Can Artificial Neural Networks Predict Psychiatric Conditions Associated with Cannabis Use?

Abstract

This data-driven computational psychiatry research proposes a novel machine learning approach to developing predictive models for the onset of firstepisode psychosis, based on artificial neural networks. The performance capabilities of the predictive models are enhanced and evaluated by a methodology consisting of novel model optimisation and testing, which integrates a phase of model tuning, a phase of model post-processing with ROC optimisation based on maximum accuracy, Youden and top-left methods, and a model evaluation with the k-fold cross-testing methodology. We further extended our framework by investigating the cannabis use attributes' predictive power, and demonstrating statistically that their presence in the dataset enhances the prediction performance of the neural network models. Finally, the model stability is explored via simulations with 1000 repetitions of the model building and evaluation experiments. The results show that our best Neural Network model's average accuracy of predicting first-episode psychosis, which is evaluated with Monte Carlo, is above 80%.

A Model for Predicting the Class of Illicit Drug Suspects and Offenders

“…The classifier yielded a 98.42% accuracy in predicting a potential drug abuser. In another study, Stamate et al (2018) held that constructing predictive models for the onset of first-episode psychosis using artificial neural networks is a novel machine learning approach. Simulations were used to explore model stability by running the model development and evaluation trials up to 1000 times.…”

Identifying psychosis spectrum disorder from experience sampling data using machine learning approaches

“…In fact, aside from the neuroimaging field (Winterburn et al, 2017), the exploration of ML in the field of clinical psychosis research is still at its early stage (Iniesta et al, 2016;Schnack, 2017), but the field is rapidly adopting these contemporary methods to handle challenges of psychosis research (Tandon and Tandon, 2018). With a pragmatic approach, the first ML applications aim to identify predictors of treatment, clinical outcomes, and diagnostic classifications (Alghamdi et al, 2016;Stamate et al, 2017;Cannon et al, 2016;Fusar-Poli et al, 2017;Jauhar et al, 2018;Koutsouleris et al, 2016;Mothi et al, 2018;Stamate et al, 2018b). These early findings demonstrate the potential of ML in handling complex data in clinical psychosis research.…”