← All teams
Cohort 13
ParamAIbringing promising medicines to the people waiting
Missing information about drug behavior makes it much harder for drug developers to choose effective, tolerable doses, increasing development costs and delaying promising medicines for patients.
Led by Anish (Max) BaggaVisit ParamAI
The problem
Mathematical models help drug developers choose doses, but the models depend on measurements of drug properties that can be hard to get. When those measurements are missing, dosing questions stay unresolved. Answering them can mean additional studies that cost millions, and researchers estimate about eight months for a full modeling project.
The solution
ParamAI is developing a machine learning model that estimates hard-to-measure drug parameters from evidence developers have already accumulated. ParamAI solves the problem by enabling teams to build pharmacology models faster, resolve dosing questions at lower cost, and accelerate regulatory review so promising medicines reach patients sooner.
How it works
01Use existing evidence
Developers already hold studies and assays. One critical question often remains, such as how quickly the drug leaves the body, and it is hard to measure in humans.
02Estimate drug properties
ParamAI estimates clearance, metabolism, and the other drug properties that set dose and timing.
03Inform dosing decisions
The estimates show which dose to favor, with the justification, evidence, and assumptions a regulatory review asks for.
The team