Machine Learning may Revolutionize accurate diagnosis of Cardiac Tumor, unravels study

Traditional diagnostic approaches often fall short in accuracy and reliability. This study pioneers the use of ML models—Support Vector Machines (SVM), Random Forest (RF), and Gradient Boosting Machines (GBM)—to optimize diagnostic processes for cardiac tumors, especially in the context of limited datasets prevalent in specialized medical fields.

The research utilized a dataset comprising clinical features from 399 patients at the Heart Hospital. The study meticulously evaluated the performance of SVM, RF, and GBM models against traditional diagnostic metrics. The primary goal was to develop and validate ML models capable of enhancing diagnostic accuracy for cardiac tumors.

The study findings were as follows:

• The RF model emerged as the superior ML model, achieving a groundbreaking accuracy of 96.25% and a perfect ROC AUC score of 0.99. This performance significantly outperformed existing diagnostic approaches.

• Clinical validation conducted at the Heart Hospital further confirmed the models' applicability and reliability. The RF model demonstrated a diagnostic accuracy of 94% in a real-world setting.

• Key predictors identified included age, echo malignancy, and echo position, underscoring the value of integrating diverse data types for enhanced diagnostic precision.

The study highlights the potential of ML to revolutionize cardiac tumor diagnostics. The RF model's superior performance and clinical validation results underscore the feasibility of implementing ML-driven diagnostic tools in clinical settings. The integration of diverse data types, such as echocardiography and pathology data, plays a crucial role in enhancing diagnostic accuracy. This study highlights the capabilities of ML in improving diagnostic precision for cardiac tumors but also paves the way for its broader application across various medical diagnostics domains.

Reference:

Sadegh-Zadeh, S.-A., Khezerlouy-aghdam, N., Sakha, H., Toufan, M., Behravan, M., Vahedi, A., Rahimi, M., Hosseini, H., Khanjani, S., Bayat, B., Ali, S. A., Hajizadeh, R., Eshraghi, A., Ghidary, S. S., & Saadat, M. (2024). Precision diagnostics in cardiac tumours: Integrating echocardiography and pathology with advanced machine learning on limited data. Informatics in Medicine Unlocked, 101544, 101544. https://doi.org/10.1016/j.imu.2024.101544