Bedda Rosario's fascination with numbers and patterns began at a young age. Growing up in Puerto Rico, she was drawn to mathematics and problem-solving, and she pursued these interests as an undergraduate at the University of Puerto Rico where she majored in mathematics. However, it wasn't until she took a probability and statistics course in her fourth year that she discovered her true passion for biostatistics and public health.
"I was missing something in mathematics," Rosario recalled. "But when I did biostatistics, everything clicked. I saw all the applications that could be done, and I learned about public health and other areas such as epidemiology and governmental health."
Rosario went on to pursue her master of public health at the University of Puerto Rico, and later her PhD at the University of Pittsburgh School of Public Health, finishing in 2008. Her research focused on neuroimaging and Alzheimer's disease, and she worked with a team to develop new statistical methods for analyzing neuroimaging data.
"My dissertation was on applications of statistical analysis techniques for neuroimaging data," Rosario explained. "The first goal was to apply three-dimensional thin plate splines to spatially normalize brain images. The idea behind that is you want to be able to compare, for example, the same area of the brain for every patient. So, we wanted to see if we could use three-dimensional thin plate splines to do spatial normalization."
However, the team soon realized that the three-dimensional thin plate splines was not the best technique for spatial normalization. "We found that it wasn't the best approach, so we focused on using statistical parametric mapping, which is a software that has several approaches to do spatial normalization," Rosario said. "But the limitation of those approaches is that you need to specify parameters for the spatial normalization. So, we wanted to assess— if we changed those parameters, what would happen?"
The team's research led to the publication of several papers, and Rosario’s dissertation was a significant contribution to the field of neuroimaging. "We were also able to find an alternative to do partial least squares analysis for voxel-based data, which is high-dimensional data," she said. "It was a great application not only for biostatistics, but for the neuroimaging field."
Rosario’s time at Pitt was marked by collaboration and innovation, from her advisor, Lisa Weissfeld, to researchers throughout the department. "Lisa was an excellent academic advisor and dissertation advisor," Rosario said. "She provided a lot of guidance and support during my PhD program."
Rosario also praised the strong background in statistics that she received during her time at Pitt. "The courses provided me with a very strong background in statistics, and that was very helpful for my career," she said.
After completing her PhD, Rosario worked as an assistant professor at the University of Pittsburgh and then went into industry, working at IBM Watson Health. While she enjoyed the work, she found that the fast-paced environment and focus on sales was not always aligned with her values. "As biostatisticians, we think, ‘this can be improved," she said. "Why not do it now instead of later?" Currently, she is an associate professor of epidemiology at Pitt Public Health.
Her passion for biostatistics and her dedication to her work have led to many accomplishments throughout her career. In addition to her dissertation work, she has also worked on multiple grants and collaborated on diverse studies. She has been an active member of the biostatistics community, presenting her research at conferences and publishing her work in reputable journals.
As a lead member of the University of Pittsburgh Biostatistics Alumni network, Rosario has showcased her exceptional organizational skills and commitment to building connections within the statistical alumni community. She has successfully organized alumni meetings at conferences in collaboration with the department, bringing together esteemed researchers and professionals from diverse backgrounds. These events have led to meaningful discussions, collaborations, and contributed to the growth of a robust network of distinguished researchers. "We get a great education, but we always have to continue learning and learning new statistical methods and their applications," she said. "Go to conferences, keep reading papers, and network. You learn a lot from your colleagues."
Rosario’s story is a testament to the power of perseverance and a passion for learning. From her early interest in mathematics to her groundbreaking work in biostatistics, she has consistently pushed herself to learn and grow. Her contributions to the field have been significant, and she continues to inspire and mentor others in the biostatistics community.
As she reflects on her journey, she is grateful for the opportunities she has had and the people who have supported her along the way. "I've had great mentors and great classmates," she said. "I continue my communications with them, and they have been a great source of support and inspiration."
Through her work and her dedication to the field, Bedda Rosario has made a lasting impact on biostatistics and the world of medical research. Her story is a reminder that with hard work, determination, and a love of learning, anything is possible.
-Calvin Dziewulski