DSF funds project for AI-based analysis of Dravet data
The effort seeks to develop an 'ontology' derived from scientific literature
The Dravet Syndrome Foundation (DSF) is funding a collaborative research project to provide a framework for artificial intelligence (AI)-based analyses of data in the published scientific literature related to Dravet syndrome.
The grant provides $240,000 over the next two years to support the project. The work will be led by Satya Sahoo, PhD, a professor at Case Western Reserve University, Cleveland, and Jeffrey Buchhalter, MD, PhD, an adjunct professor at the University of Calgary School of Medicine, Canada.
“We hope this project will create a useful publicly available tool that propels research using cutting edge AI technology that may also eventually support patient care and health system management of this complex disease,” the DSF stated in a foundation blog post.
Dravet syndrome is a severe form of epilepsy caused by genetic mutations. Children with the condition usually start having seizures in infancy with developmental and behavioral abnormalities manifesting as disease symptoms as they begin to grow.
Research on Dravet syndrome continues to grow, covering topics from better understanding how the disease manifests and affects patients to untangling its underlying biological mechanisms and developing new therapeutic strategies.
Some studies use animal models, while others use cellular models and still others rely on patient data. Along with studies that focus on Dravet syndrome, other biological and sociological research may shed light on the disorder and its impact.
With so much Dravet-related data available and so many studies being published, it can be a challenge to ascertain what is the most meaningful, a concern the project seeks to address.
Its goal is to create a Dravet ontology, or framework of words and concepts to understand the condition. This framework can then be used by AI systems to perform automated assessments of the scientific literature. That is, the AI can use the ontology to identify terms linked to Dravet syndrome, then comb through the literature to find studies that use the same terms and potentially highlight similarities and differences between animal models and patients.
“Artificial intelligence (AI) methods such as ontologies and machine learning algorithms are ideal for complex analytics over big data, which can help in knowledge discovery and identifying new research studies that bridge the gap between model organisms and humans,” the scientists wrote in their grant summary.
Machine learning is a type of AI that uses algorithms to analyze data, learn from its analyses, and then make a prediction. This type of data could help identify potential therapeutic targets and approaches for Dravet syndrome.
The Dravet ontology will be created and tested by Sahoo and Buchhalter with the help of DSF and experts in Dravet-related basic and clinical research.
Sahoo previously led a similar ontology project for epilepsy. The new project will build on that effort.
“We propose to extend the epilepsy ontology for Dravet Syndrome that, together with machine learning algorithms, can automatically index clinical literature and enable analysis of basic science data,” the scientists wrote.
About the Author
