What data do we have?

PIONEER includes de-identified/pseudonymised data from patients who were seen by an acute care provider from 1st January 2000 and will include data from patients until the project closes (2025 at the earliest).  Each dataset will be bespoke, creating to match the specific project.

Some of our datasets are listed below, but each can be finessed or expanded to meet your direct requirements, and many other conditions, pathways or therapy areas are available.   If you don’t see what you need, please contact a member of the PIONEER team.

Specialist datasets available

Synthetic Dataset: Hospitalised Patients with Thromboembolic Diagnosis

The incidence of blood clots in the lungs (PE) or limbs (DVT) is estimated to be approximately 50–150 per 100,000 people and in the UK, around 60,000 cases of PE and 200,000 cases of DVT are reported each year. However, despite the significant progress, diagnosing PE and DVT remains a challenge. This large synthetic data with up to 14.5k patients of both suspected and diagnosed thromboembolic events provides key parameters to support critical research into the condition. 

Read More »

Synthetic Dataset: Using Data-driven ML Towards Improving Diagnosis of ACS

Acute compartment syndrome (ACS) is an emergency orthopaedic condition wherein a rapid rise in compartmental pressure compromises blood perfusion to the tissues leading to ischaemia and muscle necrosis.  In this dataset, highly granular synthetic data of over 900 patients with ACS is shown to provide the key parameters to support critical research into this condition. 

Read More »

Machine Learning Frailty Index Estimates with Routine Test Results in Acute Care

Frailty is a critical measure in health care for evidence-based clinical decision making. An accurate electronic Frailty Index (eFI) at admission will be beneficial to both patients and medical service for prompt and appropriate assessment and management in acute care. An eFI that was derived from 31 routinely collected test results showed that it has promising identification power for high-risk frailty patients in aged cohort (>65), indicating the potential of a simpler and more efficient model for frailty estimation.

Read More »