Early identification of at-risk patients.
Machine learning on hospital data and clinical notes supports the early identification of people at risk of PAH.
Spot signals that warrant investigation earlier.
Pulmonary arterial hypertension (PAH) can be recognised late. The project explored how existing hospital data could highlight useful signals to guide further clinical investigation.
- Machine learning
- Clinical data
- Decision support
What we designed.
The project links information distributed across care pathways with predictive capabilities. Alerts flag elements for further investigation; diagnosis and decisions remain clinical.
Analyse care pathways
We analysed hospital utilisation data and clinical notes to identify recurring elements in patient pathways.
Identify patterns
Machine learning models were developed to investigate associations in clinical data and healthcare service use.
Support early alerts
The signals feed an alert system supporting clinicians, with the aim of identifying people for assessment and informing screening.
Care pathways and clinical notes
Machine learning patterns
Signals for clinicians
What it makespossible.
- Development of an early alert system.
- Identification of patterns in hospital care pathways.
- Support for selecting further clinical investigations.
Want to develop analytical tools for healthcare data and pathways?
Let’s start with your context, available data and the first useful outcome.
Tell us about your project
Humanoid robotics in production.
Bending, welding and stamping. An assessment of processes and costs to design the first pilot and integration roadmap.