Research On The Capacitive Coupling Detection Of Cerebral Blood Flow Pulse Signal After Subarachnoid Hemorrhage

Stroke is one of the major causes of disability and death worldwide.Subarachnoid hemorrhage(SAH)is a type of hemorrhagic stroke that is often accompanied by early brain injury from 3 to 14 days after onset.Clinical evidence suggests that changes in cerebral blood flow(CBF)are closely related to early brain injury.Therefore,monitoring cerebral blood flow in patients with subarachnoid hemorrhage can help improve the prognosis of patients.Clinical imaging methods commonly used include detecting cerebral blood flow using CT and MRI,but these devices are bulky and cannot perform continuous bedside monitoring.Transcranial Doppler(TCD)can measure cerebral blood flow,but it can only monitor blood flow of large vessels.Near infrared spectroscopy(NIRS)technology enables continuous and non-invasive monitoring of CBF,but the detection depth is limited.The existing technologies for detecting CBF have contradictions between invasive and non-invasive,continuous and intermittent,and local and global.It is necessary to study new non-invasive,continuous,and global CBF detection technologies.Capacitive coupling coupling technology originates from biological tissue electromagnetic detection technology,which has the advantages of non-invasive,non-contact,and real-time continuous monitoring.By monitoring the changes in the overall relative dielectric constant of brain tissue,the pulsation information of CBF is obtained.Therefore,capacitive coupling technology provides new ideas for the development of non-invasive,continuous,and global detection of CBF..To investigate the feasibility of capacitive coupling detection of cerebral blood flow pulsation signals in patients with SAH.The following work has been carried out in this article:(1)Construction of a capacitive coupling system for detecting cerebral blood flow pulsation.Four types of capacitive coupled sensors with different geometric structures and different capacitive coupling modes(single ended and differential)were developed,and a simple physical model of cerebral blood flow circulation was constructed.A physical experimental platform for simulating cerebral blood flow pulsation was established.Physical experiments were conducted to collect capacitive coupling response signals at different frequencies and locations.The analysis of the time and frequency domain characteristics of the signals showed that capacitive coupling can detect pulsation responses at different frequencies,with an effective detection depth that can cover the cerebral hemisphere,and a 2mm coplanar sensor has the best performance.(2)Study on the detection of cerebral blood flow pulsation in healthy volunteers by capacitive coupling.A healthy volunteer experiment was conducted to collect capacitance coupling signals from different detection sites of volunteers under normal conditions.By consuming caffeine to alter the pulsatile state of cerebral blood flow in healthy volunteers,capacitance coupled signals were collected under different states.Based on the pulse rate and cerebral oxygen value monitored synchronously by a digital oxygen pulse meter and a noninvasive cerebral blood oxygen monitor,the time domain,frequency domain,and amplitude characteristics of the capacitive coupling signal were analyzed.The results showed that the capacitive coupling signal exhibited the characteristics of cerebral blood flow pulsation signals,with the frequency domain energy component consistent with the synchronous pulse rate,and the amplitude of the capacitive coupling signal decreased after caffeine ingestion.The change trend was consistent with the change trend of cerebral oxygen.(3)Study on the detection of cerebral blood flow pulsation signals in patients with subarachnoid hemorrhage by capacitive coupling.Conduct a cerebral blood flow pulsation detection test for SAH patients,and collect capacitive coupling signals from the affected and healthy sides of SAH patients before and after surgical treatment.By comparing and analyzing the time-domain changes and amplitude trends of capacitive coupling signals in SAH patients based on patient course information,the results showed that the signal amplitude on the affected side before surgery was lower than that on the healthy side,while the signal amplitude on the affected side after surgery was higher than that before surgery.(4)Classification of cerebral blood flow pulsation capacitive coupling signals based on machine learning.Based on capacitive coupling signals from SAH patients at different locations before and after surgery,screening features were extracted,feature datasets were constructed,and training was conducted using three commonly used machine learning classification models.The evaluation was completed using six indicators,including average precision,confusion matrix,F1 score,sensitivity,ROC curve,and P-R curve.The experimental results show that the evaluation indicators of random forest needles are higher than those of SVM and decision trees,with an average accuracy of93.4%.This study verifies the feasibility of capacitive coupling detection for detecting cerebral blood flow pulsation signals in healthy volunteers and patients with SAH;Capacitive coupling detection can distinguish cerebral blood flow pulsation in healthy volunteers and patients with subarachnoid hemorrhage under different pathophysiological conditions;The capacitive coupling signals of cerebral blood flow pulsation in patients with SAH before and after surgery can be correctly classified,and an analytical and diagnostic model that can reflect the changes in cerebral blood flow pulsation levels after SAH can be established.