Newborn brain imaging system and method based on photoacoustic technology

Abstract

The invention discloses a neonatal brain imaging system and method based on a photoacoustic technology, and belongs to the field of medical instruments. The system comprises a control module, an excitation source module, a newborn brain photoacoustic data acquisition module, a data output module and a quantitative calculation module. Wherein the control module comprises two computers and a laser control instrument; the excitation source module comprises a high-frequency laser and a water cooling system thereof; the neonatal brain photoacoustic data acquisition module comprises a focusing lens, an optical fiber, a wearable flexible ultrasonic transducer and a multi-channel acquisition card; the data output module comprises a computer matched with real-time imaging MarsonicsDAQ software and an imaging algorithm embedded in the software; the quantitative calculation module comprises a multi-spectral linear unmixing algorithm and imaging software. The invention provides high-resolution and high-contrast real-time imaging of the brain tissue structure and cerebral hemodynamic parameters of the newborn, and can be used as a supplementary mode of the existing brain imaging technology for auxiliary diagnosis.

Description

Technical Field

[0001] The present invention belongs to the field of medical devices, and in particular relates to a neonatal brain imaging detection system and method based on photoacoustic technology. Background Art

[0002] With the rapid development of society and economy, the continuous advancement of modern medicine and diagnostic equipment, the early diagnosis and intervention of diseases have attracted increasing attention. Despite significant progress in the medical field, the diagnosis of neonatal encephalopathy remains a challenging task. This problem presents a series of difficulties in clinical practice, requiring continuous in-depth research and the use of advanced medical technology and scientific research results to find more accurate and effective diagnostic methods. Therefore, strengthening the research on neonatal encephalopathy and improving the level of diagnosis and treatment effectiveness have become one of the important issues that need to be addressed in the current medical field.

[0003] Intracranial hemorrhage (ICH), also known as hemorrhagic cerebrovascular disease or hemorrhagic stroke, is a brain disease caused by the rupture of a cerebral blood vessel, resulting in the extravasation of blood into the cranial cavity. It is a common disease in early neonatal life and a leading cause of death. Its diagnosis and treatment are routine in neonatal wards both domestically and internationally. Hypoxic-ischemic encephalopathy (HIE) in neonates refers to hypoxic-ischemic damage to the brain caused by perinatal asphyxia. Clinically, it presents with a range of abnormal central nervous system manifestations, such as altered consciousness, altered muscle tone, abnormal primitive reflexes, and, in severe cases, convulsions, brainstem signs, and increased anterior fontanelle tension. Both neonatal ICH and HIE are characterized by high morbidity, mortality, and disability rates.

[0004] Studies have shown that ICH occurs in approximately 13% of hospitalized newborns and as high as 40%-70% of premature infants. Furthermore, severe cerebral hemorrhage can cause permanent damage to the child, leading to serious sequelae such as hydrocephalus, hemiplegia, or intellectual disability, and can even be life-threatening. Studies have reported a mortality rate of 9.0%-18.5%. HIE occurs in 2‰-8‰ of hospitalized newborns, but the mortality rate among affected newborns is as high as 15%-20%. Furthermore, 25%-30% of surviving children often suffer from various neurological disorders that can lead to disability, including developmental delay, cerebral palsy, epilepsy, visual impairment, and learning disabilities.

[0005] Epilepsy in children is a common, recurrent neurological syndrome with complex etiologies. It is characterized by paroxysmal, transient seizures resulting from brain dysfunction. Clinical manifestations include recurrent muscle twitching and transient abnormalities in consciousness, sensation, and emotion. It is primarily caused by incomplete development of the child's nervous system, which leads to excessive and abnormal electrical discharges in the cerebral cortex due to stimulation. Hydrocephalus refers to the accumulation of cerebrospinal fluid within the ventricular system, resulting in increased intraventricular pressure. Key symptoms include developmental delay, headache, vomiting, and lethargy. Physical signs include an enlarged head circumference, a tense fontanelle, difficulty looking upward, papilledema, and abducens nerve palsy.

[0006] Neonatal cerebral hemorrhage and hypoxic-ischemic encephalopathy are characterized by neurological damage, high mortality, and severe neurological sequelae. Severe neurological damage can lead to symptoms such as loss of consciousness, abnormal heart rate, and respiratory distress, as well as the gradual necrosis of neurons and nerve cells. Pediatric epilepsy and hydrocephalus are also common neurological syndromes in childhood, significantly impacting the growth, development, and normal function of the neonatal nervous system. Early diagnosis and treatment are crucial for neonatal encephalopathy, as timely treatment can minimize potential neurological damage. Therefore, the development of a rapid and accurate brain diagnostic instrument is essential. The neonatal brain imaging system developed in this patent can, based on photoacoustic images and hemodynamic parameter assessments, pinpoint the location and severity of cerebral hemorrhage, hypoxic-ischemic encephalopathy, epileptic encephalopathy, and hydrocephalus. This system assists medical staff in developing appropriate interventional treatment plans, enabling early detection and treatment to prevent further deterioration, halt ongoing brain damage, reduce neonatal disability, and improve prognosis.

[0007] Modern medical imaging techniques for brain examinations include electroencephalography (EEG), transcranial Doppler ultrasound (TCD), computed tomography (CT), functional magnetic resonance imaging (fMRI), single-photon emission computed tomography (SPECT), positron emission computed tomography (PET), and near-infrared functional imaging (NIFI). EEG continuously monitors brain electrophysiological signals, but cannot precisely locate the location and size of lesions, making quantitative assessment difficult. TCD can quantitatively assess lesion location, but requires high skill levels and produces low contrast in brain tissue. CT provides clear images of brain tissue, but cannot quantitatively assess the condition and emits ionizing radiation, with unknown effects on newborns. FMRI offers multi-axial imaging, high resolution, and no radiation damage. However, the equipment is sensitive to environmental conditions, the examination is time-consuming, noisy, and requires the patient to remain still and immobile. Anesthesia or sedation may be necessary. In addition, functional magnetic resonance imaging equipment is expensive and the testing cost is high; single-photon emission computed tomography and positron emission tomography can semi-quantitatively or quantitatively measure local cerebral blood flow, but these two technologies not only require the injection of exogenous radioactive tracers, but also have low spatial resolution, high prices, and high equipment operation and maintenance costs; near-infrared functional imaging can diagnose brain tissue lesions, but its spatial resolution is limited, and it cannot accurately locate the damaged site, and it is difficult to detect tiny lesions.

[0008] In traditional photoacoustic brain imaging systems, linear array ultrasound transducers typically only provide imaging at a certain angle, with a limited imaging range that may not cover the entire sample area and poor resolution. Semi-circular array ultrasound transducers, on the other hand, are limited to the focal radius of the probe itself, resulting in different imaging effects for patients with different head circumferences. Therefore, the present invention utilizes a wearable flexible ultrasound transducer made of a flexible circuit. This design allows for autonomous adjustment of the transducer's shape and focal radius based on the individual patient's head circumference, maximizing signal reception and effectively enhancing the resolution and contrast of the photoacoustic image.

[0009] Furthermore, conventional photoacoustic brain imaging systems often use less accurate photoacoustic reconstruction algorithms, such as the delayed-addition algorithm and filtered back-projection algorithm. This invention employs a finite element-based photoacoustic reconstruction algorithm, solving the time-domain photon diffusion equation through iterative calculations to obtain an accurate solution for the medium / tissue absorption coefficient. This method improves the accuracy of reconstructed images and provides a reliable data foundation for quantitative analysis of imaging results. This optimized algorithm enables more accurate assessment of imaging results, enabling more efficient diagnosis and research in the field of biomedical imaging. Summary of the Invention

[0010] The present invention aims to solve the above problems of the prior art. It proposes a neonatal brain imaging system and method based on photoacoustic technology. The technical solution of the present invention is as follows:

[0011] A neonatal brain imaging system based on photoacoustic technology is characterized by comprising: a control module (3), an excitation source module (4), a neonatal brain photoacoustic data acquisition module (5) and a data output module (8); the output end of the neonatal brain photoacoustic data acquisition module (5) is connected to the receiving end of the data output module (8); wherein,

[0012] The control module includes a computer 1, a computer 2 and a laser control instrument (10), wherein the computer 1 is used to indirectly control the frequency, wavelength and energy of the laser by controlling the laser control instrument, and the computer 2 is used to control the magnification of the neonatal brain photoacoustic data acquisition module (5), the frequency range of the acquisition signal, and control the output of the image through the MarsonicsDAQ software (18);

[0013] The excitation source module (4) includes a high-frequency laser (12) and a water cooling system (11), wherein the high-frequency laser (12) is used to output pulsed laser light, and the water cooling system (11) is used to control the ambient temperature of the laser pump source;

[0014] The neonatal brain photoacoustic data acquisition module (5) includes a laser transmission module (6) and a data acquisition module (7), wherein the laser transmission module (6) includes a convex lens (13) and an optical fiber (14); the pulsed laser output by the laser is converged into the optical fiber (14) through a focusing lens; the laser is evenly irradiated onto the biological tissue to be measured after being transmitted through the optical fiber to generate a photoacoustic signal, and the ultrasonic signal is collected by a flexible ultrasonic transducer (15) worn on the neonatal brain, and the ultrasonic transducer is electrically connected to the acquisition card in a one-way manner. The photoacoustic signal received by the ultrasonic transducer is sent to a multi-channel data acquisition card (16) for data processing, and then the processed signal is transmitted to the MarsonicsDAQ software (18) for image reconstruction;

[0015] The algorithm of the data output module (8) includes a photoacoustic reconstruction algorithm based on finite element iteration; the MarsonicsDAQ software (18) is loaded into the computer 2, and the data transmitted by the acquisition card is displayed in real time through the embedded photoacoustic reconstruction algorithm based on finite element iteration;

[0016] The multi-spectral linear unmixing algorithm of the quantitative calculation module (9) is a quantitative calculation method developed based on MATLAB software. It calculates cerebral hemodynamic parameters by linearly unmixing the multi-spectral photoacoustic imaging data.

[0017] Furthermore, the high-frequency laser is used to output pulsed lasers with a repetition frequency of 500Hz and an energy peak of 26.7mJ / cm2. Its matching water cooling system can control the ambient temperature of the laser pump source at around 24°C. The high-frequency laser is also used to achieve rapid wavelength switching. When the pulse frequency is 500Hz, it can quickly switch between the two wavelengths of 730nm and 875nm, that is, only the wavelength is changed while the other experimental conditions remain unchanged.

[0018] Furthermore, the optical fiber is 1.2m long, the light inlet is a circle with a diameter of 0.9cm, and the light outlet is a rectangle with a length of 40mm and a width of 1mm. The laser is focused by a convex lens and enters the optical fiber evenly, and emits a rectangular laser with a length of 40mm and a width of 1mm and uniform energy at the optical fiber outlet. The laser energy is about 2mJ / cm 2 .

[0019] Furthermore, the ultrasonic transducer has 256 array elements, which can be freely adjusted to the corresponding curvature according to the different contours of the baby's head; by using ultrasonic coupling fluid, the ultrasonic transducer and the skin can be tightly coupled; the multi-channel data acquisition card has 256 channels, and the signal amplification factor is adjustable from 1dB to 100dB.

[0020] Furthermore, the real-time image display can determine whether the brain tissue is bleeding, hypoxic-ischemic, abnormal brain activity, and whether there is fluid accumulation. The judgment rule is: the distribution of light absorption coefficients in the brain tissue imaging area is calculated by a photoacoustic reconstruction algorithm based on finite element iteration. Bright areas will appear in areas of brain tissue bleeding and abnormal brain excitement, and dark areas will appear in areas of hypoxia-ischemic and cerebral effusion. The brain tissue structure in the reconstructed image can also be observed to determine whether the internal brain structure is dislocated. The principle of the photoacoustic reconstruction algorithm based on finite element iteration is to use the finite element method FEM to iteratively solve the radiation transfer equation RTE and the photoacoustic wave equation PWE. First, from the initial distribution of the light absorption coefficient, the distribution of the light flux is obtained by solving the finite element FE of RTE. Then, the Helmholtz equation is used to calculate the sound pressure at each transducer position along the medium / tissue surface. Then, the canonical Newton method is used to obtain the matrix equation for absorption coefficient inversion. Finally, the absorption coefficient distribution is updated by iteratively solving the equation to form a photoacoustic reconstructed image.

[0021] Furthermore, the distribution of the luminous flux is obtained by solving the finite element FE of the RTE from the initial distribution of the light absorption coefficient; then use Formula (1) is used to solve the forward solution, and formulas (2) and (3) are used to solve the inverse solution. The forward solution and the backward solution are then substituted into the time-domain photon diffusion equation to perform iterative operations to solve the absorption coefficient of the medium / biological tissue. Finally, the absorption coefficient distribution is updated by iteratively solving the equation. Specifically, the following steps are included:

[0022] Forward solution:

[0023]

[0024] Reverse solution:

[0025] (J T J+λI) Δχ=J T (p o -p c ) (2)

[0026] Δχ=Ψ o -Ψ c (3)

[0027] Light diffusion equation:

[0028] ▽·D▽Φ(r)-μ a Φ(r)=-S(r) (4)

[0029] Where p(r,t) is the sound pressure at time t at position r in the medium, 2 is the Laplace operator, v0 is the speed of sound wave in the medium, β is the thermal expansion coefficient, C p is the constant pressure specific heat capacity, ψ(r) is the spatial distribution of the initial light absorption energy, I(t) is the time-varying function of the light intensity, and J is the Jacobian matrix consisting of the partial derivatives of the sound pressure value measured on the boundary with respect to the absorbed energy density. λ is the regularization parameter, Δχ is the update vector of the absorbed energy density, and p o is the actual sound pressure distribution, p c is the sound pressure distribution obtained by forward solution given the initial absorbed energy density, ψ o is the absorbed energy density distribution obtained by finite element discretization of the wave equation, ψ c is the calculated absorbed energy density distribution, D is the diffusion coefficient, Φ(r) is the photon density, μ a is the absorption coefficient, and S(r) is the photon flux at position r.

[0030] Furthermore, the multispectral linear unmixing algorithm is specifically as follows: photoacoustic imaging of the neonatal brain is performed using two wavelength lasers of 730nm and 875nm respectively, and then the imaging data is substituted into the formula of the relationship between endogenous contrast agents to calculate oxygenated hemoglobin, deoxygenated hemoglobin, total hemoglobin, oxygen saturation and water content.

[0031] Furthermore, the imaging data is substituted into the formula of the relationship between endogenous contrast agents to calculate important hemodynamic parameter indicators such as oxygenated hemoglobin, deoxygenated hemoglobin, total hemoglobin, oxygen saturation, and water content, including:

[0032]

[0033] sO2=f*sO 2,ti +(1-f)*f*sO 2,to (7)

[0034]

[0035] Where, The photoacoustic image at wavelength λ Photoacoustic signal intensity at the position; ε HbO (λ), ε HbR (λ) and represent the absorption coefficients of oxygenated hemoglobin, deoxygenated hemoglobin and water at wavelength λ respectively; sO2 and [HbT] were The concentration distribution of oxygenated hemoglobin, deoxygenated hemoglobin, water, oxygen saturation and total hemoglobin at sO 2,ti and sO 2,to are the average oxygen saturation at the tissue inlet (artery) and outlet (venous), respectively, and f is sO 2,ti Where BF is the average blood flow of all blood vessels in the tissue and is designated as the average blood flow of the tissue, OC=MRO2 is the average oxygen consumption (oxygen metabolism) of the entire tissue, [HbT] blood is the average total hemoglobin concentration calculated for the tissue, and V is the tissue volume.

[0036] A detection method based on any of the above systems, comprising the following steps:

[0037] Step 1. Turn on the laser and multi-channel acquisition card, adjust the acquisition card trigger mode to laser pulse passive triggering, and then set the laser pulse wavelength to 730nm and the pulse laser related parameters and acquisition card sampling parameters;

[0038] Step 2: Apply appropriate ultrasonic coupling fluid to the detection position of the infant's brain, wear the flexible ultrasonic transducer on the brain of the newborn to be tested, and ensure that the detection array element is in close contact with the newborn's brain skin;

[0039] Step 3: gradually turn on the high-frequency laser pump source and the pulse laser emission switch through computer 1, and preheat the laser for about half an hour;

[0040] Step 4: The pulsed laser passes through a convex lens and an optical fiber and is irradiated onto the surface of the newborn's brain. The brain tissue absorbs the laser energy and converts it into heat energy, causing transient temperature changes. Further thermal expansion leads to local pressure changes, generating ultrasonic waves.

[0041] Step 5: Using a wearable ultrasound transducer to receive photoacoustic signals at multiple locations in the newborn's brain;

[0042] Step 6: The photoacoustic signal received by the ultrasonic transducer is transmitted wirelessly to a multi-channel data acquisition card for data processing;

[0043] Step 7: Maintain all experimental conditions unchanged and adjust the wavelength of the pulsed laser to 875 nm;

[0044] Step 8: The collected photoacoustic data are combined with MATLAB software and finite element photoacoustic reconstruction algorithm to qualitatively determine whether there is a lesion in the neonatal brain tissue;

[0045] Step 9: If the reconstructed brain tissue image is abnormal, the multispectral linear unmixing algorithm is further used to quantify the extent of brain lesions;

[0046] Step 10: Output the qualitative and quantitative reconstruction results in steps 8 and 9.

[0047] The advantages and beneficial effects of the present invention are as follows:

[0048] Existing brain imaging technology often has certain deficiencies in the diagnosis of encephalopathy such as neonatal cerebral hemorrhage, cerebral hypoxic-ischemic encephalopathy, and hydrocephalus. Compared with electroencephalography, the present invention can accurately locate the location of the lesion and provide the distribution of cerebral hemodynamic parameters around the lesion; compared with transcranial Doppler ultrasound, the present invention can perform high-contrast, high-resolution imaging; compared with computed tomography, single-photon emission computed tomography, and positron emission computed tomography, the present invention can perform continuous bedside detection without ionizing radiation; compared with magnetic resonance imaging, the present invention can perform fast, real-time imaging without contraindications; compared with near-infrared imaging, the present invention can clearly perform high-resolution imaging of the lesion and can more accurately locate the lesion.

[0049] The present invention's neonatal brain imaging system based on photoacoustic technology uses a finite element-based photoacoustic reconstruction method and a multispectral linear unmixing algorithm. The precise data solution from the finite element reconstruction algorithm is substituted into the multispectral linear unmixing algorithm, helping to more accurately assess hemodynamic changes and thus achieve a more accurate and reliable diagnosis. The system can simultaneously provide high-resolution, high-contrast real-time images of the internal structure of brain tissue, the distribution of brain tissue light absorption coefficients, and cerebral hemodynamic parameters. The system can determine the location and size of brain lesions based on the light absorption coefficient distribution map, and the extent of brain lesions based on the hemodynamic parameter image. This system can assist medical staff in developing appropriate intervention and treatment plans, enabling early detection and treatment to prevent further deterioration of the disease, inhibit ongoing brain damage, reduce neonatal disability rates, and improve prognosis.

[0050] The ingenuity of the present invention lies in: (1) The use of a wearable flexible ultrasonic transducer can be appropriately bent according to the different shapes and sizes of the infant's brain, closely fitting the infant's brain for photoacoustic imaging signal acquisition, making the acquisition operation more convenient and flexible, and the acquisition equipment more universal. At the same time, each array element on the flexible probe is precisely positioned through optical frequency domain scattering technology, and combined with a reconstruction algorithm developed specifically for the flexible probe, the key image quality parameters such as resolution and contrast of the reconstructed image can be significantly improved, thereby further optimizing the photoacoustic imaging results. In addition, the wearable flexible ultrasonic transducer can create relaxed testing conditions for infants and young children, avoiding excessive stimulation of the newborn's emotions during the testing process, which may aggravate the severity of brain pathological encephalopathy such as cerebral hemorrhage, cerebral hypoxia-ischemia, epileptic encephalopathy and hydrocephalus; (2) MATLAB software combined with the finite element-based photoacoustic reconstruction algorithm can quickly and in real time determine whether there is ischemia, hemorrhage, edema in the brain tissue and whether the internal structure of the brain tissue is displaced; (3) Combined with the multi-spectral linear unmixing algorithm, the distribution of hemodynamic parameters of the brain tissue is calculated to qualitatively and quantitatively determine the severity of the disease in the newborn (3) The reconstruction algorithm used in the present invention is a dual-grid finite element photoacoustic reconstruction algorithm, which can perform preliminary calculations on a coarser grid and then perform local refinement calculations on a fine grid, thereby reducing the overall calculation amount and improving the calculation efficiency. At the same time, it combines the global information of the coarse grid and the local details of the fine grid, which helps to improve the accuracy of the reconstructed image, especially when processing tissues with uneven acoustic characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 1 is a schematic structural diagram of a neonatal brain imaging system applicable to photoacoustic technology according to a preferred embodiment of the present invention;

[0052] Figure 2 This is a schematic diagram of the diagnostic process for neonatal patients according to an embodiment of the present invention;

[0053] Figure 3 Schematic diagram of the operation flow of the neonatal brain photoacoustic imaging system in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following will describe the technical solutions in the embodiments of the present invention in detail with reference to the accompanying drawings. The described embodiments are only a part of the embodiments of the present invention.

[0055] The technical solution of the present invention to solve the above technical problems is:

[0056] like Figure 1As shown, the neonatal brain imaging system based on photoacoustic technology of the present invention includes a control module 3, an excitation source module 4, a neonatal brain photoacoustic data acquisition module 5, a data output module 8 and a quantitative calculation module 9; the output end of the receiving module 7 in the signal neonatal brain photoacoustic data acquisition module 5 is connected to the receiving end of the data output module 8;

[0057] The control module 3 includes computers 1, 2, and a laser control instrument 10. The entire system controls the laser control instrument 10 through computer 1 to indirectly control the frequency, wavelength, and energy of the laser. Computer 2 controls the amplification factor and frequency range of the collected signal of the signal acquisition module 7, and controls the image output through MarsonicsDAQ software 18.

[0058] The excitation source module 4 includes a high-frequency laser 12 and a water cooling system 11 thereof;

[0059] Based on the above embodiment, the high-frequency laser 12 can achieve rapid wavelength switching, with a maximum repetition frequency of 500 Hz and an energy peak of 26.7 mJ / cm 2 .

[0060] On the basis of the above embodiment, the water cooling system 11 can control the ambient temperature of the laser pump source to be around 24° C., thereby ensuring the stability of the output laser.

[0061] The neonatal brain photoacoustic data acquisition module 5 includes a laser transmission module 6 and a data acquisition module 7. The laser transmission module 6 includes a convex lens 13 and an optical fiber 14. The data acquisition module 7 includes a wearable flexible ultrasonic transducer 15 and a multi-channel acquisition card 16. The pulsed laser output by the laser is converged into the optical fiber 14 through the focusing lens 13. After being transmitted through the optical fiber 14, the laser is uniformly irradiated onto the biological tissue to be measured to generate a photoacoustic signal. The optical fiber 14 is a silica optical fiber with a small absorption coefficient in the specific wavelength range of 730nm to 875nm to reduce energy loss in laser transmission. The laser is transmitted through the optical fiber 14 to irradiate the biological tissue to be measured to generate a photoacoustic signal. The ultrasonic signal is collected by the flexible ultrasonic transducer 15 worn on the neonatal brain. The ultrasonic transducer 15 is electrically connected to the acquisition card 16 in a unidirectional manner. The photoacoustic signal received by the ultrasonic transducer 15 will be sent to the multi-channel data acquisition card 16 for data processing, and then the processed signal will be transmitted to the MarsonicsDAQ software for image reconstruction.

[0062] Based on the above embodiment, the optical fiber 14 is 1.2m long, the light inlet is a circle with a diameter of 0.9cm, and the light outlet is a rectangle with a length of 40mm and a width of 1mm. The laser is focused by a convex lens and enters the optical fiber evenly, and emits a rectangular laser with a length of 40mm and a width of 1mm and uniform energy at the optical fiber outlet. The laser energy is about 2mJ / cm2 , far below the laser energy safety standard of 10mJ / cm 2 .

[0063] Building on the above-mentioned embodiment, the wearable flexible ultrasonic transducer 15 comprises 256 elements and offers excellent free-bending properties, allowing it to be adjusted to a specific curvature based on the contours of a baby's head. By using ultrasonic coupling fluid, the ultrasonic transducer is tightly coupled to the skin, thereby improving the sensor's sensitivity and accuracy.

[0064] On the basis of the above embodiment, the multi-channel data acquisition card 16 has 256 channels and receives data collected by the ultrasonic transducer through a wireless connection; its signal amplification factor is adjustable from 1dB to 100dB.

[0065] The MarsonicsDAQ imaging software 18 of the data output module 8 is developed based on the multi-channel acquisition card 16 in the acquisition module. The algorithm includes an embedded finite element-based photoacoustic reconstruction algorithm 17. The MarsonicsDAQ software 18 is loaded into the computer 2, and the data transmitted by the acquisition card 16 is displayed in real time using its embedded algorithm.

[0066] Furthermore, based on the above embodiment, the rule for determining whether there is hemorrhage or hypoxia-ischemia is as follows: the light absorption coefficient distribution in the brain tissue imaging area is calculated by a finite element-based photoacoustic reconstruction algorithm. A bright area will appear in the brain tissue hemorrhage area, and a dark area will appear in the hypoxia-ischemia area. The principle of the photoacoustic reconstruction algorithm based on finite element iteration is to use the finite element method (FEM) to iteratively solve the radiation transfer equation (RTE) and the photoacoustic wave equation (PWE). First, from the initial (guessed) distribution of the light absorption coefficient, the distribution of the light flux is obtained by solving the finite element (FE) of the RTE. Then, the Helmholtz equation is used to calculate the sound pressure at each transducer position along the medium / tissue surface. Then, the canonical Newton method is used to obtain the matrix equation for the absorption coefficient inversion. Finally, by iteratively solving the equation, the absorption coefficient distribution is updated to form a photoacoustic reconstructed image.

[0067] The multi-spectral linear unmixing algorithm of the quantitative calculation module is a quantitative calculation method developed based on MATLAB software. It calculates cerebral hemodynamic parameters by linearly unmixing the multi-spectral photoacoustic imaging data.

[0068] Furthermore, based on the above embodiment, real-time image display can determine whether there is hemorrhage, hypoxia-ischemia, abnormal brain excitability, or cerebrospinal fluid in the brain tissue. The judgment rule is as follows: the specific values ​​of oxygenated hemoglobin, deoxygenated hemoglobin, total hemoglobin, oxygen saturation, and water content parameters in the newborn brain are calculated using a multispectral linear unmixing algorithm 19. By comparing the difference with the normal newborn brain hemodynamic parameters, the severity of the newborn's illness can be determined.

[0069] Furthermore, the principle of the multispectral linear unmixing image reconstruction algorithm16 is to use two wavelengths of 730nm and 875nm to clearly image the brain area, and then substitute them into the formula of the relationship between endogenous contrast agents to calculate oxygenated hemoglobin, deoxygenated hemoglobin, total hemoglobin, blood flow velocity, oxygen saturation, oxygen metabolic rate and water content. These indicators can be used as direct criteria for early diagnosis of the degree of brain lesions.

[0070] like Figure 3 As shown, the neonatal brain imaging system and method based on photoacoustic technology of the present invention includes the following steps:

[0071] Step 1. Turn on the laser and multi-channel acquisition card, adjust the acquisition card trigger mode to laser pulse passive triggering, and then set the laser pulse wavelength to 730nm and the pulse laser related parameters and acquisition card sampling parameters;

[0072] Step 2: Apply appropriate ultrasonic coupling fluid to the detection position of the infant's brain, wear the flexible ultrasonic transducer on the brain of the newborn to be tested, and ensure that the detection array element is in close contact with the newborn's brain skin;

[0073] Step 3: Use the computer to gradually turn on the high-frequency laser pump source and pulse laser emission switch, and preheat the laser for about half an hour;

[0074] Step 4: The pulsed laser passes through a convex lens and an optical fiber and is irradiated onto the surface of the newborn's brain. The brain tissue absorbs the laser energy and converts it into heat energy, causing transient temperature changes. Further thermal expansion leads to local pressure changes, generating ultrasonic waves.

[0075] Step 5: Using a wearable ultrasound transducer to receive photoacoustic signals at multiple locations around the neonate's brain;

[0076] Step 6: The photoacoustic signal received by the ultrasonic transducer is transmitted wirelessly to a multi-channel data acquisition card for data processing;

[0077] Step 7: Maintain all experimental conditions unchanged and adjust the wavelength of the pulsed laser to 875 nm;

[0078] Step 8: The collected photoacoustic data are combined with MATLAB software and a finite element-based photoacoustic reconstruction algorithm to qualitatively determine whether there is a lesion in the neonatal brain tissue;

[0079] Step 9: If the reconstructed brain tissue image is abnormal, the multispectral linear unmixing algorithm is further used to quantify the extent of brain lesions;

[0080] Step 10: Output the qualitative and quantitative reconstruction results in steps 8 and 9.

[0081] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.

[0082] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0083] The above embodiments should be understood as merely illustrating the present invention and not as limiting the scope of protection of the present invention. After reading the contents of the present invention, technicians may make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A neonatal brain imaging system based on photoacoustic technology, characterized in that: include: A control module (3), an excitation source module (4), a neonatal brain photoacoustic data acquisition module (5) and a data output module (8); the output end of the neonatal brain photoacoustic data acquisition module (5) is connected to the receiving end of the data output module (8); wherein, The control module includes a computer 1, a computer 2 and a laser control instrument (10), wherein the computer 1 is used to indirectly control the frequency, wavelength and energy of the laser by controlling the laser control instrument (10), and the computer 2 is used to control the magnification of the neonatal brain photoacoustic data acquisition module (5), the frequency range of the acquisition signal, and control the output of the image through the MarsonicsDAQ software (18); The excitation source module (4) comprises a high-frequency laser (12) and a water cooling system (11), wherein the high-frequency laser (12) is used to output pulsed laser, and the water cooling system (11) is used to control the ambient temperature of the laser pump source; The neonatal brain photoacoustic data acquisition module (5) comprises a laser transmission module (6) and a data acquisition module (7), wherein the laser transmission module (6) comprises a convex lens (13) and an optical fiber (14); the pulsed laser output by the laser is converged into the optical fiber (14) through a focusing lens; the laser is transmitted through the optical fiber and uniformly irradiates the biological tissue to be tested to generate a photoacoustic signal, and the ultrasonic signal is collected by a flexible ultrasonic transducer (15) worn on the neonatal brain, the ultrasonic transducer is electrically connected to the acquisition card in a unidirectional manner, and the photoacoustic signal received by the ultrasonic transducer is sent to a multi-channel data acquisition card (16) for data processing, and then the processed signal is sent to the MarsonicsDAQ software (18) for image reconstruction; The algorithm of the data output module (8) includes a photoacoustic reconstruction algorithm based on finite element iteration; the MarsonicsDAQ software (18) is loaded into the computer 2, and the data transmitted by the acquisition card is displayed in real time through the photoacoustic reconstruction algorithm based on finite element iteration embedded in the software; The multi-spectral linear unmixing algorithm of the quantitative calculation module (9) is a quantitative calculation method developed based on MATLAB software. By linearly unmixing the photoacoustic imaging data under multi-spectral conditions, cerebral hemodynamic parameters are calculated.

2. A neonatal brain imaging system based on photoacoustic technology according to claim 1, characterized in that: The high-frequency laser is used to output pulsed laser with a repetition frequency of 500Hz and an energy peak of 26.7mJ / cm2. Its matching water cooling system can control the ambient temperature of the laser pump source at around 24°C. The high-frequency laser is also used to achieve rapid switching of wavelengths. When the pulse frequency is 500Hz, it can quickly switch between the two wavelengths of 730nm and 875nm, that is, only the wavelength is changed and the other experimental conditions remain unchanged.

3. The neonatal brain imaging system based on photoacoustic technology according to claim 1, characterized in that: The optical fiber is 1.2m long, the light inlet is a circle with a diameter of 0.9cm, and the light outlet is a rectangle with a length of 40mm and a width of 1mm. The laser is focused by a convex lens and then enters the optical fiber, and a rectangular laser with a length of 40mm and a width of 1mm and uniform energy is emitted from the optical fiber light outlet. The laser energy is about 2mJ / cm 2 .

4. The neonatal brain imaging system based on photoacoustic technology according to claim 1, characterized in that: The ultrasonic transducer has 256 array elements, which are used to freely adjust to the corresponding curvature according to the contours of different infant heads; by using ultrasonic coupling fluid, the ultrasonic transducer and the skin can be tightly coupled; the multi-channel data acquisition card has 256 channels, and the signal amplification factor is adjustable from 1dB to 100dB.

5. The neonatal brain imaging system based on photoacoustic technology according to claim 1, characterized in that: The real-time image display can determine whether the brain tissue is bleeding, hypoxic-ischemic, abnormal brain activity, and whether there is fluid accumulation. The judgment rule is: the distribution of light absorption coefficients in the brain tissue imaging area is calculated by a photoacoustic reconstruction algorithm based on finite element iteration. Bright areas will appear at the bleeding sites of brain tissue and abnormal excitement of brain areas, and dark areas will appear at the sites of hypoxic-ischemic and cerebral effusion. It is also possible to determine whether the internal brain structure is dislocated by observing the brain tissue structure in the reconstructed image. The principle of the photoacoustic reconstruction algorithm based on finite element iteration is to use the finite element method FEM to iteratively solve the radiation transfer equation RTE and the photoacoustic wave equation PWE. First, from the initial distribution of the light absorption coefficient, the distribution of the light flux is obtained by solving the finite element FE of RTE. Then, the Helmholtz equation is used to calculate the sound pressure at each transducer position along the medium / tissue surface. Then, the canonical Newton method is used to obtain the matrix equation for absorption coefficient inversion. Finally, the absorption coefficient distribution is updated by iteratively solving the equation to form a photoacoustic reconstructed image.

6. The neonatal brain imaging system based on photoacoustic technology according to claim 1, characterized in that: The method is described as obtaining the distribution of light flux from the initial distribution of light absorption coefficient by solving the finite element FE of RTE; then using formula (1) to solve the forward solution, using formulas (2) and (3) to solve the inverse solution, and then substituting the forward solution and the backward solution into the time-domain photon diffusion equation to perform iterative calculation to solve the absorption coefficient of the medium / biological tissue; finally, updating the absorption coefficient distribution by iteratively solving the equation, specifically comprising the following steps: Forward solution: Reverse solution: (J T J+λI) Δχ=J T (p o -p c ) (2) Δx=Ψ o -P c (3) Light diffusion equation: Where p(r,t) is the sound pressure at position r in the medium at time t, is the Laplace operator, v0 is the propagation speed of sound waves in the medium, β is the thermal expansion coefficient, C p is the constant pressure specific heat capacity, ψ(r) is the spatial distribution of the initial light absorption energy, I(t) is the function of the light intensity changing with time, and J is the Jacobian matrix composed of the partial derivatives of the sound pressure value measured on the boundary with respect to the absorption energy density λ is the regularization parameter, Δχ is the update vector of the absorbed energy density, and p o is the actual sound pressure distribution, p c is the sound pressure distribution obtained by forward solution given the initial absorbed energy density, ψ o is the absorbed energy density distribution obtained by finite element discretization of the wave equation, ψ c is the calculated absorbed energy density distribution, D is the diffusion coefficient, Φ(r) is the photon density, μ a is the absorption coefficient, and S(r) is the photon flux at position r.

7. A neonatal brain imaging system based on photoacoustic technology according to claim 6, characterized in that: The multi-spectral linear unmixing algorithm is specifically as follows: photoacoustic imaging of the neonatal brain is performed using two wavelength lasers of 730nm and 875nm respectively, and then the imaging data is substituted into the formula of the relationship between endogenous contrast agents to calculate oxygenated hemoglobin, deoxygenated hemoglobin, total hemoglobin, oxygen saturation and water content.

8. The neonatal brain imaging system based on photoacoustic technology according to claim 7, characterized in that: Substituting the imaging data into the formulas (5)-(8) of the relationship between endogenous contrast agents, important hemodynamic parameter indicators such as oxygenated hemoglobin, deoxygenated hemoglobin, total hemoglobin, oxygen saturation and water content are calculated, including: sO2=f*sO 2,ti +(1-f)*f*sO 2,to (7) In the formula, The photoacoustic image at wavelength λ The photoacoustic signal intensity at the position; ε HbO (λ), ε HbR (λ) and Respectively represent the absorption coefficients of oxygenated hemoglobin, deoxygenated hemoglobin and water at wavelength λ; sO2 and [HbT] were The concentration distribution of oxygenated hemoglobin, deoxygenated hemoglobin, water, oxygen saturation and total hemoglobin at the 2,ti and sO 2,to are the average oxygen saturation at the tissue inlet (artery) and outlet (venous), respectively, and f is sO 2,ti coefficient; where BF is the average blood flow of all vessels in the tissue and is designated as the average blood flow of the tissue, OC=MRO2 is the average oxygen consumption (oxygen metabolism) of the entire tissue, [HbT] blood is the average total hemoglobin concentration calculated over the tissue, and V is the tissue volume.

9. A detection method based on the system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Turn on the laser and multi-channel acquisition card, adjust the acquisition card trigger mode to laser pulse passive trigger, and then set the laser pulse wavelength to 730nm and the pulse laser related parameters and acquisition card sampling parameters; Step 2: Apply appropriate ultrasonic coupling liquid to the detection position of the infant's brain, wear the flexible ultrasonic transducer on the brain of the newborn to be tested, and ensure that the detection array element is in close contact with the newborn's brain skin; Step 3, gradually turn on the high-frequency laser pump source and the pulse laser emission switch through computer 1, and preheat the laser for about half an hour; Step 4: The pulsed laser is irradiated onto the surface of the newborn's brain through a convex lens and an optical fiber. The brain tissue absorbs the laser energy and converts it into heat energy, causing transient temperature changes, further generating thermal expansion, leading to local pressure changes, and generating ultrasonic waves. Step 5, using a wearable ultrasound transducer to receive photoacoustic signals at multiple locations in the newborn's brain; Step 6: The photoacoustic signal received by the ultrasonic transducer enters the multi-channel data acquisition card through wireless transmission for data processing; Step 7, keeping all experimental conditions unchanged, adjust the wavelength of the pulsed laser to 875nm; Step 8: The collected photoacoustic data are combined with MATLAB software and finite element photoacoustic reconstruction algorithm to qualitatively determine whether there is a lesion inside the neonatal brain tissue; Step 9: If the reconstructed imaging of brain tissue is abnormal, the degree of lesions in the brain region is quantified by further combining the multispectral linear unmixing algorithm; Step 10: Output the qualitative and quantitative reconstruction results in step 8 and step 9.

Images