NEW YORK – Ontada on Thursday said it will use Microsoft's Azure AI platform to process more than 150 million unstructured oncology document components and extract 100 data elements across 39 cancer types.
The Boston-based oncology technology and data company said that up to 80 percent of vital health data exists in unstructured forms such as clinical notes, lab tests, medical images, sensor readings, genomic data, and operational and financial data, and 97 percent of that data is lost to analysis. It hopes to use this data to inform personalized treatment and improve patient outcomes.
Within the collaboration, Microsoft will make available to Ontada its Azure AI platform, a set of artificial intelligence-driven services and tools that allows scientists to analyze unstructured text, images, speech, and other data formats. According to the companies, Azure AI will "allow for enhanced data quality, accuracy, and the construction of a richer view of the patient journey" and bolster Ontada's ability to analyze real-world data using its own AI platform ON.Genuity and the iKnowMed electronic health record system.
"Our strategic collaboration with Microsoft enables us to extend the boundaries of artificial intelligence-driven oncology research by generating meaningful and actionable data at unprecedented scale and speed," Ontada President Christine Davis said in a statement.
In 2023, the US Food and Drug Administration contracted with Ontada to explore the natural history of rare cancers using real-world data from community oncology clinics. Ontada has also previously partnered with BeiGene, Merck, and Amgen to apply real-world data insights to improve cancer care. In 2021, Ontada provided real-world data on 3,500 patients drawn from electronic health records for the MYLUNG study, in which researchers are investigating real-world barriers keeping non-small cell lung cancer patients from receiving molecular testing and precision medicine.