NYC Gazette

Researchers have introduced a new algorithm to detect drug interactions at the 2024 Joint Statistical Meetings conference in Portland, Oregon. The team, led by Dr. Honggang Wang, chair of the graduate Department of Computer Science and Engineering at the Katz School, presented advancements in deep learning and federated learning applied to medical research.

“Our presentations underscored the importance of integrating advanced mathematical tools like the Choquet integral into data science, particularly in the healthcare sector,” said Dr. Wang. “Our research not only advances the field of medical data analysis but has the potential to transform how healthcare providers approach patient care, especially in the era of big data and decentralized research.”

The conference focused on "Statistics and Data Science: Informing Policy and Countering Misinformation." Dr. Wang, Matthew Fried, a Ph.D. candidate in mathematics at the Katz School, and Semyon Lomasov, a research assistant pursuing graduate studies at Stanford University, contributed to discussions on statistical learning and data science.

In one presentation titled “A New Choquet Activation Function Based Deep Neural Network for Drug Interaction Detection,” Fried and Dr. Wang introduced a method for improving neural networks' ability to detect drug interactions using a new activation function based on the Choquet integral.

Dr. Wang and Lomasov also presented their development of an algorithm based on the Choquet integral for handling complex medical data without prior statistical assumptions. Their research showed that this approach could model synergistic and antagonistic effects of different drugs more accurately than conventional techniques.

In another session, Lomasov and Dr. Wang discussed “Federated Choquet Regression with Categorical Variables for Outcome Prediction in Longitudinal Trial Data.” This addressed privacy concerns when aggregating data from multiple clinical sites by developing a federated regression algorithm based on the Choquet integral.

Their algorithm was tested on synthetic and real medical data, showing strong performance in global data analysis but varying effectiveness in decentralized scenarios depending on aggregation methods used.

“As the need for more sophisticated data analysis methods continues to grow, our work sets the stage for future research that could lead to more accurate predictions and better health outcomes,” said Dr. Wang.