Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
February 28, 2026Journal of Radiation Research and Applied SciencesOpen Access

Enhancing the early diagnosis of bipolar disorder with machine learning models

View Full Paper
Ask AI
Bookmark
Share

Authors

FAFahd S. Alharithi

Discussion

Loading...

Member takes

Overview

Multimodal analysis improves early detection of bipolar disorder, suggesting advances in precision psychiatry.

Key Points

  • The research aims to enhance early diagnosis of bipolar disorder using machine learning models that integrate various data modalities.
  • Used a multimodal dataset including fMRI neuroimaging and clinical records.
  • Developed BDiagNet-3D, a 3D Convolutional Neural Network for extracting brain activation patterns.
  • Combined features using a Deep Neural Network with clinical and cognitive data.
  • Processed data with Random Forest and Support Vector Machine for interpretable results.
  • Achieved 64% accuracy with Random Forest, 73.2% with Support Vector Machine, and 99.1% with BDiagNet-3D.
  • Precision, recall, and F1-scores followed similar performance trends.
  • Demonstrated strong potential for AI tools in clinical decision-making.

Cite This Study

Fahd S. Alharithi (2026) studied this question.

synapsesocial.com/papers/69a2877b0a974eb0d3c033bbhttps://doi.org/10.1016/j.jrras.2026.102245
View Full Paper
Ask AI
Bookmark
Share