With an eye toward democratizing predictive analytics and machine learning, healthcare Big Data company Health Catalyst is moving into the open-source world. Thursday, Health Catalyst introduced healthcare.ai, which the Salt Lake City-based company called the first healthcare-specific repository of open-source machine-learning software.
“We want to enable machine learning,” said Levi Thatcher, Health Catalyst’s director of data science. “We’re trying to form a healthcare machine-learning community,” Thatcher said.
“By open sourcing healthcare.ai, we hope to facilitate industrywide collaboration and advance the adoption of machine learning, making it easy for healthcare organizations to learn from and enhance these tools together, without the need for a team of data scientists,” Health Catalyst Executive Vice President Dale Sanders added in a press release.
In an interview with MedCity News, Thatcher noted that predictive analytics in healthcare so far has been the nearly exclusive province of academic medical centers. Health Catalyst itself grew out of Intermountain Healthcare, which has several residency programs and a large research operation.
Community hospitals — even big ones — simply have not had the resources to hire data scientists, and are badly behind in the world of analytics. “Many are relying on heuristics that are a decade or two old,” Thatcher said.
“We’re looking at mid-size to large hospitals,” Thatcher said.
A Deep-dive Into Specialty Pharma
A specialty drug is a class of prescription medications used to treat complex, chronic or rare medical conditions. Although this classification was originally intended to define the treatment of rare, also termed “orphan” diseases, affecting fewer than 200,000 people in the US, more recently, specialty drugs have emerged as the cornerstone of treatment for chronic and complex diseases such as cancer, autoimmune conditions, diabetes, hepatitis C, and HIV/AIDS.
With healthcare.ai, healthcare facilities will be able to optimize analytics and processes based on that organization’s own history of practice and patient outcomes. “[Machine-learning] models will learn based on past cases,” according to Thatcher.
From there, organizations will be able to stratify patient risk based on what has worked best in the past and what needs improvement, he explained. The technology should be able to help with operational efficiency, financial efficiency and clinical excellence, Thatcher said.
Users can go to the healthcare.ai website to download and then customize software packages written in either the R or Python languages, which Health Catalyst said are popular in healthcare data science. Current software “of course integrates very well with our late-binding data warehouse,” Thatcher said.
He said Health Catalyst would be an active member of the open-source community. The company already has leaned on healthcare.ai to build predictive models for reducing hospital readmissions, determining the likelihood of patients to pay and for optimizing clinician schedules.
Photo: Flickr user David Lofink
