Page 314 - Invited Paper Session (IPS) - Volume 2
P. 314
IPS254 Thaddeus Tarpey et al.
Statistical modeling for psychiatric nosology
Thaddeus Tarpey , Eva Petkova , R. Todd Ogden 2
1
1
1 New York University
2 Columbia Univesity
Abstract
The problems of defining, diagnosing and treating psychiatric disorders are
difficult and are of high-interest. This paper focuses on statistical modeling
strategies for addressing this problem. In particular, the unsupervised learning
methods of finite and infinite mixtures are contrasted. In precision medicine,
the focus is often on treatment outcome. Can data on treatment outcomes aid
in the developing neurobiological categorization approaches to differentiate
psychiatric diseases?
Keywords
Aliasing; convolution; mixtures; placebo effect
1. Introduction
The term nosology refers to the branch of medical science dealing with
the classification of diseases. Nosology has been a major issue in the field of
psychiatric research. It is difficult to define mental illnesses such as depression
and bipolar disorder etc. It is also difficult to diagnosis such medical conditions
as well. Due to these difficulties, in psychiatric nosology, there is high interest
in discovering biosignatures from biological measures that can be used to
characterize and diagnose diseases (Insel et al., 2010). Currently, the problem
of defining and diagnosing mental diseases is often based on the DSM-V, or
Diagnostic and Statistics Manual (American Psychiatric Association., 2013)
which is often quite problematic. For example, Ostergaard et al. (2011) show
there are 1497 combinations of symptoms that can lead to a diagnosis of
depression.
Highlights of the problems and issues related to psychiatric nosology are
indicated by the following selected quotes from RDoC Insel et al. (2010):
“Diagnostic categories based on clinical consensus fail to align with findings
emerging from clinical neuroscience and genetics. The boundaries of these
categories have not been predictive of treatment response... One consequence
has been to slow the development of new treatments targeted to
underlying pathophysiological mechanisms... the critical test is how well the
new molecular and neurobiological parameters predict prognosis or
treatment response... to implement1 neuroscience-based psychiatric
classification.”
301 | I S I W S C 2 0 1 9