Indicators of research to translate into clinical applications in biomedical research are hard to understand and harder to predict. The need for finding such indicators has become more pressing than ever, unarguably. Decades of interval between a scientific discovery and actual applications reiterate this need. So, a new data-driven decision-making model based on artificial intelligence and machine learning is likely to change this soon.
Researchers from Office of Portfolio Analysis (OPA) in collaboration with the U.S. Agency, the National Institutes of Health (NIH), has discovered that looking over the metric they call Approximate Potential to Translate (APT) can be the starting point. Of note, the APT values can be easy to access and can prove helpful in detecting signs of translational potential of a biomedical research. Furthermore, the first step toward getting these signs correct is getting hold of the research paper’s citation value.
Open Citation Collection brings Vast Data into Public Realm
Most of the times, APT values are hard to get through as they are not available publicly. Propriety protocols and licensing agreements hinder their availability. The team of researchers circumvented this problem. They created open citation collection (NIH-OCC), a place that pulled in citation data from various sources available publicly. Further, the scope of the collection is vast. It presently has 420 million citation links, and are likely to be updated frequently.
The NIH-OCC is aimed to quantify APT values without the need for large data sets and huge time lags. Even two years of publication data after publication of research findings would suffice, the researchers contended. Interested stakeholders can use the model through the iCite web service through a tool. Moreover, the tool among other things enable one to get information on citation(s) data by clinical articles, and the APT values.
The researchers consider the model to expand their understanding of bench-to-bedside research translation—one of the pillars of to measure impact of science to real world. Hence, focusing on research articles that are better candidate for clinical translation in biomedical science will eventually benefit human health in years to come.