Division of Data Science and Artificial Intelligence
Study Program of Anesthesiology and Intensive Therapy
Faculty of Medicine, Universitas Udayana
A tiered, cohort-based curriculum bridging traditional clinical statistics and modern data science for anesthesiology residents. Built around reproducibility, causal reasoning, and real clinical datasets.
This curriculum is not a generic data science course. Every concept is anchored to anesthesia and critical care practice. Residents leave as researchers who can build, or builders who can publish.
| Tier | Duration | Language | Goal |
|---|---|---|---|
| Introductory | 4 weeks | R (toolchain only) | Shared competence: R, RStudio, Git, GitHub, Quarto |
| Basic | 16 weeks | R | Critical consumer + reproducible researcher |
| Intermediate | 20 weeks | Python | Independent analyst, researcher/builder entry |
| Advanced | 24 weeks | Python + AI-assisted | Clinical tool builder or preprint-ready researcher |
Progression: Linear. Introductory → Basic → Intermediate → Advanced.
Cohort size: 4–6 residents per cohort.
Cadence: Rolling. All three tiers run concurrently at steady state.
Contact hours: 2 hours/week.
Assessment: Capstone project per tier.
- Basic — Quarto report reproducing and critiquing a published anesthesia paper + R code on GitHub
- Intermediate — Quarto manuscript + reproducible Python pipeline using VitalDB or eICU
- Advanced — Deployed clinical tool prototype or preprint-ready manuscript (builder/researcher track)
mds-curriculum/
├── intro/ # Introductory Course (4 weeks)
│ ├── week-01/
│ │ ├── index.qmd # Resident-facing module
│ │ ├── facilitator.qmd # Facilitator notes
│ │ └── data/ # Module dataset or pointer
│ └── ...
├── basic/
│ ├── week-01/
│ │ ├── index.qmd # Resident-facing module
│ │ ├── facilitator.qmd # Facilitator notes
│ │ └── data/ # Module dataset or pointer
│ └── ...
├── intermediate/
├── advanced/
├── datasets/
│ ├── basic/
│ ├── intermediate/
│ └── advanced/ # MIMIC-IV pointers only — data not stored here
├── facilitator/
│ ├── onboarding.qmd
│ └── rubrics/
├── capstone-archive/ # Past cohort capstones (with consent)
├── certificate/
│ ├── template.qmd # Parameterized Quarto PDF template
│ ├── params-example.yml # Example parameter file
│ ├── assets/ # Seal, signature scan
│ └── REGISTRY.yaml # Public verification record
├── _quarto.yml
└── ARCHITECTURE.md
| Tier | Dataset | Access |
|---|---|---|
| Basic | Curated synthetic/public anesthesia datasets | Open |
| Intermediate | VitalDB, eICU-CRD | Open / PhysioNet credentialed |
| Advanced | MIMIC-IV | PhysioNet credentialed (local copy required) |
MIMIC-IV data is never committed to this repository. Modules use relative paths against a local
data/directory which is.gitignored. Reproducibility is guaranteed by code, not data.
Advanced graduates are trained as junior facilitators. Facilitator onboarding is documented in facilitator/onboarding.qmd and versioned with the curriculum. Facilitator handoff is a repo permission grant, not a document transfer.
Residents who complete a tier capstone receive a Certificate of Competency issued by the Division of Data Science and Artificial Intelligence, Study Program of Anesthesiology and Intensive Therapy, Faculty of Medicine, Universitas Udayana.
Signatories:
- Koordinator Program Studi — Prof. Dr. dr. Tjokorda Gde Agung Senapathi, SpAn-TI, Subsp.A.R (K)
- Kepala Divisi Data Science dan Artificial Intelligence — dr. I Made Agus Kresna Sucandra, SpAn-TI, Subsp.T.I (K)
Verification: Every certificate carries a unique ID resolvable at balinesthesia.github.io/mds-curriculum/verify against certificate/REGISTRY.yaml.
Template: Generated from certificate/template.qmd — parameterized Quarto PDF, versioned with the curriculum.
This curriculum follows Semantic Versioning:
- Patch — typo fix, broken link, dataset pointer update
- Minor — module content revision, new exercise, facilitator note addition
- Major — tier restructure, language change, capstone brief rewrite
Each cohort records the curriculum version they completed.
Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
Free to use and adapt for non-commercial academic and clinical education purposes with attribution. Commercial use requires explicit written permission from Balinesthesia.
Attribution: Division of Data Science and Artificial Intelligence, Study Program of Anesthesiology and Intensive Therapy, Faculty of Medicine, Universitas Udayana
This curriculum is maintained by Balinesthesia. Contributions from Advanced graduates are welcome via pull request. External collaborators should fork and submit PRs — see ARCHITECTURE.md for governance details.
See ARCHITECTURE.md for design decisions and rationale before opening a PR.