inVerse Data Science components

Volv’s approach is unique and provides state of the art performance defining disease cohorts that respond to specific therapies more accurately.

Interest in personalized medicine has taken off, and increasingly payers and regulators are looking to understand which treatment is tailored to the individual characteristics of patients. Achieving the objectives of precision healthcare will require clinically-grounded, evidence-based approaches, which in turn demands rigorous, scalable predictive analytics.

Standard strategies for deriving prediction models for medicine involve acquiring ‘training’ data for large numbers of patients, labelling each patient according to the outcome of interest, and then using the labelled examples to learn to predict the outcome for new patients.

Unfortunately, labelling individuals is time-consuming and expertise-intensive in medical applications and thus represents a major impediment to practical personalized medicine. We overcome this obstacle with a novel machine learning algorithm that enables individual-level prediction models to be induced from aggregate-level labelled data, which is readily-available in many health domains. We have demonstrated the utility of the our Volv learning from aggregates methodology in the following fields:

1. leveraging US county-level mental health statistics to create a screening tool which detects individuals suffering from depression based upon their Twitter activity;
2. designing a decision-support system that exploits aggregate clinical trials data on multiple sclerosis (MS) treatment to predict which therapy would work best for the presenting patient;
3. employing group-level clinical trials data to induce a model able to find those MS patients likely to be helped by an experimental therapy
   

Volv is exploring the efficacy of the approach for other diseases that have a high impact on healthcare cost burden.