Department of Statistics of St. Xavier’s University, Kolkata (SXUK) organized a seminar on “A Bayesian Multimodal Model for the Identification of Biomarkers in Neuroimaging Studies” on April 17, 2026 at the university campus.

The resource person was Dr. Dulal Kumar Bhaumik, Professor of Psychiatry, Biostatistics and Bioengineering at the University of Illinois, Chicago, USA. 

He introduced a Bayesian multimodal modeling framework designed to identify biomarkers, or neuromarkers, by integrating structural and functional neuroimaging data. The proposed model focused on comparative studies involving individuals with late-life depression and healthy control groups. He highlighted how identifying such biomarkers can support targeted therapeutic interventions and improve treatment outcomes, particularly in treatment-resistant cases.

The talk also explored how structural connectivity can enhance the interpretation of functional brain activity, thereby improving the precision of biomarker detection. 

Addressing the challenges posed by high-dimensional neuroimaging data, Dr. Bhaumik explained a novel statistical method for controlling the false discovery rate in small-sample settings. He further demonstrated the effectiveness of the proposed framework through simulation studies and real-data applications related to late-life depression and cognitive function.

The seminar witnessed enthusiastic participation from students, faculty members, and researchers.

Dr. Soma Debnath.