An approach for cancer outcomes modelling using a comprehensive synthetic dataset.
Unlock the power of synthetic data! This novel approach generates a comprehensive dataset that mirrors real-world cancer patient information, enabling robust machine learning models to predict outcomes. Breakthrough findings pave the way for accelerated cancer research and per...
This study presents a novel approach for generating a comprehensive synthetic dataset that accurately mimics real-world cancer patient data, including computed tomography-based radiomic features and clinical information. Using a non-small cell lung cancer dataset, the researchers trained and tested machine learning models to predict two-year overall survival, comparing performance on real and synthetic data. The synthetic dataset closely matched the real data, and the models achieved comparable predictive accuracy, demonstrating the potential of this method for augmenting limited patient data and accelerating cancer outcomes research. While further validation with larger datasets is needed, this approach shows promise for application to other cancer types and endpoints, paving the way for more efficient and personalized cancer care.