NuMedii is advancing the field of personalized medicine for idiopathic pulmonary fibrosis (IPF) by creating a new advisory board that includes some of the world’s leading specialists in interstitial lung disease to guide a new approach to drug discovery based on artificial intelligence (AI).
The California-based company has been exploring the potential of big data and AI since 2010 to accelerate the discovery of precision therapies designed to address high unmet medical needs.
One of the company’s new areas of technology is called Artificial Intelligence for Drug Discovery (AIDD), which was first developed by researchers at Stanford University and which NuMedii adapted to help develop new treatments for IPF.
The technology employs a keen knowledge of human biology from literature and molecular, pharmacological, and clinical data that researchers at the company curated themselves. These data are combined with proprietary machine-learning and network-based algorithms designed to discover and advance new treatment candidates and biomarkers in several therapeutic areas, including rare diseases like IPF.
“The formation of our IPF Advisory Board and our achievement in building our proprietary knowledge base for IPF ahead of schedule reflects our steadfast commitment to this rare disease,” Gini Deshpande, PhD, chief executive officer of NuMedii, said in a press release.
“We thank all our IPF Advisors for their dedication, as their unparalleled insights and experiences will be invaluable to NuMedii as we continue to further advance our work in IPF, a horrible lung disease for which there remains significant unmet medical need,” Deshpande added.
A complete list of NuMedii’s IPF Advisory Board members is available here.
NuMedii has also created a development partnership with Three Lakes Partners, an organization that boosts collaboration between fields and stakeholders, to discover new therapies for IPF based on AIDD technology.
A 2017 research article by researchers at Stanford University, titled “Low Data Drug Discovery with One-Shot Learning,” discussed the latest advances in machine learning, their significant contribution to drug discovery, and the path that led to AIDD development.
The study details the development of a new kind of deep learning called one-shot learning that “can be used to significantly lower the amounts of data required to make meaningful predictions in drug discovery applications,” the researchers wrote.
This is an important advance in the field, as a longstanding challenge to AI in drug development is that AI algorithms traditionally need thousands to trillions of data points.