Key result
A new Lagrangian Support Vector Machine algorithm effectively classified aortic valve dynamics as normally operating or permanently closed using a mathematical cardiovascular-LVAD model.
A novel LSVM-based algorithm can accurately detect aortic valve opening and closing in a simulated LVAD system, which may help optimize bridge-to-recovery device control.
Answers are grounded in this paper and cite their sources.
May guide LVAD control algorithm development in simulation; leaves open patient validation and outcome impact.
Aortic valve dynamics - which implies continuous opening and closing of the aortic valve in each cardiac cycle during the feedback control of the rotary Left Ventricular Assist Devices (LVAD) support - has important clinical implications for patients with mild congestive heart failure. When the LVAD is implanted in such patients as a bridge-to-recovery device, permanent closure of the aortic valve must be avoided by maintaining proper control on the power delivered to the device. In this paper, a new aortic valve dynamics detection algorithm based on a Lagrangian Support Vector Machine (LSVM) model is presented. A detection indicator is derived from the systemic vascular flow signal in the circulatory system using a nonlinear mathematical model of the combined cardiovascular-LVAD system and forms the input to the LSVM classifier. The LSVM classifier is trained and tested to classify the aortic valve dynamics into two states: aortic valve opening and closing (i.e. operating normally) and aortic valve permanently closed. Our results show that the proposed algorithm can detect the aortic valve dynamics effectively in terms of classification accuracy and stability. This classifier will be an integral part in the development of a feedback controller for the LVAD when used on patients as a bridge-to-recovery device. The output of the classifier will be used to adjust the power delivered to the LVAD to ensure that the aortic valve opens and closes normally within each cardiac cycle while at the same time making sure that the physiological demands of the patient are met.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2014) studied mild congestive heart failure. Lagrangian Support Vector Machine (LSVM) model for aortic valve dynamics detection was evaluated on classification accuracy and stability of aortic valve dynamics. A new Lagrangian Support Vector Machine algorithm effectively classified aortic valve dynamics as normally operating or permanently closed using a mathematical cardiovascular-LVAD model.
Synapse has enriched 10 related papers. Consider them for comparative context: