PM.20.039 – Towards data-driven and applicable personalized medicine
Clinical practice guidelines are recommendations to optimize patient care informed by a systematic review of evidence and assessment of benefits and harms of management options. Guidelines are based on ‘the average patient’ and need translation to the individual patient in clinical practice.
We developed an early version of a personalized decision tool in the form of an app for treatment decision making in hepatitis C. This tool holds the potential to be extended across a wide range of diseases and innovate clinical medicine profoundly. Further research is required regarding data acquisition, data analysis and synthesis, and clinical implementation.
This research proposal focuses on fundamental scientific research as well as technological implementation and user adoption.
Fundamental research questions are:
1. How can we synthesize individual patient data from classical published studies and from electronic patient registries or records into a large trustworthy data set, to be used for data-driven personalized medicine? How can AI contribute to automated rigorous updating of databases?
2. Which statistical and machine learning methods are most appropriate to estimate individualized treatment benefit and harm? How can we synthesize forthcoming evidence to keep estimates up-to-date?
The scientific methods that emerge from addressing these questions will be translated into an improved clinical decision support tool, which displays personalized information on efficacy, adverse effects and costs of therapies. It is to be fitted into the electronic workflow of doctors, with the aim of revolutionizing the use of decision support in medical practice.
A further research question is:
3. How to optimize user adoption? What will encourage clinicians to adopt tools for data-driven personalized medicine?
As medical areas of interest, we plan to expand from hepatitis C to treatment decision-making in hypertension and heart failure.
app, Artificial Intelligence, clinical data synthesis, decision support, individual treatment effects, Personalized medicine, prediction modelling, user adoption
Erasmus Medical Center (EMC), ExpertDoc B.V., MUMC+, TherapySelector B.V., Zorgkeuzelab B.V.
|Organisation||LUMC, Leiden, Netherlands|
|Name||Prof. dr. E. (Ewout) Steyerberg|