Transforming Clinical Narratives into Computational Logic: Machine-Actionable Representation of Diagnostic Knowledge and Its Applications

Activity: Talk or presentationOral presentationscience to science / art to art

Description

Accurate diagnosis is crucial for the timely and effective treatment of
disorders. The complexity of disorder definitions and the need to consider all possible manifestations during diagnosis add to this challenge. For instance, the first diagnostic criterion of Major Depressive Disorder (MDD) alone can result in over 7,000,000 possible manifestations. This complexity is
further compounded by the overlap between multiple disorder groups, making discrimination between disorders essential. Over years, established disorders have seen iterative refinement of diagnostic criteria, improving differentiation. However, there is no systematic representation explicitly stating successful discrimination between these disorders. The emergence of new clinical conditions, such as Long Covid, which presents with symptoms similar to depressive disorders and chronic fatigue syndrome, underscores the need for quick and reliable diagnostic pathways. In such cases, where iterative refinement is not yet available, it is particularly important
to systematically compare the similarities and verlaps with established disorders.
https://www.plus.ac.at/psychologie/fachbereich/sgp/vortragsreihe/
A systematic, machine-actionable representation that can quantify the similarities and overlaps
of established disorders is needed. This would facilitate accurate and reliable diagnosis, especially in the context of emerging disorders.
Currently, diagnostic criteria are predominantly available in narrative form, which limits their
utility for AI applications.
Here we present a systematic representation of diagnostic criteria using logical formalisms and
to quantify the similarities and overlaps of disorders. By transforming narrative diagnostic criteria into a structured, logical format, we can leverage computational methods to enhance diagnostic accuracy. This approach not only aids in the discrimination of established disorders but
also provides a robust framework for the integration of new clinical conditions into existing diagnostic paradigms. Ultimately, this systematic representation will support healthcare professionals in making more accurate diagnoses and improve patient outcomes through timely and
appropriate treatment interventions.
Period12 Nov 2024
Held atFachbereich Psychologie & Salzburger Gesellschaft für Psychologie, Austria

Fields of Science and Technology Classification 2012

  • 501 Psychology