Output details

Quantifying short-term dynamics of Parkinson's disease using self-reported symptom data from an internet social network

<28> Shows how a recent phenomenon in computer science, massive health datasets from social networks, can be processed with computationally efficient data analysis algorithms, to perform networked medical discovery. Self reported symptom progression data is processed by new techniques which introduce ‘convex optimization’ into pharmacodynamic/pharmacokinetic regression methods – the basis of all pharmaceutical industry drug treatment efficacy studies. This work contributed towards the award of the second half of an MIT-Wellcome Trust Fellowship (WT090651, worth approx. $338,000 over 4 years) which forms the basis of ongoing pharmaceutical industry research collaborations (Sage Bionetworks, friend@sagebase.org, Roche, anirvan.ghosh@roche.com).