A Model for Estimation of the Pulmonary Capillary Pressure

Abstract

Abstract:To estimate the pulmonary capillary pressure, a theory was introduced by Holloway and coworkers. Based upon this, a mathematical model describing the measured data was developed. Because the physiologic data are embedded in noise and the pulmonary capillary pressure cannot be measured directly, we simulated an extensive series of data. The noise properties of the data were as analyzed to design a signal-processing tool, that cancels the noise from the measured data. The signal processing tool developed for the current application consists of pre-processing with a moving time average filter and post-processing with a neural network. After a verification procedure the tool can be applied to measured data, hence a more reliable measurement of the pulmonary capillary pressure is achieved.

Model-Based Biosignal Interpretation II

“…Jensen et al [5] describe the principles behind, and the design of a signalprocessing' tool based on a theoretical Model for estimation of the pulmonary capillary pressure. The design involves pre-processing with a moving timeaverage filter for damping the amplitude of the noise, and post-processing by a layered neural network to compensate for the non-ideal shape of the lowpass filter.…”

“…Among the four papers in this Section, two have a cognitive outcome [2,3] and one is decision oriented [4]. The last paper from Jensen, Demant and Sanchez [5] is very different in its objectives: it describes a new noise-cancelling biosignal processing tool which has been designed assuming that both the model of the signal and of the additive noise are known.…”

Performance of Multi-Layer Feedforward Neural Networks to Predict Liver Transplantation Outcome

“…Neural networks are an important class of pattern classifiers, with growing acceptance in medical and biological research. In recent years, neural networks have been successfully applied in various medical and biological fields, such as diagnosis [1][2][3], outcome prediction [4,5], predicting length of ICU stay [6], medical imaging and signal processing [7][8][9][10], and biochemical analysis [11][12][13], to name just a few.…”