We don’t have better algorithms than anyone else; we just have more data Peter Norvig, Director of Research, Google
Artificial intelligence (AI) is the dominant theme of our time. From futuristic self-driving cars and smart assistants to the more mundane photo apps and smartphones, AI is at the center of transforming our society. But beyond attention grabbing applications, AI is also helping every industry optimize services, lower costs, accelerate processes, and make better decisions.
The healthcare sector is no different. McKinsey estimates that AI and advanced analytics can unlock $122B in value for the life sciences industry. Though in its initial stages, intelligent algorithms are being used for early detection, recommending customized treatments and for optimizing clinical trials.
Yet for all its promise, realizing the benefits of AI in healthcare isn’t easy. The biggest challenge with building smart systems is the availability of good data. By some estimates almost 80% to 90% of all advanced analytics projects’ time is spent gathering and cleaning the data, not in building models.
For most life science companies, the hurdle of clean, linked data is simply too onerous, hampering their efforts to realize the full benefits of advanced analytics and AI in their applications.
Unfortunately the choices for advanced healthcare intelligence have been limited. Legacy data solutions haven’t kept up with the rapid changes in technology. Their data aggregation methodology are an artifact of a previous generation, relying on manual surveys, phone verification and web scraping. Consequently their data is often riddled with incorrect, incomplete or stale information.
Enter Healthbase. Built using the largest set of real-world claims activity and public data, Healthbase uses proprietary machine-learning algorithms and big data techniques to map the US healthcare market. From large health systems and accountable care organizations (ACOs) to individual practitioners and physician groups, Healthbase gives a clear picture into entities and their relationships.
Figure 1. A snapshot of the graph-linked structure of Mayo Clinic Health System
The full data and feature set are quite exhaustive, but a few reasons on why we believe Healthbase is the data solution for next generation healthcare AI and analytics.
- Maps affiliations between HCOs (Healthcare Organizations) and HCPs (Healthcare Professionals) in real-time by triangulating data from multiple data sets that include real world data, tax filings, CMS records and regulatory documents, among others.
- Computes the strength and nature of relationship between entities. The strength is recorded as a simple score between 1-100, allowing applications to run custom rules based on affiliation type.
- The data model underlying Heathbase is a graph database that enables dynamic and complex relationship between entities. The graph database also extends analytic insights by essentially creating an adaptable query engine.
Figure 2: Mayo Clinic Health System as viewed in Healthbase’s web app
No other solution in the market offers such advanced data capabilities. The Healthbase platform is available in different form factors from a simple user-driven web app to the more complex relational and graph databases.
Healthbase’s data has benchmarked against competing solutions and the quality and depth of our data has been noticeably superior.
Over the coming weeks and months, we’ll be using this space to share insights derived from our data. If you like what you’ve read, head over to gethealthbase.com to learn more.
Tech companies are getting serious about their intention to remake health care by leveraging AI and machine learning. But instead of the “big bang” change, perhaps this disruption will occur in small steps.