Application Of Magnetic Induction Phase Shift Method Into Monitoring Of Intracranial Pressure Increase Due To Cerebral Hemorrhage

Monitoring of intracranial pressure (ICP) is extremely important in the fields ofneurosurgery and neurology. There are invasive and non-invasive monitoring techniques.The invasive techniques are accompanied with risk of complications, such as hemorrhageand infection, and some microtransducers are even encountered with the problem of zerodrift. Noninvasive ICP monitoring methods include Magnetic Resonance Imaging (MRI)&Computer Tomography (CT), Optic Nerve Sheath Diameter (ONSD), Fundoscopy andPapilledema, Tympanic Membrane Displacement (TMD), and Transcranial DopplerUltrasonography (TCD), but the noninvasive methods are still faced with three problems.(1)Problems of large individual difference, low accuracy, and low adaptability.(2) Thecontact-type measurement requires the placement of a number of electrodes into thepatient’s head, which is not only inconvenient for clinical use, but also raises the patient’sdiscomfort.(3) These techniques do not apply to the personalized group, aging group, orother special groups, who need monitoring of health condition under the natural state.Therefore, an ICP monitoring method is needed urgently, which can overcome thelimitations of contact-type noninvasive ICP monitoring, and also can realize noninvasivenon-contact real-time and field-independent monitoring.Magnetic induction phase shift (MIPS) method has been used to study the brainmoisture content, dielectric constant, brain hematoma and cardiopulmonary activities.Electrode-less measurement of changing conductivity in the human body can be realized bydetecting the effect of the inducted eddy currents. Different biological tissues, which haveunique electromagnetic properties, can be distinguished on basis of these properties. Thismechanism has been well established. Furthermore, distinct approaches for detection ofelectromagnetic changes in tissues by non-contact means are available. Part I:(1) Research of new sensors. A coil sensor was made with Archimedes helix using thedifferential method, which improves the system sensitivity. Compared withtraditional coil sensors, the new coil sensor not only offsets the incentive effect ofthe magnetic field, but also improves the signal sensitivity.(2) Composition of the MIPS system. The system consists of a signal generator(AFG5232), an excitation coil, a detection coil, a data acquisition card (NIPCI-5124), a homemade preamplifier and filter module, a phase detector module, apersonal computer, and a biological signal collecting and processing part.(3) Evaluation of system performance. System stability test results show that at theexcitation frequency of7.7MHz, the excitation voltage is5Vpp, the amplitudesignal sampling rate is100MHz, the number of FFT points is400000, themagnification size is100, and the temperature drift of MIPS within30min is only0.370.00522°. Then simulation research and salt water simulation experimentwere carried out. Salt water experiment results show that this system’s mostsensitive frequency point is7.7MHz. The largest sensitivity occurs at a horizontaldistance of5mm from the coil sensor. The time to obtain one MIPS data is about0.057s, which is much shorter compared with previous hardware-based phasedetector systems (3-7s). The above experiments indicate that this system can helpresearchers to study the relationships among cerebral hemorrhage, ICP and MIPS.Part II:(1) Theory study of MIPS and ICP. To study the theory relationships among cerebralhemorrhage, intracranial pressure and MIPS, we built a simple brain conductivitymodel based on brain electrical conductivity distribution. An exponential functionbetween MIPS and ICP was deduced:P PeKA0. This equation is mainly aimedat the first stage of ICP increase, or namely, when CSF compensation regulationhas a material effect on the overall cerebral conductivity.(2) Relationship among cerebral hemorrhage, ICP and MIPS. The experimental group(28rabbits) and the control group (10rabbits) were anaesthetized with urethane(25%,5ml/kg) via ear vein. We established the model of ICP increase with ACH bystereotactic methods. Each rabbit in the experimental group was injected with3ml of autologous blood at a speed of0.33ml/min. The changes in ICP were measuredwith a Camino MPM-1ICP monitor. MIPS signals were collected by the systemdescribed above, and were preprocessed with wavelet transform. To facilitate dataanalysis, the signals of ICP and MIPS were re-sampled at an interval of2Hz,which each resulted in two19points of discrete variables. The experimental resultsshow that MIPS dropped by about0.77065±0.353309o,0.69082±0.371077o, and0.57636±0.225052owith the first, second and third1ml of injection, respectively.MIPS fell down with the increase of blood volume and the downtrend graduallyslowed down. Our early experimental results showed that the MIPS changes werein a strict linear relationship with blood volume. In this experiment, however, thedowntrend of MIPS was slow. We believe that MIPS changes embody not only thevolume change of cerebral hemorrhage, but also the brain compensatory changes influid flow. The changes of MIPS contain a combination of brain blood andcerebrospinal fluid (CSF). Therefore, the relationship between cerebral hemorrhageand MIPS was not strictly linear. In the control group, MIPS changed from0to0.00036°±0.013806°within the540-s period. MIPS changed very weakly due tothe low electrical conductivity of cerebral blood and the small bleeding volume.The experimental results demonstrate that ICP increases faster than blood volumedid, which agrees with the literature theory. With the increase of cranial content atthe early stages of ICP increase, since the CSF compensation dominates, ICPincreases very slowly. With the increase of cranial contents, the cerebral bloodregulating function dominates, but its adjustment ability is weakened, and ICPbegins to rise. At the third stage, the regulation abilities of CSF and brain blood areboth reduced, and ICP rises increasingly faster. In the control group, ICP changesfrom12.6±2.716207to14.7±3.056868mmHg, and basically remains unchangedcompared to the large difference in the experimental group.(3) The sensitivity of MIPS and ICP at early stage. To study the initial sensitivity ofMIPS, the ICP and MIPS data from the experimental group and the control groupwere put into independent sample t-test. Results show that the ICP data betweenthe two groups are significantly different after180s. For the New Zealand whiterabbits, the statistical results show that the CSF regulation ability is about1mL. The MIPS data between the two groups are significantly different after30s. Theresults indicate that MIPS is more sensitive than ICP at the primary stage of ICPincrease.(4) Correlation between MIPS and ICP. The19points of ICP and MIPS data in theexperiment group were applied into correlation analysis. The results show that withthe28rabbits of ICP and MIPS, the correlation coefficient is between0.71485and0.98067(p <0.05), which proves that the ICP and MIPS are negatively correlated.(5) Nonlinear regression between MIPS and ICP. Nonlinear regression analysis showsthat relationship between MIPS and ICP can be expressed asP9.404e-0.91.According to the theoretical derivation of MIPS and ICP, the normal ICP of ableed-less rabbit is about9.404mmHg, and the ICP of rabbits increasesexponentially with MIPS. The change rule is decided not only by the bleedingvolume, but also by the overall changes of cerebral conductivity. Statistical resultsshow that at the early stages of pressure increase, MIPS can reflect the change ofICP in the form of exponential function.Conclusions:Compared with previous MIPS measurement systems, we established a rapidsoftware-based phase detector system based on PCI-5124acquisition card. With this newsystem, the acquisition of one MIPS signal only takes0.057s. Archimedes spiral anddifferential combination sensor were used in this system to improve sensitivity. Theexperimental results show that the relationship between MIPS change and cerebralhemorrhage volume change is not strictly linear. For New Zealand white rabbits, thestatistical results show that CSF regulation ability is about1mL. The results indicate thatMIPS is sensitive than ICP at the early stage of ICP increase. These results suggest thatMIPS technology can help clinicians to quickly find physiological brain changes, provideearly warning for clinical use, and improve early diagnosis and treatment. Since thecorrelation between MIPS and ICP signals is linear, we find through the nonlinearregression that the relation between MIPS and ICP can be expressed as.Statistical results show that at the early stages of the pressure increase, MIPS can reflect thechange of ICP in the form of exponential function. Overall, MIPS mehod provides one mean of ICP monitoring, especially at the initial stage of ICP increase. The relationshipbetween MIPS and ICP at the middle and late stages of ICP increase needs further research.