AI algorithms are only as powerful as the data behind them. Once large and high-quality biological data sources are accessed, machine learning can start to unlock new information that could help to develop drugs, diagnose patients, and determine the course of treatment. This is why Nucleai has partnered with medical centers around the world, such as Jefferson, Sheba, and more, to tap into a database of pathology and clinical information. The database helps build and enrich our platform, developing AI-powered image analysis models and ultimately extracting spatial insights from pathology images. In turn, our pharmaceutical partners, such as Debiopharm and Merck Group, gain access to our advanced pre-trained models which provide novel insights that enhance their drug research and clinical trial success.
Importance of Pathology Data in Spatial Biology
Although there is great promise in deep learning, there are still substantial challenges to overcome before being implemented in a clinical environment and affecting treatment decisions. One of the greatest limitations of current deep learning algorithms is the lack of medical data in both high quantity and quality.
This problem is especially apparent in the pathology field that hasn’t fully adapted to the digital transformation yet. DNA, RNA, proteins, radiology, and patient medical records are all digitized and represented in bits. Computational analysis of this data is being utilized for drug development, biomarker discovery, and clinical trial design. Pathology tissue biopsies, on the other hand, are analog by nature and therefore underutilized in medical research today. As a result, we have robust insights on how diseases behave at a molecular level, but still lack the understanding of how they work on a spatial level. Spatial characterization of tissues will complement genomics, proteomics, and radionics data, and can produce the next generation of diagnostics, biomarkers, and drugs.
Partnerships with Jefferson, Sheba and more help us access robust pathology and clinical data
How, then, can we unlock digital spatial biology data out of the analog tissue biopsies? At Nucleai, we are working to build the biggest spatial biology atlas in the world.
Through our unique data partnerships, and utilizing the highly centralized Israeli healthcare system, we have been able to digitize and analyze tens of thousands of slides to date, where each slide size can reach several gigabytes and contain hundreds of thousands of cells. We generate hundreds of human interpretable features from each slide instead of the simple information extracted from the slides today by the human eye. More importantly, for every patient analyzed, we have access not only to the pathological slide but to the genomic data, demographic data, and clinical outcome data from the EMR. Fusing spatial biology data with phenotype data is where the magic happens – it allows us to understand how spatial biology affects disease prognosis and drug response. These slides are only the tip in the iceberg – there are millions of other glass slides that are buried in pathological labs and can be used for spatial biology research.
Merck, Debiopharm, and all our Pharma Partners unlock novel insights for drug development
While this database is being used by Nucleai to discover novel biomarkers and drug targets, it also allows us to partner with pharma companies in clinical trial design and supports their mission to develop better drugs. The data we access originates from a variety of laboratories and health systems around the world, which enables Nucleai to create a robust and heterogenous platform of image analysis models. The speed and generalizability of the models provide our partners with a catalog of novel spatial insights. Ultimately, these findings, which we combine with genomics and outcomes data, provide a necessary layer of information to the drug discovery and clinical trial processes. Incorporating spatial biology into drug development, though underutilized by Pharma companies today, can deliver immense value and improve the efficacy of treatments for patients worldwide.
