Quantifying the Efficacy of Magnetic Nanoparticles for MRI and Hyperthermia Applications via Machine Learning Methods

Magnetic nanoparticles are a prospective class of materials for use in biomedicine as agents for magnetic resonance imagining (MRI) and hyperthermia treatment. However, synthesis of nanoparticles of high efficacy is resource-intensive experimental work. In turn, the use of machine learning (ML) methods is becoming useful in materials design and serves as a great approach to designing nanomagnets for biomedicine. In this work, for the first time we develop an ML-based approach for prediction of main parameters of material efficacy, i.e., specific absorption rate (SAR) for hyperthermia and r₁/r₂ relaxivities in MRI, with parameters of nanoparticles as well as experimental conditions as descriptors. For that, we assemble a unique database with more than 980 magnetic nanoparticles collected from scientific articles. Using this data, we train several tree-based ensemble models to predict SAR, r1 and r2 relaxivity. After hyperparameter optimization, models reach performance of R² = 0.86, R² = 0.78 and R² = 0.75, respectively. Testing the models on samples unseen during the training shows no performance drops. Finally, we develop DiMag, an open-access resource created to guide synthesis of novel nanosized magnets for MRI and hyperthermia treatment with machine learning and boost development of new biomedical agents.

In this repository, you can find the collected databases, the machine learning models described in the original work, as well as the validation data and processing steps. The structure of the repository is as follows:

• database: CSV files of manually collected and processed databases for training and testing of ML models.
• model_selection: python files for ML models involved in model selection.
• models: python files with ML models of the best performance.
• validation: python files for the best ML models and CSV data files used in the validation process.