Open source site aims to boost use of machine learning

Register now

A vendor with experience and products in data warehousing and analytics is creating a community for open-source software intended to streamline efforts to use machine learning applications in healthcare.

Health Catalyst has created as a repository of healthcare-focused open source machine learning software, saying that it’s important for the industry to benefit from the technology and democratize machine learning in healthcare. In addition to creating the web site and contributing its tools and algorithms to the open-source community, the company is offering ongoing support to maintain it.

The Salt Lake City-based company says the site will provide one site to download algorithms and tools, contribute code, read documentation and communicate with other healthcare professionals who are interested in using the technology to improve patient care.

Also See: UPMC, Health Catalyst join forces to cut healthcare costs

The initiative hopes to spread interest and use of machine learning and artificial intelligence by serving as a repository for coding work in the healthcare arena.

Health Catalyst says facilitates the development of predictive and pattern recognition models using a healthcare organization’s own data, and it features packages for two common languages in healthcare data science—Python and R.

The packages are intended to streamline healthcare machine learning by simplifying the workflow of creating and deploying models and delivering functionality specific to healthcare. Specifically, they pay attention to longitudinal questions, offer an easy way to do risk-adjusted comparisons, and provide easy connections and deployment to databases.

Both packages provide an easy way to create models on healthcare organizations’ data. This includes linear and random forest models, ways to handle missing data, guidance on feature selection, proper performance metrics and easy database connections.

The use of machine learning and predictive analytics has been limited in healthcare because of challenges in programming solutions, typically by highly trained data scientists, mostly in the nation’s top academic medical centers. But aims to make machine learning accessible to healthcare professionals who have little or no data science skills but who are interested in using the technology to improve patient care. It’s targeted at BI developers, data architects and SQL developers who want to create appropriate and accurate models with healthcare data.

By making its central repository of proven machine learning algorithms freely available, enables a large, diverse group of technical healthcare professionals to quickly use machine learning tools to build accurate models.

“Machine learning and artificial intelligence are going to transform healthcare,” says Dale Sanders, executive vice president of Health Catalyst. “We are seeing amazing results and yet we are barely getting started. With machine learning, the data is talking to us, exposing insights that we’ve never seen before with traditional business intelligence and analytics.

“By open sourcing, we hope to facilitate industry-wide 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,’ Sanders adds “All of us have seen what open source software has achieved in other industries and we want to be a part of that in healthcare.”

Health Catalyst expects to gain benefits by supporting the use of open-source applications for machine learning, he adds.

“These tools will democratize machine learning in a realm that needs it most—because everyone benefits when healthcare is made safer, more efficient and effective,” Sanders says. “We are not just being altruistic here. By submitting our tools and algorithms to the open source community, we and our clients will benefit from the collective intelligence that exists beyond our team of data scientists.”

For reprint and licensing requests for this article, click here.