Vega Partners with Lygase to Advance Imaging with AI

Ontario, Calif., November 27, 2024 –Vega Imaging Informatics, under core division Vega Tech, a leader in imaging AI technology solutions, and Lygase, a burgeoning and transformative AI software company, proudly announce a partnership to transform the imaging and healthcare landscapes through AI. Together, both companies will work to revolutionize these spaces by improving the processes of data curation and AI model development. This partnership marks a significant step forward in the world-wide effort to supply data curators and AI developers with the solutions they need to complete real world imaging data projects and develop AI models at increased efficiency, scale, and speed.

Lygase has developed state-of-the-art AI technology for a large-scale medical imaging Foundation Model and a world-class imaging data de-identification and validation platform, known as Medimask. Lygase’s Foundation Model provides the essential data, model and services necessary for the development of AI models, adapts to various medical imaging modalities, fine-tunes to specific clinical tasks or imaging environments, and supports diverse applications such as segmentation and classification within a single unified architecture. Moreover, Medimask leverages cutting-edge Natural Language Processing (NLP) and transformer-based technology to anonymize and validate DICOM files, offering seamless workflow integration and unparalleled performance to imaging data curators world-wide.

“Through our partnership with Lygase, Vega’s de-identification and validation platform has become best-in-class, not only positioning us to deidentify imaging data at higher volumes and with greater accuracy than any other company in the world, but also to equip other organizations with this same technology through resale,” said Luke Bideaux, CEO and President of Vega Imaging Informatics. “Additionally, adding Lygase’s state-of-the-art medical imaging Foundation Model to our existing access to millions of imaging studies for fine tuning, positions us as an industry leader for providing imaging data for AI developers.”

Medimask will be utilized internally through Vega Data and made available for resale through Vega Tech, allowing access to industry-leading tools at competitive price points. This best-of-breed medical image de-identification and validation tool will help healthcare and AI organizations create high-quality, de-identified datasets with extreme speed and without compromise of data privacy. Alongside these tools, Vega will offer managed services from Certified Imaging Informatics Professionals (CIIPs) to support its clients through adoption of this technology.

Lygase’s Foundation Model will also be available for resale through Vega Tech. Pre-trained utilizing large scale, multi-institutional datasets for robust generalization across diseases, this Foundation Model can accelerate development for a variety of services such as pathology detection, lesion tracking, physiologic function analysis, anatomical segmentation and labeling, imaging data normalization, image quality control, generative reporting, noise reduction, digital pathology image analysis, image acquisition assistance, automatic measurement, advanced visualization, synthetic image data generation, disease risk prediction, and many more.

Lygase will leverage Vega’s world-wide reach in the medical imaging AI community, including its connection to over 400 imaging facilities, along with its team of certified professionals to help with the implementation and support of Medimask and the development of its Foundation Model, ensuring seamless integration to its future customers.

“Through our partnership with Vega Imaging Informatics, Lygase has broader access to imaging data than our competitors, and their first-hand experience in de-identifying imaging data provides extremely valuable insights to our development team,” said Bilal Mohammed, CEO of Lygase. “Our Foundation Model, combined with Vega’s access to millions of real-world images, gives us a leg up over our competitors when providing a robust model and the data necessary to fine-tune it to virtually any medical imaging task. We are thrilled to be able to help Vega and their clients overcome some of the toughest challenges in the advancement of AI for medical imaging.”

About Vega Imaging Informatics

Founded in 2018, Vega Imaging Informatics’ mission is to create value for its clients through the delivery of quality imaging informatics services and solutions, to serve as an industry leader in the field of imaging informatics and to be a driver for change in healthcare that leads to improved patient outcomes. Vega’s team of CIIPs offers cutting-edge imaging informatics technology solutions alongside professional services to help medical imaging organizations overcome the most formidable IT and imaging informatics challenges. Through a diverse array of services and solutions, Vega empowers these organizations to optimize costs, address problems, and enhance operational efficiency and overall quality of care.

About Lygase

Founded in 2023, Lygase’s mission is to transform healthcare through AI. Their image de-identification platform, Medimask, leverages Natural Language Processing and transformer-based technology to anonymize and validate DICOM data files, and their large-scale medical imaging Foundation Model is transforming the way medical imaging AI technology is being developed.

Cautionary Note Regarding Forward-Looking Statements

This press release contains forward-looking statements, which are based on our current expectations and predictions about future performance, plans or prospects, along with information currently available to us, such as those focused on business prospects and opportunities. Forward-looking statements are not historical in nature, and often include use of such words as “will,” “should,” “could,” “anticipates,” “expects,” “believes,” “projected,” “potential,” or “hopes,” along with other similar terminology. Forward-looking statements are just predictions and are not guaranteed to be accurate. Furthermore, the forward-looking statements discussed in this press release are only accurate as of the date they were made, may not occur and/or actual events and results may differ materially and are subject to risks, uncertainties and assumptions about us. We are not required to publicly update these statements.