Follow Datanami:
May 3, 2021

Spark NLP Crosses Five Million Downloads, John Snow Labs Announces

Oliver Peckham

Enterprise AI applications are booming, and the pandemic seems to have only accelerated their growth. Today, John Snow Labs – which primarily develops AI and NLP tools for healthcare applications – is announcing that its flagship NLP library, Spark NLP, has crossed five million downloads, having doubled its growth over the last four months.

The company says that over the last year alone, paid customers for its commercial Spark NLP for Healthcare tool (along with its Spark OCR tool) have tripled, with new customers including a series of major pharmaceutical and healthcare companies. Further, the company is claiming that 54% of all healthcare-oriented AI use Spark NLP, which includes more than 1,400 pre-trained models aimed at the sector.

John Snow Labs attributes its strong growth to “a long series of updates and enhancements,” citing new software releases every two weeks for the last 40 months. In addition to general improvements to Spark NLP’s speed and performance, the company has added better named entity recognition and multi-modal learning to the tool.

“Trends including the urgent industry need for more personalized medicine and accelerating clinical trials, paired with clinicians’ continued reliance on free text and new AI breakthroughs is making NLP a foundational healthcare technology,” said David Talby, the CTO of John Snow Labs. “We are proud to help our customers accelerate the progress of healthcare by translating the most recent AI research advances into production-grade software that continues to exceed industry standards for accuracy, scalability, and privacy.”

2020 saw John Snow Labs, like many tech firms, lend its hand to the global effort (and make a case for the efficacy of its platform) by offering many of its healthcare AI products and healthcare datasets free of charge to researchers working on COVID-19-related issues. John Snow Labs also took the opportunity last year to announce its inaugural NLP Summit, a virtual (of course) two-week conference highlighting “best practices, real-world case studies, and the latest software, models, and transformers” for the NLP world. Around the summit, John Snow Labs announced a new entity recognition model for adverse drug events, the latest in a series of additions to its product line.

John Snow Labs, no doubt aided by its apparent market leadership within the sector, has maintained a positive outlook on medical applications of AI even as other AI healthcare projects fail to make the connection between promising ideas and harsh realities. Just a few months ago, Talby told Datanami that while “it’s going to take a long time” to perfect, “the technology fundamentally works” for NLP in healthcare, which he says can “match or exceed human levels of accuracy” for most tasks.

Datanami