MSP: Maternal Sleepstages for predicting newborn attributes

This is a guiding template for short-term exploratory projects. Many sections are optional either in their presence or scope and can be excluded based on the project at hand.

• Status: 🟡 In Progress
• Duration: 28th July 2025 - ???
• Type: Data Analysis
• Repository: GitHub Repository
• Team: Fabricio

What we're exploring: We are exploring whether hypnodensities or hynograms of mothers during their pregnancy can be used for predicting properties of infants.

• Take a look at the correlation between the HRV of the child, the SpO2 of the mother, and the HRV of the mother.

A description of the used hardware and software. The scope is up to the judgement of the team.

Suggestions for projects with software:

• Programming Language & Package Versions, Docker image, …
• Notes on setup / required steps to get things working.
• Git Repo (Code, Dockerfiles, Configs, …)

Suggestions for projects with data:

• Data source and when it was acquired/downloaded
• Description of changes to data, ideally with scripts + dependencies in git repo.

• Bullet points of successful approaches

• Failed approaches (and why)
• Performance bottlenecks discovered
• Incompatibilities found

• Unexpected discoveries
• Hidden complexities
• Simpler alternatives found

Should we proceed?: Yes/No/Maybe

If yes, what needs to happen?

• Paper: Neural network analysis of sleep stages enables efficient diagnosis of narcolepsy
• Sleepdata: MSP Dataset