EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Data diversity vs. model complexity in the prediction of pediatric bipolar disorder: Evidence from academic and community clinical samples

Zhuoyu Shi, Eric A Youngstrom, Yinuo Liu, Jennifer K Youngstrom, Robert L Findling

Journal of Affective Disorders · 2026-12 · 414 · 122378

DOI: 10.1016/j.jad.2026.122378 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Affective Disorders, 2026-12. Verfasst von Zhuoyu Shi und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Psychology and Neuroscience, University of North Carolina at Chapel Hill, 235 E. Cameron Avenue, Chapel Hill, 27599, NC, USA. Hauptthemen laut MEDLINE: Bipolar Disorder. Die Arbeit nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Stanley Medical Research Institute, National Institutes of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Pediatric bipolar disorder is challenging to diagnose accurately due to symptom heterogeneity. More standardized and data-driven approaches are needed to enhance diagnostic reliability. We evaluated a clinical decision tool (nomogram), statistical methods (logistic regression, LASSO), machine learning (support vector machine, random forest, k-nearest neighbors, extreme gradient boosting), and deep learning (multilayer perceptron) for pediatric bipolar disorder prediction across two datasets collected in academic (N = 550) and community (N = 511) clinical settings. We compared three modeling strategies: cross-dataset validation, cross-dataset with interaction terms, and pooled-dataset. We assessed model performance using discrimination, calibration, and predictor importance ranking. In the baseline cross-dataset approach, all models showed good internal discrimination in the academic dataset, but external discrimination in the community dataset substantially declined. Interaction-enhanced models slightly improved internal discrimination but not external performance or calibration. Recalibration substantially improved cross-dataset calibration. Models trained on the pooled sample showed strong performance on held-out samples drawn from the heterogeneous pooled cohort, with good calibration for most models. Across models and training strategies, PGBI-10M was consistently identified as the most important predictor. Predictive models for pediatric bipolar disorder showed strong internal performance but limited cross-setting generalizability due to dataset shift and miscalibration. Within the present study, increasing model complexity did not improve external performance, whereas training on pooled data improved performance on held-out samples from the heterogeneous pooled cohort. These findings suggest that training-data diversity may provide greater practical benefit than increasing model complexity for developing robust psychiatric prediction models, underscoring the importance of open and collaborative datasets.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Bipolar DisorderHumansChildPredictive Learning ModelsFemaleMachine LearningPrediction AlgorithmsClassification AlgorithmsAdolescentMaleRandom ForestReproducibility of Results

Stichwörter der Autor:innen

Bipolar disorderEvidence-based diagnosisMachine learning

Autor:innen und Institutionen

  • Department of Psychology and Neuroscience, University of North Carolina at Chapel Hill, 235 E. Cameron Avenue, Chapel Hill, 27599, NC, USA.
  • Department of Psychology and Neuroscience, University of North Carolina at Chapel Hill, 235 E. Cameron Avenue, Chapel Hill, 27599, NC, USA; Nationwide Children's Hospital, 700 Children's Drive, Columbus, 43205, OH, USA;…
  • Nationwide Children's Hospital, 700 Children's Drive, Columbus, 43205, OH, USA.
  • Virginia Commonwealth University, 501 N. 2nd Street, Richmond, VA, 23298-0308, USA.

Förderung

  • Stanley Medical Research Institute
  • National Institutes of Health

Zitieren

APA

Shi, Z., Youngstrom, E. A., Liu, Y., Youngstrom, J. K., & Findling, R. L. (2026). Data diversity vs. model complexity in the prediction of pediatric bipolar disorder: Evidence from academic and community clinical samples. Journal of Affective Disorders, 414, 122378. https://doi.org/10.1016/j.jad.2026.122378

Vancouver

Shi Z, Youngstrom EA, Liu Y, Youngstrom JK, Findling RL. Data diversity vs. model complexity in the prediction of pediatric bipolar disorder: Evidence from academic and community clinical samples. Journal of Affective Disorders. 2026;414:122378. doi:10.1016/j.jad.2026.122378

BibTeX
@article{Shi2026,
  author = {Shi, Zhuoyu and Youngstrom, Eric A and Liu, Yinuo and Youngstrom, Jennifer K and Findling, Robert L},
  title = {Data diversity vs. model complexity in the prediction of pediatric bipolar disorder: Evidence from academic and community clinical samples},
  journal = {Journal of Affective Disorders},
  year = {2026},
  volume = {414},
  pages = {122378},
  doi = {10.1016/j.jad.2026.122378}
}
Kostenlos starten →