Skip to main content

Machine Learning for Healthcare and Life Sciences

IBM Research - Haifa

Real World Evidence


Recent years have seen a dramatic increase in the availability of data collected in the practice of heathcare. If analyzed properly, this data (known as real world evidence or RWE) has the potential to transform the healthcare and life sciences space. Analyzing RWE can benefit all major healthcare stakeholders. Payers and providers can better personalize care, based on data about treatments and outcomes in practice. They can also perform comparative effectiveness studies to compare drugs and treatments. Regulators and policy makers can better define policies that increase healthcare value and safety. Pharmaceutical companies are especially looking into RWE as a new growth driver. Pharma companies use RWE to deepen their understanding of diseases and treatments and better direct the design of products; improve their mechanism for recruiting patients for clinical trials; gain and defend market access; and lead innovative smart solutions, such as clinical decision support for personalized treatments, personalized dosage optimization, and adherence analysis.

The Big Data comprising RWE originates from diverse sources: Electronic medical records (EMR), claims data, literature, and even social media. It can come in a structured form or as free text and includes data about every aspect of patients' journeys. Thus, RWE is not only vast but also varied in type and source. This poses an analytical challenge – how can we gain the most insights, information, and ultimately benefit from RWE?

Our team focuses on using machine learning and deep analytics to derive benefit from RWE data. We have conducted several projects analyzing RWE that span various data types, and various disease areas including epilepsy, diabetes, metabolic syndrome, and mental illnesses.

For example, in 2013 we partnered with UCB, a global biopharma company, to provide a solution for optimizing treatment choices for epilepsy patients, using predictive analytics and patient similarity. See more info in the press release and brochure.

In conjunction with our RWE research project, we have built a unique stack of analytics tools to support semi-automatic data mining and predictive analytics workflows for RWE analysis.

Contact: Yaara Goldschmidt