People on This Project
- Principal Investigator: Zahid Butt
- Lead: Kiran Saqib
Machine Learning & Mental Health
Machine learning-based predictive models are gaining popularity for combining huge amount of data into single model and evaluating the model’s predictive value for previously unseen individuals e.g., at risk and new patients. We utilized a scoping review methodology to rapidly map the research activity in the field of ML for predicting PPD. All studies used supervised learning techniques to predict PPD. Support vector machine and random forests were the most commonly employed algorithms. ML models were found effective in predicting PPD using electronic health records and data from social media platforms. As ML algorithms continue to be refined and improved, it may assist clinicians to identify maternal mental illnesses at an earlier stage when interventions may be more effective.