning/om/infrastruktur/kostholdsmetodikk/digikost.html): Automated HTML reports using RMarkdown. Data ingested from UiO Nettskjema in TSD
[MinMat](https://www.med.uio.no/imb/om/aktuelt/aktuelle-saker/2021/bli-med-pa-forskning-pa-kosthold-hos-eldre.html): Wrote R code that parses UiO Nettskjema data in TSD. Data ingested via a mobile app through UiO nettskjema. Later rewrote code to python.
[iTandem](https://www.oslo-universitetssykehus.no/itandem): Wrote R code that parses UiO Nettskjema data in TSD. Data ingested via a mobile app through UiO nettskjema. Later rewrote code to python.
[NORTRAS](https://www.sv.uio.no/psi/forskning/prosjekter/nortras/): Wrote an R Shiny app that vizualises data (graphs and tables) for the NORTRAS study
[MinDag](https://www.med.uio.no/om/aktuelt/blogg/2022/app-hjelpe-forsta-kompliserte-psykiskiske-lidelser.html): Wrote an R Shiny app that creates HTML reports based on data delivered from an app (connected to UiO Nettskjema)
# UiO-IT Hackaweek 2026
Participated in the [UiO-IT internal hackaweek 2026](https://www.usit.uio.no/om/aktuelt/2026/hackaweek.html) (March). The topic of the hackaweek was AI: UiO-IT let gave us API access to (new) models [GPT.uio.no](https://gpt.uio.no/), with a lot more resources
than normal. Some of these were local models hosted at [NTNU](https://www.hpc.ntnu.no/idun/documentation/ai-coding-assistant-and-large-language-models-llms-on-idun/) (i.e. Kimi and GLM), although we could also make use of fully external services such as Sonnet models via Claude Code. The "section" I am in (data management) decided to collectively work on improving the [Unimusportal](https://www.unimus.no/portal/), which is a webservice
that lets you search information and photographs across the five university museums in Norway. Specifically, we configured coding agents (e.g. Opencode) to use the models to
improve specific features of the Unimusportal. What made this endeavour an interesting "exercise", is that the codebase is written in flutter and dart, something only the maintainer is proficient in
<br>
<br>
In the end, I ended up deploying two new features:
* navigating photos belonging to the same "object" with a "carousel"
* option of generating a PDF of an object/finding with requisite (meta)data
{width=80%}
<br>
*Original view: see see [link](https://unimus.no/portal/#/photos/adea6b79-b234-4a29-bab5-8cfb2bf7e302?thing=43b05f2c-d066-4eec-ad68-0a6be9152fe0)*
{width=80%}
<br>
*View after implementing photo "carousel": see [link](https://unimus.no/portal-utv/#/photos/adea6b79-b234-4a29-bab5-8cfb2bf7e302?thing=43b05f2c-d066-4eec-ad68-0a6be9152fe0)*
{width=80%}
<br>
*Example of (new) PDF button: compare [before](https://unimus.no/portal/#/things/4db27529-03e6-4ce0-84e1-d7efdc05697f) and [after](https://unimus.no/portal-utv/#/things/4db27529-03e6-4ce0-84e1-d7efdc05697f)*
{width=60%}
<br>
*Example of generated PDF file (can test in the [after](https://unimus.no/portal-utv/#/things/4db27529-03e6-4ce0-84e1-d7efdc05697f) link above)*
# Talks and presentations
{{< video files/ML_lynkurs_IT_konferanse_2021.mp4 width="600" height="350" title="IT konferanse" >}}
Crash course (Norwegian) in how to use tidymodels for building machine learning models.
<br>
From [IT conference at UiO 2021](https://www.uio.no/for-ansatte/enhetssider/arrangement/it-konferansen/2021/)
<br>
{{< video files/NORTRAS_app_data_at_uio_2022.mp4 width="600" height="350" title="NORTRAS demo" >}}
Presented NORTRAS shiny app (Norwegian) with dummy data (see projects section), at [data@UiO 2022](https://www.usit.uio.no/om/arrangementer/data/2022/)
[Talk](files/DMN_R_in_TSD.pdf) at [UiO Data Manager Network](https://www.ub.uio.no/english/libraries/dsc/data-managers-network/) october 2022 about how R is used in production in TSD.
<br>
Mentions some of the projects in [Selected work projects](#sec-work-proj)
[Talk](files/UiO_ConcePTION_Workshop_march_2023.pdf) discussing an outsider (UiO IT) perspective of the [ConcePTION](https://www.imi-conception.eu/) project.
<br>
<br>
{{< video files/R_live_educloud_ondemand.mp4 width="600" height="350" title="R educloud live" >}}
Live coding in R (short notice backup), showcasing [Educloud On Demand](https://www.uio.no/tjenester/it/forskning/beregning/open-on-demand/) (run software on HPC interactively) at [data@UiO 2024](https://www.usit.uio.no/om/arrangementer/data/2024/)
# Scientific articles
From my days as a PhD researcher or in my spare time
## PhD
PhD thesis:
<br>
[Elucidating depression heterogeneity using clinical, neuroimaging and genetic data](https://www.duo.uio.no/handle/10852/71645)
Article 1:
<br>
[Data-driven clustering reveals a link between symptoms and functional brain connectivity in depression](https://www.sciencedirect.com/science/article/abs/pii/S2451902218301290)
Article 2:
<br>
[Brain connectome mapping of complex human traits and their polygenic architecture using machine learning](https://www.sciencedirect.com/science/article/abs/pii/S0006322319317913)
Article 3:
<br>
[Multimodal fusion of structural and functional brain imaging in depression using linked independent component analysis](https://onlinelibrary.wiley.com/doi/full/10.1002/hbm.24802)
## Selected co-author papers
[Inflammation and cognition in severe mental illness: patterns of covariation and subgroups](https://www.nature.com/articles/s41380-022-01924-w)
[Data-driven health deficit assessment improves a frailty index’s prediction of current cognitive status and future conversion to dementia: results from ADNI](https://link.springer.com/article/10.1007/s11357-022-00669-2)