Vortrag: Sarah Vetsch (MSc), Andreas Müller, Dr.phil I: anlässlich IPEG

Sarah Vetsch, MSc, und Andreas Müller referieren anlässlich des internationalen Kongresses der EEG-Pharmako Gesellschaft in Zürich (25.11.2018: 9:00, Psychiatrische Universitätsklinik Zürich, Lenggstrasse 31, 8032 Zürich):

EEG/ERP – Biomarker/Neuroalgorithms in adults with ADHD: Development, reliability and application in clinical practice.

Andreas Müller, Sarah Vetsch, Ilia Pershin (Brain and Trauma foundation, Grison Switzerland)

Abstract

Background: Although countless research work on mental disorders and neurobiology has been carried out in recent years, this has had little effect on clinical processes. Electrophysiological characteristics of Attention-Deficit/Hyperactivity Disorder (ADHD) in conjunction with recent machine learning methods promise easy-to-use approaches that can complement the existing diagnostic elements if sufficiently large samples are taken as a basis. Neuro-algorithms are understood as models of multidimensional brain networks by means of which ADHD patient data can be separated from healthy controls data.

Methods: Spontaneous electroencephalographic (EEG) and event-related potential (ERP) data were collected three times over the course of a two-year period in a multicenter sample of adults consisting of 181 ADHD patients and 147 healthy controls. Spectral power as well as ERP amplitude and latency measures were used as input data for a semi-automatic machine learning framework incorporating the training, selection and evaluation of multiple classification models based on nested cross-validation.

Results: ADHD patients and healthy controls could be classified with a sensitivity ranging from 75% to 83% and specificity values of 71% to 77%. In the analysis of the repeated measurements, sensitivity values of a selected logistic regression model remained high (72% and 76%), while specificity values slightly decreased over time (64% and 67%).

Conclusions: The implementation in clinical practice requires the facility to track the affected networks as well as expertise in neuropathophysiology. In light of this, the use of neuroalgorithms can enhance the diagnostic process by making it less subjective, more reliable and linked to the underlying pathology.

(1-3):

  1. Kropotov JD. Functional neuromarkers for psychiatry: Applications for diagnosis and treatment: Academic Press; 2016.
  2. Mueller A, Candrian G, Grane VA, Kropotov JD, Ponomarev VA, Baschera GM. Discriminating between ADHD adults and controls using independent ERP components and a support vector machine: a validation study. Nonlinear Biomed Phys. 2011;5:5. doi: 10.1186/1753-4631-5-5. PubMed PMID: 21771289; PubMed Central PMCID: PMCPMC3149569.
  3. Müller A, Candrian G, Kropotov J. ADHS-Neurodiagnostik in der Praxis: Springer-Verlag; 2011.

Keyword: ADHD, Biomarker, Neuroalgorithms, Machine Learning

Schreiben Sie einen Kommentar

Ihre E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert.