Statistical signals spot seizures

In This Story

People Mentioned in This Story
Body

Ruishan Lin, a PhD student in George Mason University’s Statistics Department, won first place in the Statistical Significance Competition poster session of the 2025 American Statistical Association Joint Statistical Meeting (JSM) in Nashville, “one of the largest statistical events in the world.”   

Her research, supervised by Department of Statistics assistant professor Abolfazl Safikhani, applied statistical modeling to analyze brain patterns using electroencephalogram (EEG) data. By detecting changes in certain brain patterns, she found that health practitioners can anticipate when a patient with epilepsy may be about to experience a seizure. “It helps in reducing future risks and pre-detecting seizures that are going to happen; we can thus take precaution as quickly as possible,” she said. 

Lin look to student peers to learn how to make her poster more visually appealing. 

She worked closely with Safikhani. “He's a very established researcher in change point detection (identifying sudden shifts in statistical properties in a time series). I took his time series class and then he asked if I wanted to work on this project with him,” said Lin. The locations of EEG channels on the scalp of a patient create a natural network, making such signals ideal for applying change point detection methods.  

Lin said the supportive culture at George Mason allows PhD students to flourish, citing the department’s R. Clifton Bailey Travel Award that allowed her to attend JSM. She also took advantage of poster sessions at Innovation Week, StatConnect, and the college’s Whiskey and Widgets event to exhibit and learn from her peers. She realized her early versions were too technical and not aesthetically appealing. “I talked to other students who made some great posters and chatted with them to learn the underlying logic of how to make a good poster. Conveying my research is as important as doing my research.”  

She said that she will be working with Safikhani to apply the same research to larger public health data sets, focusing on different health issues.