Machine Learning Model Predicts Schizophrenia More Accurately Than Bipolar Disorder: JAMA
A new study has found that machine learning algorithms can forecast the progression of schizophrenia and bipolar disorder from typical clinical information in electronic health records (EHRs). Schizophrenia and bipolar disorder are typically diagnosed years later when symptoms first occur, although they tend to present during late adolescence or early adulthood. Inconvenient delay in diagnosis can disrupt timely treatment as well as complicate outcomes among patients. This study was conducted by Lasse H. and colleagues published in the journal of JAMA Psychiatry.
The study utilized EHR data from the Central Denmark Region's Psychiatric Services. It involved 24,449 patients aged 15 to 60 with at least two psychiatric service contacts, at least three months between them, between January 1, 2013, and November 21, 2016. The analysis was conducted from December 2022 to November 2024. Predictors were medications, diagnoses, and clinical notes extracted from the EHRs.
To forecast the diagnostic conversion to schizophrenia or bipolar disorder at five years, researchers used two machine learning models: elastic net regularized logistic regression and extreme gradient boosting (XGBoost). The performance of the model was assessed with the area under receiver operating characteristic curve (AUROC).
Key Findings
Conversion to schizophrenia or bipolar disorder was predicted by XGBoost model with an AUROC of 0.70 (95% CI, 0.70-0.70) for the training dataset and 0.64 (95% CI, 0.63-0.65) for the test dataset.
At a predicted positive rate of 4%, the model had:
• Sensitivity: 9.3%
• Specificity: 96.3%
• Positive Predictive Value (PPV): 13.0%
When specifically predicting schizophrenia, the model showed an increased AUROC of 0.80 (95% CI, 0.79-0.81), with:
• Sensitivity: 19.4%
• Specificity: 96.3%
• PPV: 10.8%
• For bipolar disorder, the model was less accurate, with an AUROC of 0.62 (95% CI, 0.61-0.63), sensitivity of 9.9%, specificity of 96.2%, and PPV of 8.4%.
The study authors concluded that machine learning algorithms, specifically XGBoost, were able to predict schizophrenia and bipolar disorder based on EHR information. With increased refinement and testing, AI models could be great assets in psychiatric facilities to help enhance early treatment and diagnosis and ultimately improve patient outcomes.
Reference:
Hansen L, Bernstorff M, Enevoldsen K, et al. Predicting Diagnostic Progression to Schizophrenia or Bipolar Disorder via Machine Learning. JAMA Psychiatry. Published online February 19, 2025. doi:10.1001/jamapsychiatry.2024.4702
