Computational Based Neuropsychiatric Research
As scientists we have two fundamental responsibilities: to create knowledge and disseminate it. We create knowledge with the resources that are entrusted to us. We then disseminate this new-found knowledge by teaching, mentoring, publishing, using it in our clinics and collaborating with commercial organizations. In our lab we do all of these activities, as they relate to neuropsychiatric innovations.
Answering today's scientific questions requires a diverse team of experts. In fact, one would be hard pressed to find many biomedical problems where a solo investigator sits alone searching for a solution. The questions are just too complex.
Our collaborations include working with experts in early identification of epilepsy, neurosurgery, automated suicide classification, machine-learning exploration of spiritual needs, the language of depression, and childhood sexual abuse. In all our efforts we are part of teams that include computational and data scientists, clinicians, clinical trials experts, software developers and project managers, and so forth.
Together we have developed new ways to diagnose and treat the children we serve. Most of our work includes applying machine learning, natural language processing and spreading activation to understand neuropsychiatric illness.
For example, Dr. Pestian collaborated with Drs. Tracy A. Glauser, Richard J. Wenstrup, and Alexander A. Vinks to identify methods that determine the optimal neuropsychiatric drug. Those methods were patented and commercially licensed. By December 2016, this discovery had been used to help more than 250,000 people, and hundreds more are helped every day. In another example, Drs. Pestian and Glauser, et al, developed computational algorithms that are based on the human decision making process. We call this spreading activation and it now serves as the basis of our approach for computationally based machine learning that is used in our project.