Biomedical subjects

K Kück

Publications and source records attributed to K Kück.

3 recordsLinked to original sources

PubMed1997

Theoretical analysis of non-invasive oscillometric maximum amplitude algorithm for estimating mean blood pressure.

A theoretical analysis is performed to evaluate the effect of arterial mechanical and blood pressure pulse properties on the accuracy of non-invasive oscillometric maximum amplitude algorithm (MAA) estimates of the mean blood pressure obtained using air-filled occlusive cuffs. Invasively recorded blood pressure pulses, selected for their varied shapes, are scaled to simulate a wide range of blood pulse pressures (diastolic blood pressure minus systolic blood pressure). Each scaled blood pressure pulse is transformed through an exponential model of an artery to create a series of blood volume pulses from which a simulated oscillometric waveform is created and the corresponding MAA estimate of the mean blood pressure and error (mean blood pressure minus MAA estimate) are determined. The MAA estimates are found to depend on the arterial blood pressure. The errors are found to depend on the arterial mechanical properties, blood pressure pulse shape and blood pulse pressure. These results suggest that there is no direct relationship between the mean blood pressure and MAA estimate, and that multiple variables may affect the accuracy of MAA estimates of the mean blood pressure obtained using air-filled occlusive cuffs.

Algorithms↗

PubMed1995

Intelligent monitor for an anesthesia breathing circuit.

A competent breathing circuit is mandatory to the safe and effective delivery of oxygen and anesthetic gases to the patient. Studies have shown that failures in the circuit are the most likely causes of anesthetic mishaps. Unfortunately, the complexity of the system renders traditional monitoring methods ineffective. We have developed a hierarchical artificial neural network monitor that is capable of examining ventilator signals. It was trained to identify 23 faults in the breathing circuit during ventilator controlled breathing and 21 faults during spontaneous breathing. The networks correctly identified a fault condition in 92% and 83% of cases for ventilator and spontaneous data, respectively. The correct fault type was found in 76% and 68% of cases for ventilator and spontaneous data, respectively. Results show that the network met our criteria for a holistic, specific, and vigilant monitoring system.

Anesthesia, General↗

PubMed1992

Differential features for a neural network based anesthesia alarm system.

We have developed a neural network based alarm system that identifies 19 specific faults in the anesthesia breathing circuit, such as "Inspiratory Hose Leak," or "Y-Piece Disconnection." CO2, pressure, and expired flow waveforms, along with ventilator settings, were sampled by a personal computer. Fifty-two features, such as "maximum CO2" or "minimum pressure", were extracted from each breath, converted to "differential" features, normalized, and used as the inputs of a three layered feed-forward neural network. The network was trained, using backward error propagation with momentum, to classify each breath as normal or containing one of 19 faults. To collect the neutral network training set, seven dogs were anesthetized and ventilated using controlled ventilation. Each of 19 faults were created over a range of ventilator settings and fresh gas flows. The neural network correctly identified 83.1% of 550 events presented to it during testing. These preliminary results are an encouraging example of neural network applications in the field of clinical monitoring.

Anesthesia, Closed-Circuit↗