Our focus is on development of algorithms for risk prediction, preventive care, and diagnostics using multiple disparate sources of data. Our developments are based on advances in machine learning, data sciences, and epidemiology.
Identifying which patients are most likely to benefit from an intervention, treatment decisions, prevention of avoidable hospital admissions, rapid prototyping and evaluation of risk stratification algorithms using partially missing clinical, molecular, and/or telemetric measurements.
Using summary level data to develop quantitative, analytical approaches to evaluating optimal healthcare, identifying “policy markers” –bottlenecks preventing optimal care, identifying causes of inefficiency or readmission, analysis of multiple scenarios to evaluate evidence-based policy implementation.
We are able to provide tailored high-performance data cleaning and integration solutions that can deliver a x100 cost reduction (from 3 person-months to 6 laptop-hours for an -omics platform).
The absolute minimal data sets required for this work include: 1) patient-reported outcomes, 2) social determinants of health data, and 3) activity-based costing. Without these, organisations can never achieve value-based care- managing populations of health and creating better patient outcomes for an efficient cost.