Patents
Literature
1 results about "Hypsarrhythmia" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
Hypsarrhythmia is very chaotic and disorganized brain electrical activity with no recognizable pattern, whereas a normal brain electrical activity shows clear separation between each signal and visible pattern. It is an abnormal interictal pattern, consisting of high amplitude and irregular waves and spikes in a background of chaotic and disorganized activity seen on electroencephalogram (EEG), and frequently encountered in infants diagnosed with infantile spasms, although it can be found in other conditions.
Electrocardiosignal classification method based on dual-threshold neuromorphic reservoir
PendingCN122087509ARich feature dimensionImprove the problem of poor linear separabilityBiological modelsSensorsEcg signalLinear coding
The invention discloses an electrocardiosignal classification method based on a dual-threshold neuromorphic reservoir, which comprises the following steps of: 1, receiving an electrocardiogram time sequence signal, and carrying out adjacent time sequence nonlinear coding operation on the electrocardiogram time sequence signal to generate an enhanced feature vector containing an original signal and a second-order interaction feature; step 2, inputting the enhanced feature vector into a dual-threshold neuromorphic reservoir, and generating a dual-channel pulse response sequence by using a dual-threshold pulse neuron to distinguish signal strength states; and step 3, constructing a joint feature vector according to the two-channel pulse response sequence, and inputting the joint feature vector into a classifier to obtain a classification result of the electrocardiosignals. According to the method, through cooperation of feature enhancement and a double-threshold mechanism, the problems that a traditional liquid state machine is poor in linear separability and neurons are prone to saturation are effectively solved, the number of the neurons is greatly reduced, and meanwhile the accuracy and the energy efficiency ratio of arrhythmia classification are improved.
Owner:ZHEJIANG UNIV OF SCI & TECH