We developed two methods to screen for white-coat hypertension and masked hypertension. First method utilizes deep learning and dimensionality reduction with Tab-PFN and t-SNE. Second method uses machine learning classifiers such as Random Forest. Features were ranked with SHAP values, and supervised feature selection was performed.

Our team is the first to publish related technology, utilizing machine learning and deep learning. We developed original screening methods, combining blood pressure measurements and lab data to maximize clinical use. We applied multi-dimensional data analysis to address current diagnostic limitations. Compared to international benchmarks, our technology excels in various performance metrics. External validation also showed excellent results, proving its applicability in various environments.

White-coat hypertension and masked hypertension have high prevalence and significant market potential. Our technology holds two Taiwanese patents and is technically unique. It can be used for preliminary screening and is suitable for portable and home health monitoring devices, such as blood pressure monitors, smartphones and wearables. It can also integrate with cloud services or connect to healthcare HIS systems.