Healthcare · Clinical Trial Site Selection ML

Tier-1 Pharmaceutical (via Deloitte).

Role: Data Scientist (Consulting)Life Sciences · ML · Trial OperationsPrior experience — Michele Sanna at DeloitteStatus: Completed

Site-performance prediction model over real-world data, EHR-derived epidemiology, investigator publication history, and regulatory inspection records. Combined with a portfolio optimiser (mixed-integer programming) for site mix selection.

Python
LightGBM
SHAP
Gurobi
Azure ML
Databricks

Scroll to explore →

first-patient-in speed across two adopted oncology trials

enrolment target hit rate (vs. ~65% historical baseline)

saved in avoided timeline extensions

Next project

Regione Sardegna (via Accenture)

Want this kind of work in your business?

Start with the problem, not the platform. Tell us what's slow, manual or breaking.

Send a project brief Or just say hello