A coronary perfusion pressure prediction system and cardiopulmonary resuscitation system

By combining the LSTM-Seq2Seq model of PETCO2 and EEG signals, the problem of accurate monitoring of coronary perfusion pressure during cardiopulmonary resuscitation was solved, achieving high-quality CPR, reducing the risk of brain damage, and improving patient survival rate.

CN116763290BActive Publication Date: 2025-10-14SHANDONG UNIV +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310720720.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-16
Publication Date
2025-10-14
Estimated Expiration
2043-06-16

AI Technical Summary

Technical Problem

Existing technologies make it difficult to accurately monitor coronary perfusion pressure during cardiopulmonary resuscitation, resulting in low success rate and high risk of ischemic injury. Traditional signal prediction methods cannot take into account the randomness and uncertainty of the signal.

Method used

PETCO2 and EEG signals are combined with the LSTM-Seq2Seq neural network model to predict coronary perfusion pressure in real time. The encoder and decoder process variable-length sequences to enhance the adaptability of the prediction framework and adjust rescue measures in a timely manner.

Benefits of technology

It improves the accuracy of coronary perfusion pressure prediction and CPR quality, reduces brain damage caused by ischemia, and improves patients' survival rate and neurological prognosis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116763290B_ABST
    Figure CN116763290B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of coronary perfusion pressure prediction, and discloses a coronary perfusion pressure prediction system and a cardiopulmonary resuscitation system, which comprises a data acquisition module, which is used for acquiring the end-tidal carbon dioxide partial pressure signal and the electroencephalogram signal of a patient at a plurality of continuous time points; a coronary perfusion pressure prediction module, which is used for predicting the coronary perfusion pressure value of the patient by using a neural network based on the end-tidal carbon dioxide partial pressure signal and the electroencephalogram signal; wherein the neural network encodes the variable-length sequence composed of the end-tidal carbon dioxide partial pressure signal and the electroencephalogram signal at the plurality of continuous time points by using an encoder composed of a plurality of LSTM neural networks, and maps the variable-length sequence composed of the coronary perfusion pressure value at the plurality of continuous time points by using a decoder. Not only can brain injury caused by ischemia be intervened in time, but also the sequence-to-sequence prediction model supports variable-length input and output, and the adaptability of the whole prediction framework is enhanced.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Physiological parameter processing device and method

    CN103815966A

  • Intelligent decision-making system and method for cardio-pulmonary resuscitation

    CN115670902A