Study to Develop and Validate a Model Predicting Depression After Ischemic Stroke Using Patient Data and Machine Learning

What this Trial Is About

Post-stroke depression PSD is a common neuropsychiatric condition affecting 20% to 60% of patients after a stroke. It is linked to higher death rates, slower recovery, more cognitive problems, and increased financial and caregiving burdens for families and society. Early detection of PSD is challenging, which may worsen outcomes. This research aims to use machine learning to combine various patient data to create a risk prediction model for PSD, helping healthcare providers assess PSD risk early and accurately for each patient. The study groups patients based on whether they have been diagnosed with PSD after ischemic stroke. It observes patients admitted within seven days of stroke symptom onset and collects multi-dimensional indicators to support the development and clinical validation of the PSD prediction model. No treatments or drugs are assigned, as this is an observational study focusing on diagnosis and risk assessment. Participants will be evaluated for PSD occurrence three months after their stroke. Researchers will monitor the number of PSD cases at this time point. Patients undergo neuropsychological examinations unless excluded due to severe cognitive or communication difficulties. Consent is required from patients or family members. The study will provide insights to improve early, individualized assessment and management of PSD over the follow-up period.

Construction and Clinical Validation Study of a Prediction Model for Depression After Ischemic Stroke