Predicting functional outcome using machine learning and large language models: A transdiagnostic approach in a young, help-seeking population (Preprint)

<sec> <title>BACKGROUND</title> Functional impairments associated with mental health conditions are on the rise. Predicting functional outcomes may improve the targeting of preventive interventions. While prognostic models have primarily focused on psychosis, early recognition services require a transdiagnostic approach. </sec> <sec> <title>OBJECTIVE</title> Predict global functioning within 2-year follow-up using baseline clinical and structural MRI data in a population-based sample of young, help-seeking individuals presenting with affective and anxiety symptoms as well as ADHD. </sec> <sec> <title>METHODS</title> We classified 357 help-seeking individuals aged 18-35 years recruited from 9 sites as “impaired” (Global Assessment of Functioning (GAF)≤60; N=228) or “non-impaired” (GAF&gt;60; N=129) at year one and/or year two follow-up. GAF classification group status at follow up was predicted using linear support vector machine (SVM), decision tree and large language model (LLM) Llama-3 using clinical assessments and/or structural MRI. Leave-one-site-out (SVM) or external sample (LLM) were used for validation. </sec> <sec> <title>RESULTS</title> SVM achieved balanced accuracy 69.2% using clinical features only. Items related to baseline occupational functioning, interpersonal relationships, cognitive functioning, psychotic and affective symptoms as well as presence of anxiety disorder were most predictive. The decision tree further reduced the feature set to 5 predictive items achieving balanced accuracy 76.6%. Although amygdala and hippocampal subregions achieved balanced accuracy 57.1%, structural MRI did not improve the overall prediction. Llama-3 performed comparably well to SVM, (balanced accuracy 72.6%). </sec> <sec> <title>CONCLUSIONS</title> Machine learning demonstrated good performance in predicting global functioning. Interestingly, the out-of-the-box LLM performed comparably well without being trained or fine-tuned, highlighting the potential of leveraging free-text data for mental health prognosis. </sec>

Authors

Publication Details

Published
2025-09-26
DOI
https://doi.org/10.2196/preprints.84424
Primary Topic
Machine Learning in Healthcare
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Predicting functional outcome using machine learning and large language models: A transdiagnostic approach in a young, help-seeking population (Preprint)

Vera Flasbeck, Christoph Mulert, Pavol Mikoláš, Karolina Leopold et al.
Machine Learning in Healthcare
article

Predicting functional outcome using machine learning and large language models: A transdiagnostic approach in a young, help-seeking population (Preprint)

Vera Flasbeck, Christoph Mulert, Pavol Mikoláš, Karolina Leopold, Andreas Jansen, Thomas Ethofer, Michael Marxen, Isabella C. Wiest, Falk Gerrik Verhees, Georg Juckel, Christina Berndt, Christoph Vogelbacher, Jonathan Repple, Michael Bauer, Philipp Ritter, Anne Rau, Jana Fiebig, Florian Thomas‐Odenthal, Julia Martini, Andreas Reif, Silke Matura, Gregor Leicht, Birgit Maicher, Christoph Ulrich Correll, Martin Lambert, Eva Mennigen, Udo Dannlowski, Tilo T. J. Kircher, Andreas J. Fallgatter, Felix Bermpohl, Andrea Pfennig, Jakob Nikolas Kather, Thomas Stamm, Irina Falkenberg, Fabian Huth, Vivien Kraft, Andreas Bechdolf, Paula Marie Henneberg, Kyra Bröckel-Bundt
article en

Abstract

<sec> <title>BACKGROUND</title> Functional impairments associated with mental health conditions are on the rise. Predicting functional outcomes may improve the targeting of preventive interventions. While prognostic models have primarily focused on psychosis, early recognition services require a transdiagnostic approach. </sec> <sec> <title>OBJECTIVE</title> Predict global functioning within 2-year follow-up using baseline clinical and structural MRI data in a population-based sample of young, help-seeking individuals presenting with affective and anxiety symptoms as well as ADHD. </sec> <sec> <title>METHODS</title> We classified 357 help-seeking individuals aged 18-35 years recruited from 9 sites as “impaired” (Global Assessment of Functioning (GAF)≤60; N=228) or “non-impaired” (GAF&gt;60; N=129) at year one and/or year two follow-up. GAF classification group status at follow up was predicted using linear support vector machine (SVM), decision tree and large language model (LLM) Llama-3 using clinical assessments and/or structural MRI. Leave-one-site-out (SVM) or external sample (LLM) were used for validation. </sec> <sec> <title>RESULTS</title> SVM achieved balanced accuracy 69.2% using clinical features only. Items related to baseline occupational functioning, interpersonal relationships, cognitive functioning, psychotic and affective symptoms as well as presence of anxiety disorder were most predictive. The decision tree further reduced the feature set to 5 predictive items achieving balanced accuracy 76.6%. Although amygdala and hippocampal subregions achieved balanced accuracy 57.1%, structural MRI did not improve the overall prediction. Llama-3 performed comparably well to SVM, (balanced accuracy 72.6%). </sec> <sec> <title>CONCLUSIONS</title> Machine learning demonstrated good performance in predicting global functioning. Interestingly, the out-of-the-box LLM performed comparably well without being trained or fine-tuned, highlighting the potential of leveraging free-text data for mental health prognosis. </sec>

Openalex Percentile: Top 86%
Machine Learning in Healthcare
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.