Category: Translational

Symposium

Symposium 32 - Computational Clinical Science: New Techniques to Improve Methods, Theory, Diagnosis, and Prediction

Friday, November 17
12:00 PM - 1:30 PM
Location: Sapphire Ballroom M & N, Level 4, Sapphire Level

Keywords: Change Process / Mechanisms | Translational Research | Cognitive Processes
Presentation Type: Symposium

Computational clinical science refers to the application of computational modeling techniques originally developed in various basic sciences to problems in clinical science (Redish & Gordon, 2016). Computational modeling may help to advance clinical methods, theory, diagnosis, and prediction (Huys, Maia, & Frank, 2016). Four aspects of the computational clinical science movement are of special interest to the ABCT community. First, computational models may be able to model the dynamics of behavioral processes with orders-of-magnitude more precision than standard behavioral measures (Maia & Frank, 2011). Second, extant clinical knowledge can be used in combination with computational modeling to “bootstrap” (see Meehl & Cronbach, 1955, p. 286) from imprecise and underspecified concepts to precise and specific constructs (Paulus, Huys, & Maia, 2016). Third, construct development can help to refine the diagnostic system (Flagel et al., chapter 10, in Redish & Gordon, 2016). Fourth, consistent with dawning recognition in the field of machine learning that improving feature specification is essential to improving prediction (Bengio, Courville, & Vincent, 2013), precise measures of well-specified constructs may accelerate our ability to answer the key “for whom” questions in clinical science (Shoham & Insel, 2011). Examples of such questions include for whom specific treatments are most beneficial (DeRubeis, Cohen, et al., 2014) and who benefits from intensive treatment as opposed to for whom such treatment wastes scarce resources (Fournier et al., 2008, 2010; Lorenzo-Luaces, DeRubeis, et al., 2017).


Computational clinical science is in its infancy, but the presentations in this symposium point to its potential and the diversity of its potential applications. Specifically, the presentations showcase a variety of applications: methods improvements (Peter Hitchcock), construct refinement (Henry Chase), diagnostic specificity (Vanessa Brown), and outcome prediction (Katia Harlé).

Learning Objectives:

Peter F. Hitchcock

Graduate Student
Drexel University

Presentation(s):

Send Email for Peter Hitchcock

Richard J. McNally

Harvard University

Presentation(s):

Send Email for Richard McNally

Peter F. Hitchcock

Graduate Student
Drexel University

Presentation(s):

Send Email for Peter Hitchcock

Henry W. Chase

Research Assistant Professor
University of Pittsburgh

Presentation(s):

Send Email for Henry Chase

Vanessa M. Brown

Graduate Student
Virginia-Tech Carilion Research Institute

Presentation(s):

Send Email for Vanessa Brown

Katia M. Harlé

Assistant Clinical Professor
University of California, San Diego

Presentation(s):

Send Email for Katia Harlé


Assets

Symposium 32 - Computational Clinical Science: New Techniques to Improve Methods, Theory, Diagnosis, and Prediction



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Computational Clinical Science: New Techniques to Improve Methods, Theory, Diagnosis, and Prediction