This study investigates a model that uses genetic, clinical, and laboratory data to predict the risk of recurrent ischemic strokes in patients with laboratory aspirin resistance. By analyzing data from 296 ischemic stroke patients, the researchers identified several genetic markers linked to aspirin resistance and stroke recurrence, including the ITGB3, GPIba, and TBXA2R genes. The study uses machine learning algorithms, such as CatBoost and Random Forest, to enhance prediction accuracy.
The machine learning model was developed using various data points like patient age, body mass index, and platelet aggregation levels, along with genetic information. With a high Area Under Curve (AUC) score of 0.89, this model demonstrates impressive predictive capabilities, allowing for personalized risk assessments in patients.
By identifying genetic and clinical factors that increase the likelihood of recurrent strokes, this model offers a tool for targeted prevention, enabling healthcare providers to customize treatment plans for high-risk patients. This approach has the potential to improve outcomes for patients with aspirin resistance by allowing for more tailored and effective stroke prevention strategies.
Comments
Post a Comment