May 05, 2020 – A team from Northwestern University has developed an artificial intelligence platform that can quickly identify research that has the most potential to produce COVID-19 treatments and solutions.
For more coronavirus updates, visit our resource page, updated twice daily by Xtelligent Healthcare Media.
As the pandemic continues, scientists are conducting massive amounts of research related to coronavirus. With federal agencies like the FDA and HHS planning to accelerate clinical trials, hundreds of researchers are testing possible new treatments and vaccines, adding to the already-immense body of COVID-19 data.
To determine which research has the most potential to yield real solutions, the scientific community has historically used the Defense Advanced Research Projects Agency’s Systematizing Confidence in Open Research and Evidence (DARPA SCORE) program. The program relies on scientific experts to review and rate submitted research studies based on how likely they are to be replicable.
On average, this process can take about 314 days, which is a long time to wait during a global pandemic.
With the AI-powered tool, researchers can prioritize resources for the most promising studies, and ignore studies that are unlikely to yield valuable solutions. The model is just as accurate as the human scoring system at making these predictions, and it can scale up to review a larger number of papers in minutes instead of months.
“The standard process is too expensive, both financially and in terms of opportunity costs,” said Brian Uzzi, Richard L. Thomas Professor of Leadership at Kellogg and co-director of the Northwestern Institute on Complex Systems.
“First, it takes take too long to move on to the second phase of testing and second, when experts are spending their time reviewing other people’s work, it means they are not in the lab conducting their own research.”
The artificial intelligence tool bypasses the human-scoring method, allowing the research community and policymakers to make faster decisions about prioritizing time and funds on studies that are likely to succeed.
To build the platform, the Northwestern team developed an algorithm that would predict which studies’ results are likely to be replicable – a critical sign that a study’s results are valid. The model’s prediction of the likelihood of replicability may be more accurate than the traditional human-scoring prediction.
Because it considers more of the narrative of the study – and expert reviewers tend to focus on the strength of relational statistics – the model may have a better idea of which studies could generate viable solutions.
“There is a lot of valuable information in how study authors explain their results,” Uzzi said. “The words they use reveal their own confidence in their findings, but it is hard for the average human to detect that.”
The algorithm analyzes the words of thousands of papers, recognizing word-choice patterns that humans might not recognize. Additionally, the tool has a much bigger schema to draw upon for its predictions, making it an ideal supplement to human expertise. Researchers can use the model to analyze COVID-19 -related research papers and quickly determine which show the most promise.
“This tool is particularly useful in this crisis situation where we can’t act fast enough,” Uzzi said. “It can give us an accurate estimate of what’s going to work and not work very quickly. We’re behind the ball, and this can help us catch up.”
Used alone, the tool has an accuracy comparable to the DARPA SCORE method. Together, human experts and the algorithm predict which findings will be replicable with even greater accuracy than either method on its own.
“This tool will help us conduct the business of science with greater accuracy and efficiency,” Uzzi said. “Now more than ever, it’s essential for the research community to operate lean, focusing only on those studies which hold real promise.”
With this AI-powered tool, researchers can come closer to finding a potential treatment for COVID-19.
“In the midst of a public health crisis, it is essential that we focus our efforts on the most promising research,” said Uzzi. “This is important not only to save lives, but also to quickly tamp down the misinformation that results from poorly conducted research.”