Research on key algorithm optimization strategy of oscillographic blood pressure measurement

Abstract In this paper, adaptive filtering and wavelet de-baseline drift algorithms are proposed for the signal preprocessing of the oscillographic blood pressure measurement algorithm. On feature extraction, fuzzy logic classification and dynamic programming tracking methods are designed. The blood pressure model of support vector machine and adaptive neural network is constructed. Active learning and cross-domain modeling are introduced to correct individual differences. This study comprehensively optimizes the key algorithm of oscillographic blood pressure measurement, and significantly improves the measurement accuracy, stability, and individual adaptability of the system.