Working with an international group of human/computer interaction researchers, Brahnam has published widely and performed several studies that investigate the extent of agent abuse. In addition, she has examined agent abuse as an ethical issue and has proposed methods for designing agents so they are less likely to elicit these negative responses from users.
Brahnam uses her technology expertise in multiple other research projects — one of the primary areas being medical decision support systems.
An example of her contribution in this field is the development of the pilot Infant COPE (Classification of Pain Expressions) database and evaluation protocols, which includes 204 photographs collected of 26 healthy term infants.
The infants were photographed at Mercy Hospital in the neonatal unit while they were experiencing a number of benign nonpain stressors and an acute pain stimulus (the heel lance needed for the state-mandated blood exam).
This study was the first to investigate automatic neonatal pain detection using facial expressions.