How Cornerstone AI makes data ‘right’ for the healthcare sector

    To further strengthen our commitment to providing industry-leading data technology coverage, VentureBeat is pleased to welcome Andrew Brust and Tony Baer as regular contributors. Look for their articles in the Data Pipeline.

    AI has the potential to transform healthcare. Whether predicting the risk of terminal diseases or developing new drugs, companies are leveraging data-driven algorithms to improve the quality of patient care in every way. The use cases are expected to only grow from here, but there are certain hurdles along the way as well. Example: the lack of high-quality data sets.

    Health organizations cumulatively generate about 300 petabytes of data every day. This information is stored in various systems, but is not used effectively due to poor preparation. In short, data teams, who tend to create manual rules for data cleansing, are struggling to keep up with the growing amounts of information. They spend most of their time, nearly 80%, preparing the data — to make it accurate, connected, and standardized — rather than actually exploring and analyzing it for potential, life-saving AI applications.

    Cornerstone AI . Comprehensive Solution

    To fix this, the San Francisco-based company, cornerstone AI has launched a solution that automatically characterizes, harmonizes and cleans healthcare data in a fraction of the time it takes traditional methods. The company also announced that it has raised $5 million in seed funding.

    According to Cornerstone, his platform’s algorithm uses a combination of custom Python and R code to scan every table and data point — to infer their structure and validity — then organize the tables for analysis, while keeping all notable ones errors are removed and corrected.

    “A data team doesn’t need to configure anything in the system other than telling them what the patient ID field is. The system automatically learns the data structure and then automatically learns the patterns in the data. Data teams can get started in the system on day one and see the AI ​​findings in the user interface,” said Michael Elashoff, the company’s co-founder and CEO.

    After correction, the findings are shared as part of a data quality report.


    Although the company is still in its early stages, it has implemented its solution at quite a few healthcare companies. In one case, a medical device company that previously spent six months cleaning up data was able to speed up the process twenty times or just nine days. The system already covers the full range of structured and semi-structured healthcare data, starting with medical records, clinical trials, registry data, claims, digital health and sensor data.

    “In a recent validation study we did, the system identified 98% of data issues, with a specificity of about 99.9%,” added the CEO, claiming the platform can run 750 million records in about two hours.

    He also clarified that unstructured information, such as faxes or pathology reports, will remain out of reach of the platform at least from now on.

    Plan ahead

    With this funding round, led by Healthy Ventures, Cornerstone plans to further develop its product and rope for more customers, possibly long-term contracts.

    “Customers have told us that the machine learning (ML) models our system builds for data cleansing have applications that go beyond getting a high-quality data set,” said Elashoff. “For example, the patterns and relationships the system identifies can be used to identify patients whose treatment response or surgical recovery is different from what it should be. In those cases, the software identifies potential clinical insights that may have been hidden in the complexity of the data.” So we use the funding to build out this functionality so that companies can get much more out of their data.”

    Other notable players in data cleansing and preparation are Datadog and New Relic, but they are not specific to the healthcare sector, such as Cornerstone AI.

    “We developed the algorithm specifically to work with medical data, with its high complexity and high error rate. We had to develop brand new ML techniques to ensure that our models were not thrown off guard by the errors we are trying to find,” emphasized the CEO.

    Apart from this, unlike other systems, the company’s platform generates an explanation for every problem it finds and provides a built-in regulatory-grade audit trail that tracks all changes.

    The mission of VentureBeat is a digital city square for tech decision makers to gain knowledge about transformative business technology and transactions. Learn more about membership.

    Recent Articles

    Related Stories

    Leave A Reply

    Please enter your comment!
    Please enter your name here

    Stay on op - Ge the daily news in your inbox