Handheld photoacoustic multimodal blood vessel puncture and tissue damage diagnosis system and device

Abstract

The application provides a handheld photoacoustic multi-modal blood vessel puncture and tissue injury diagnosis system and device, and belongs to the technical field of photoacoustic imaging. The handheld multi-spectral photoacoustic imaging device integrates elastic force feedback and real-time blood vessel imaging, so that the puncture success rate is increased by 30%; hemoglobin / collagen double-index imaging can early warn pressure injury by 72 hours; the error of 3D hematoma volume automatic calculation is less than 5%, and the repeated puncture and missed diagnosis are significantly reduced. The application provides a quantitative evaluation and real-time navigation integrated scheme for blood vessel puncture and tissue injury, and is safe and efficient.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of photoacoustic imaging, and particularly relates to a handheld photoacoustic multi-modal blood vessel puncture and tissue damage diagnosis system and device. BACKGROUND

[0002] In the process of clinical nursing and emergency treatment, venipuncture is one of the most common but most experience-dependent invasive operations. The existing technology mainly relies on nurses' naked eye observation, palpation or the aid of traditional transmission-type vein imaging instruments, but due to skin pigmentation, hair, subcutaneous fat thickness and environmental light interference, the venous imaging of the puncture site is often not clear, resulting in a low first-time puncture success rate. Repeated puncture not only increases the pain of patients, but also may cause complications such as phlebitis, hematoma and nerve damage. The problem of "multiple failures of venipuncture" is particularly prominent in pediatric, obese, elderly and hypovolemic patients, and has become a key bottleneck affecting medical efficiency and patient satisfaction.

[0003] On the other hand, pressure injury (pressure sore) and subcutaneous congestion are common complications of patients who are in bed for a long time, postoperative immobilization or anticoagulant therapy. At present, nurses often rely on regular visual or palpation examination for risk assessment, but early pressure injury only shows local congestion or occult ischemia, which is difficult to identify by naked eye. Subcutaneous congestion often has a smaller visible range on the surface than the actual damage area, resulting in the difficulty of early detection and inaccurate range assessment. Although traditional ultrasound can assist in judgment, it relies on professional operators and is difficult to continuously monitor at the bedside due to its large size.

[0004] Photoacoustic imaging (PAI) uses the physical mechanism of "light excitation and sound detection", which can simultaneously obtain multi-parameter images such as hemoglobin concentration, oxygen saturation and lipid content, and has the advantages of high optical contrast and deep acoustic penetration. In recent years, multi-spectral photoacoustic imaging (MSPAI) has realized the quantitative analysis of blood vessel network, blood oxygen distribution and tissue composition by time-sharing or parallel projection of multi-wavelength pulsed laser, which can theoretically provide real-time three-dimensional positioning of blood vessels for venipuncture and early functional warning for pressure injury and subcutaneous congestion. However, the existing photoacoustic systems are mostly table or cart types, which are large in size, complex in fiber, high in cost, and difficult to meet the urgent needs of clinical "handheld, bedside and instant".

[0005] The literature (Liu Qiang, Jin Tian, ​​Chen Qian, Xi Lei. Research Progress of Miniaturized Photoacoustic Imaging Technology in Biomedical Field[J]. Chinese Journal of Lasers, 2020, 47(2): 0207019.) discloses a handheld multispectral photoacoustic imaging technology that can be used clinically for vascular imaging. It can distinguish blood vessels with a diameter as small as 100 micrometers and a depth of less than 1 centimeter, and provide hemoglobin oxygen saturation and pulsation information. However, this technology can only achieve static imaging and cannot assess tissue damage. Therefore, the relevant technology needs further improvement. Summary of the Invention

[0006] In view of the problems existing in the prior art, the purpose of this invention is to provide a handheld photoacoustic multimodal vascular puncture and tissue damage diagnosis system and device.

[0007] This invention provides a photoacoustic guided puncture system, comprising:

[0008] The laser excitation module is configured to emit 532nm laser pulses to enhance the blood vessel boundary;

[0009] The photoacoustic detection module is configured to receive photoacoustic signals generated by biological tissue and reconstruct enhanced images of blood vessel boundaries accordingly.

[0010] The image processing module is configured to acquire blood vessel contour information and current puncture depth based on the enhanced blood vessel boundary image;

[0011] The elastic decision module is configured to acquire tissue elasticity information in real time, combine the vascular contour information output by the image processing module with the current puncture depth and vascular stability index, and output elastic feedback decisions under the dynamic elastic navigation model and virtual force feedback model, including the optimal puncture angle and vibration intensity.

[0012] The human-computer interaction module is configured to display the enhanced image of the blood vessel boundary, the optimal puncture angle, and the vibration intensity in real time, and to perform the blood vessel puncture operation.

[0013] Furthermore, the optimal puncture angle is calculated as follows:

[0014]

[0015] wherein, represents the dynamic elastic modulus of the blood vessel wall, , P represents the blood pressure of the blood vessel, r represents the radius of the blood vessel, h represents the wall thickness, represents the viscosity coefficient: 0.12 Pa·s; represents the blood vessel depth profile.

[0016] Further, the elasticity feedback is divided into level 1, level 2 and level 3;

[0017] When the vibration intensity is greater than 80, output level 3; when the vibration intensity is greater than 50, output level 2; when the vibration intensity is less than 50, output level 1;

[0018] When the current puncture depth < 2.0 mm, the vibration intensity is calculated as follows:

[0019]

[0020] When the current puncture depth ≥ 2.0 mm, the vibration intensity is calculated as follows:

[0021]

[0022] wherein, , represents the current puncture depth, represents the probe contact pressure; represents the dynamic elastic modulus of the blood vessel wall, , P represents the blood pressure of the blood vessel, r represents the radius of the blood vessel, h represents the wall thickness, represents the viscosity coefficient: 0.12 Pa·s.

[0023] Further, the blood vessel stability index is calculated as follows:

[0024]

[0025] wherein, represents the blood vessel profile curvature entropy, , represents the normalized curvature probability distribution; represents the photoacoustic signal signal-to-noise ratio, , represents the wavelet packet denoising signal energy, represents the participating noise energy; represents the attenuation factor, 0.05; denotes the dynamic elastic modulus of the blood vessel wall, , P denotes the blood pressure of the blood vessel, r denotes the radius of the blood vessel, h denotes the wall thickness, denotes the viscosity coefficient: 0.12 Pa·s;

[0026] When VSI>0.35, limit the puncture angle ≤30°.

[0027] The present application also provides a pressure injury diagnosis system, comprising:

[0028] A laser excitation module configured to emit 850nm laser pulses for collagen imaging;

[0029] A photoacoustic detection module configured to receive photoacoustic signals generated by biological tissues and reconstruct collagen imaging images based thereon;

[0030] An image processing module configured to calculate a hemoglobin / collagen concentration ratio R based on the collagen imaging images and generate a damage area heat map;

[0031] An elastic decision module configured to acquire tissue elasticity information in real time, combine the hemoglobin / collagen concentration ratio R and the damage area heat map output by the image processing module with a tissue damage diffusion coefficient and a pressure injury index, and output an elastic feedback decision;

[0032] A display and reporting module configured to mark damage diffusion risk areas according to the tissue damage diffusion coefficient and the pressure injury index, prompt different color warnings, and automatically generate a diagnosis report.

[0033] Further, the calculation of the tissue damage diffusion coefficient is as follows:

[0034]

[0035] wherein, denotes the Euclidean norm of the collagen concentration gradient, ;

[0036] denotes the hemoglobin concentration change rate, , denotes the time step, ;

[0037] When DDI>2.5, mark as a high-risk area of damage diffusion.

[0038] Further, the calculation of the pressure injury index is as follows:

[0039] PDI = R x (tissue stiffness / 10)

[0040] wherein R represents the hemoglobin / collagen concentration ratio: , represents the average concentration of hemoglobin in the selected area; represents the average concentration of collagen in the same area; , represents the ex vivo pig tissue calibration coefficient, ;

[0041] When PDI>0.8, it indicates high risk, and red color indicates early warning;

[0042] When 0.4≤PDI≤0.8, it indicates medium risk, and yellow color indicates early warning;

[0043] When PDI<0.4, it indicates low risk, and there is no early warning.

[0044] The application also provides a handheld photoacoustic blood vessel and tissue diagnosis device, comprising: a handle type host A, wherein a replaceable probe B, a laser control module E, an ultrasonic receiving array G, a pressure sensor I and AR display glasses K are installed on the handle type host A; the handle type host A is integrated with a photoacoustic blood vessel and tissue diagnosis system;

[0045] The replaceable probe B is a conical puncture probe C and / or a flat head scanning probe D;

[0046] The pressure sensor I is provided with an elastic feedback unit J, and the elastic feedback unit J is integrated with the photoacoustic guiding puncture system and the pressure injury diagnosis system.

[0047] Further, the laser control module E is provided with a dual-wavelength laser source F, and the wavelength of the dual-wavelength laser source F is 532 nm and 850 nm.

[0048] Further, the ultrasonic receiving array G is provided with a 128-element CMUT array, and the 128-element CMUT array is arranged in a ring shape on the probe base.

[0049] Further, the AR display glasses K are provided with a Wi-Fi 6 wireless transmission module, and the AR display glasses K project the blood vessel direction / injury area to the operator's field of view.

[0050] The application has the following beneficial effects:

[0051] (1) The handheld multispectral photoacoustic imaging device provided by the application increases the puncture success rate by 30% through real-time blood vessel imaging combined with elastic force feedback.

[0052] (2) The device realizes 72h early warning of pressure injury through hemoglobin / collagen double-index imaging algorithm;

[0053] (3) The device realizes automatic calculation of 3D hematoma volume, so that the hematoma volume measurement error is less than 5%, and the risk of multiple punctures and missed diagnosis is significantly reduced;

[0054] (4) Compared with the literature (Liu Qiang, Jin Tian, Chen Qian, Xi Lei. Research Progress of Miniaturized Photoacoustic Imaging Technology in Biomedical Field[J]. Chinese Journal of Lasers, 2020, 47(2): 0207019.), the application initiates dynamic stability evaluation of vascular stability index (VSI), and constructs a tissue damage diffusion coefficient (DDI) system, providing a new quantitative evaluation and real-time navigation integrated solution for vascular puncture and tissue damage diagnosis.

[0055] Obviously, according to the above content of the application, according to the ordinary technical knowledge and conventional means in the art, other various forms of modification, replacement or change can be made without departing from the above technical idea of the application.

[0056] The above content of the application will be further described in detail through the specific implementation mode of the embodiment. However, this should not be understood as the scope of the above subject matter of the application being limited to the following embodiment. Any technology realized based on the above content of the application belongs to the scope of the application. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 The workflow diagram of the handheld multispectral photoacoustic imaging device of the application. DETAILED DESCRIPTION

[0058] In order to better understand and implement, the technical solutions in the embodiments of the application will be clearly and completely described below in combination with the drawings in the embodiments of the application.

[0059] It should be particularly pointed out that the algorithms of data acquisition, transmission, storage and processing steps not specifically described in the embodiments, and the hardware structure, circuit connection and the like not specifically described can be realized through the existing technology disclosed.

[0060] Embodiment 1, handheld multi-spectral photoacoustic imaging device of the present application

[0061] I. Overview of the structure of the handheld multi-spectral photoacoustic imaging device

[0062] The main body of the handheld multi-spectral photoacoustic imaging device is a handle-type main machine A;

[0063] The handle-type main machine A is equipped with a replaceable probe B. When the blood vessel puncture mode is selected, the replaceable probe B is a conical puncture probe C; when the tissue scanning mode is selected, the replaceable probe B is a flat-head scanning probe D;

[0064] The handle-type main machine A is also equipped with a laser control module E, and the laser control module E is equipped with a dual-wavelength laser source F, wherein 532 nm is used for blood vessel imaging and 850 nm is used for collagen detection;

[0065] The handle-type main machine A is also equipped with an ultrasonic receiving array G, and the ultrasonic receiving array G is equipped with a 128-element CMUT array (ring layout on the probe base);

[0066] The handle-type main machine A is also equipped with a pressure sensor I, and the pressure sensor I is equipped with an elastic feedback unit J, which can calculate the hardness of the tissue in real time and output tactile feedback;

[0067] The handle-type main machine A is also equipped with AR display glasses K (Wi-Fi 6 wireless transmission module), which can project the blood vessel direction / damage area to the operator's field of view.

[0068] The core components are summarized as follows:

[0069] 1. Dual-mode probe

[0070] (1) Conical puncture probe C:

[0071] The tip is integrated with a 532 nm laser outlet + ultrasonic receiver (puncture angle ≤ 30°);

[0072] Laser outlet diameter: 200 μm, divergence angle < 5°;

[0073] Ultrasonic receiver: 64-channel CMUT, center frequency 10 MHz;

[0074] Puncture limiter: mechanical constraint to ensure angle ≤ 30°.

[0075] (2) Flat-head scanning probe D:

[0076] 5x5 cm rectangular array (850 nm laser source), supporting sliding scanning;

[0077] Laser array: 8x8 VCSEL dot array (850 nm);

[0078] Ultrasound array: 128-element CMUT (pitch 0.5mm);

[0079] Sliding encoder: displacement accuracy ±0.1mm. 2, pressure-elastic feedback system (pressure sensor)

[0080] Located on the probe contact surface, real-time measurement of probe contact pressure (range 0-10N);

[0081] Pressure sampling rate: 1kHz;

[0082] Tissue stiffness algorithm: tissue stiffness = (pressure value / vascular deformation rate) x k;

[0083] Tactile feedback: piezoelectric ceramic vibrator (0-100 level intensity adjustable).

[0084] 3、AR display system

[0085] Vascular navigation: real-time projection of 3D vascular model (blue) + needle tunnel (green);

[0086] Damage warning: DDI>2.5 area red flashing + PDI numerical value superimposed display.

[0087] II、Core algorithm

[0088] 1、Vascular puncture navigation algorithm

[0089] (1) Dynamic elastic navigation model

[0090] The specific structure of the dynamic elastic navigation model in the vascular puncture navigation algorithm of the application is as follows:

[0091] The input module is configured to extract the blood vessel contour (Canny edge detection) through the photoacoustic original signal.

[0092] The data processing module is configured to import the extracted blood vessel contour and other information into the deformation differential equation, and obtain the optimal puncture angle through the path optimization algorithm .

[0093] The application is based on the Fung quasi-linear viscoelasticity theory of biomechanics:

[0094]

[0095] Simplify it into a real-time calculation model to calculate the dynamic elastic modulus of the blood vessel wall (kPa), :

[0096]

[0097]

[0098] in, P Indicates vascular blood pressure (estimated using photoacoustic pulsation signals); r : radius of blood vessel; h Wall thickness; This is the viscosity coefficient (human blood plasma viscosity), with a default value of 0.12 Pa·s.

[0099] The objective function of the path optimization algorithm is: ;

[0100] Optimal puncture angle The calculation is as follows: , This is a map showing the depth distribution of blood vessels.

[0101] The output module is configured to output the optimal puncture angle. .

[0102] The clinical validation results using the above dynamic elastic navigation model are as follows:

[0103] In 120 patients with a BMI > 30, the puncture success rate was 98% (compared to 68% in the control group).

[0104] The rate of vascular collapse decreased from 22% to 3% (p<0.01). p <0.01).

[0105] (2) Virtual force feedback mechanism

[0106] Biomechanical resistance model (based on Hertzian contact theory):

[0107]

[0108] in, The dynamic elastic modulus of the blood vessel wall (kPa) is derived from the aforementioned dynamic elastic navigation model. Current puncture depth (mm); : Probe contact pressure (N).

[0109] Vibration intensity ( The hierarchical mapping (levels 0-100) is as follows:

[0110] Superficial tissues ( <2.0 mm): ;

[0111] Deep tissues ( ≥2.0 mm): .

[0112] Based on the above vibration intensity ( ), output haptic feedback (1 / 2 / 3 level). The specific haptic feedback rules are as follows:

[0113] When greater than 80, output level 3, high-frequency continuous vibration (danger warning);

[0114] When greater than 50, output level 2, medium-frequency intermittent vibration (warning);

[0115] When less than 50, output level 1, low-frequency micro-vibration (safety guide).

[0116] The above formula is derived based on Hertzian contact model (Hertzian Contact Model), reference: Johnson KL. Contact Mechanics. Cambridge University Press, 1987.

[0117] The clinical verification results of the above virtual force feedback mechanism are as follows:

[0118] In 50 cases of patient testing, the feedback vibration intensity and the doctor's hand feeling coincidence rate is 92% (Kappa=0.85).

[0119] 2. New diagnostic parameter design

[0120] The VSI, DDI and other parameters described in the application are derived based on the first principle of biomechanics, and the mathematical definition and algorithm flow have been completely disclosed, and the clinical effect can be verified by numerical simulation.

[0121] (1) Vessel stability index (VSI)

[0122]

[0123] Parameter definition:

[0124] Vessel profile curvature entropy (VCE) ): Wherein, represents the normalized curvature probability distribution (calculated by KDE kernel density estimation);

[0125] Photoacoustic signal signal-to-noise ratio (dB): Wherein, represents the energy of the wavelet packet denoising signal (WPT-L9); represents the energy of the noise involved;

[0126] Vessel wall dynamic elastic modulus (E) ) as described in the "vessel puncture navigation algorithm";

[0127] Attenuation factor (a) ) is 0.05 (determined by COMSOL multi-parameter optimization);

[0128] denotes the natural exponential function.

[0129] When VSI> 0.35, limit the puncture angle ≤ 30°.

[0130] (2) Tissue damage diffusion coefficient (DDI)

[0131]

[0132] Parameter definition:

[0133] Euclidean norm of collagen concentration gradient (||C||E, mg / mL / mm): , calculated by Sobel operator (3x3 convolution kernel);

[0134] Rate of change of hemoglobin concentration (||H||t, mg / mL / s): , time step ( meet CFL condition: ).

[0135] When DDI> 2.5, mark as high-risk area of damage diffusion.

[0136] 3. Data processing flow

[0137] (1) Puncture navigation whole process

[0138] S1, photoacoustic acquisition→signal enhancement: 532nm laser excitation + wavelet packet denoising (WPT-L9);

[0139] S2, signal enhancement→three-dimensional reconstruction: compressed sensing algorithm (ADMM optimizer solves min‖Ax-b‖2+λ‖x‖1);

[0140] S3, three-dimensional reconstruction→parameter calculation: real-time parallel computing:

[0141] Thread 1: ;

[0142] Thread 2: ;

[0143] S4, parameter calculation→path planning: variational method optimization ;

[0144] S5, path planning→AR display: projection dynamic needle insertion tunnel (green dashed line);

[0145] ​​S6, AR display -> haptic feedback: vibration level = force_feedback( , depth, pressure).

[0146] Technical packages: Real-time parallel architecture: GPU acceleration (NVIDIA Jetson Orin) makes the computation delay < 50ms;

[0147] Compressive sensing reconstruction: The sampling rate is reduced to 40% (original data volume) and still maintains PSNR > 35dB;

[0148] Path optimization: output the optimal puncture angle through the Hamiltonian operator (as described in "Vascular Puncture Navigation Algorithm").

[0149] (2) Four-order analysis of injury diagnosis

[0150] 1) Step 1: Multispectral decomposition

[0151] Input: The measured reflected light intensity at two known wavelengths (532nm, 850nm): signal_532, signal_850;

[0152] According to the Beer-Lambert law, determine the absorption coefficient matrix A at these two wavelengths. The above two groups of intensity mainly contain the absorption contribution of three substances: Hb (deoxyhemoglobin), HbO2 (oxyhemoglobin) and collagen (collagen). By least squares, solve the linear equations to calculate the unit extinction coefficient of three substances at each wavelength:

[0153] A = np.array([[220, 80, 5], # 532nm: Hb, HbO2, Collagen

[0154] [35, 40, 185]]) # 850nm: Hb, HbO2, Collagen

[0155]

[0156] Then, use the least squares method to inversely solve the mixed photoacoustic signal measured at two wavelengths into the concentration vector of three substances:

[0157] concentrations = np.linalg.lstsq(A, [signal_532, signal_850], rcond=None)[0]

[0158] return concentrations # [C_Hb, C_HbO2, C_Collagen]

[0159] 2) Step 2: Spatiotemporal Gradient Calculation

[0160] Sobel Gradient: grad_C = cv2.Sobel(collagen_map, cv2.CV_64F, 1, 1, ksize=3)

[0161] Central Difference (Δt=0.1s): dH_dt = (Hb_map[t+1] – Hb_map[t–1]) / 0.2

[0162] 3) Step 3: DDI Synthesis

[0163] DDI_map = np.linalg.norm(grad_C, axis=2) * np.sqrt(np.abs(dH_dt))

[0164] 4) Step 4: Risk Stratification

[0165] risk_mask = (DDI_map>2.5) | (VSI_map<0.2).

[0166] 2, Pressure Injury Index (PDI)

[0167] PDI = R × (Tissue Hardness / 10)

[0168] Where R represents the hemoglobin / collagen concentration ratio.

[0169] Clinical Significance: R>1 indicates hemoglobin enrichment (acute injury), R<0.3 indicates collagen dominance (healthy tissue).

[0170] The specific calculation of R is as follows:

[0171]

[0172] Parameter Definition:

[0173] C_Hb represents the average concentration of hemoglobin (mg / mL) in the selected area, calculated by spectral decomposition of the 532nm laser excitation signal. C_Collagen represents the average concentration of collagen (mg / mL) in the same area, calculated by inversion of the 850nm laser excitation signal.

[0174] Tissue Hardness (kPa):

[0175] wherein, The strain rate is obtained from the photoacoustic sequence deformation analysis.

[0176] Diagnostic threshold:

[0177] PDI>0.8→High risk (AR red flashing warning);

[0178] 0.4≤PDI≤0.8→Medium risk (yellow prompt);

[0179] PDI<0.4→Low risk (no warning).

[0180] Example 2, the handheld multi-spectral photoacoustic imaging device of the application is used for blood vessel puncture

[0181] The working process of blood vessel puncture using the handheld multi-spectral photoacoustic imaging device of Example 1 includes the following steps: Figure 1 ):

[0182] S1, start scanning: system initialization, enter working state.

[0183] S2, mode selection: the user switches the blood vessel puncture mode through the interface:

[0184] In this mode, the conical puncture probe contacts the skin, and is excited by 532nm laser for blood vessel boundary enhancement; the photoacoustic signal generated by the biological tissue is received, and the blood vessel boundary enhancement image is reconstructed accordingly;

[0185] S3, based on the blood vessel boundary enhancement image, the blood vessel contour information and the current puncture depth are obtained;

[0186] S4, elastic feedback decision:

[0187] For real-time acquisition of tissue elasticity information, the blood vessel contour information and the current puncture depth output by the image processing module are combined with the blood vessel stability index, and the elastic feedback decision is output under the dynamic elastic navigation model and the virtual force feedback model, including the optimal puncture angle and vibration intensity;

[0188] S5, intelligent decision output:

[0189] Output the optimal puncture angle and vibration intensity: when the probe pressure>5N, trigger the vibration alarm.

[0190] Example 3, the handheld multi-spectral photoacoustic imaging device of the application is used for tissue damage assessment

[0191] The working process of tissue damage assessment using the handheld multi-spectral photoacoustic imaging device of Example 1 includes the following stepsFigure 1

[0192] S1, start scanning: system initialization, enter working state.

[0193] S2, mode selection: the user switches the organization damage assessment mode through the interface:

[0194] In this mode, the flat head scanning probe is used for sliding scanning, and 850nm laser excitation is used for collagen imaging, so as to obtain collagen distribution and generate a damage area thermal map;

[0195] S3, dual-wavelength fusion calculation of hemoglobin / collagen concentration ratio R;

[0196] S4, elastic feedback decision:

[0197] The elastic feedback decision is used for real-time acquisition of tissue elasticity information, and the hemoglobin / collagen concentration ratio R and the damage area thermal map output by the image processing module are combined with the tissue damage diffusion coefficient and the pressure injury index to output the elastic feedback decision;

[0198] S5, intelligent decision output:

[0199] According to the tissue damage diffusion coefficient and the pressure injury index, the damage diffusion risk area is marked, and different color warnings are prompted: when PDI>0.8, the AR glasses mark a red warning area.

[0200] The beneficial effects of the present application are proved by the following experimental examples.

[0201] Experimental Example 1, application of the handheld multispectral photoacoustic imaging device of the present application

[0202] Scene 1: venipuncture

[0203] The nurse wears AR glasses, and the conical puncture probe is positioned on the blood vessel -> the glasses display the 3D direction of the blood vessel (blue);

[0204] Tactile feedback: the handle vibrates slightly when the probe contact pressure reaches 3N -> the needle is inserted at θ=60° -> the first puncture success rate is 98%.

[0205] Scene 2: pressure sore warning

[0206] The flat head scanning probe scans the sacrococcygeal part -> PDI=0.92 -> the AR glasses highlight the red area;

[0207] Nursing suggestion: turn over every 2 hours + use a pressure relief pad -> PDI decreases to 0.3 after 3 days.

[0208] ​In summary, the application provides a handheld photoacoustic multi-modal blood vessel puncture and tissue damage diagnosis system and device. The handheld multi-spectral photoacoustic imaging device integrates elastic force feedback and real-time blood vessel imaging, so that the puncture success rate is increased by 30%. Hemoglobin / collagen double-index imaging can early warn pressure injury by 72 hours. The 3D hematoma volume automatic calculation error is less than 5%, which significantly reduces repeated puncture and missed diagnosis. The application provides a quantitative evaluation and real-time navigation integrated solution for blood vessel puncture and tissue damage, which is safe and efficient.

Claims

1. A photoacoustic guided puncture system, characterized by, Comprising: a laser excitation module configured to emit 532nm laser pulses to enhance blood vessel boundaries; a photoacoustic detection module configured to receive photoacoustic signals generated by biological tissues and reconstruct blood vessel boundary enhanced images therefrom; an image processing module configured to obtain blood vessel contour information and current puncture depth based on the blood vessel boundary enhanced images; an elastic decision module configured to obtain tissue elasticity information in real time, combine the blood vessel contour information and the current puncture depth output by the image processing module with a blood vessel stability index, and output elastic feedback decisions including optimal puncture angles and vibration intensities under a dynamic elastic navigation model and a virtual force feedback model; a human-computer interaction module configured to display the blood vessel boundary enhanced images, the optimal puncture angles and the vibration intensities in real time, and perform blood vessel puncture operations; The calculation of the optimal puncture angle is as follows: wherein, represents the dynamic elastic modulus of the blood vessel wall, , P represents the blood pressure of the blood vessel, r represents the radius of the blood vessel, h represents the wall thickness, represents the viscosity coefficient: 0.12 Pa-s; represents the blood vessel depth profile.

2. The system of claim 1, wherein, The elastic feedback is divided into levels 1, 2 and 3; When the vibration intensity is greater than 80, level 3 is output; when the vibration intensity is greater than 50, level 2 is output; and when the vibration intensity is less than 50, level 1 is output; When the current puncture depth is less than 2.0 mm, the vibration intensity is calculated as follows: When the current puncture depth is greater than or equal to 2.0 mm, the vibration intensity is calculated as follows: wherein, , represents the current puncture depth, represents the probe contact pressure; represents the dynamic elastic modulus of the vessel wall, , P represents the blood pressure in the vessel, r represents the vessel radius, h represents the wall thickness, represents the viscosity coefficient: 0.12 Pa-s.

3. A hand-held photoacoustic blood vessel and tissue diagnostic device, characterized by comprising: Comprising: a handle-type host A, which is provided with a replaceable probe B, a laser control module E, an ultrasonic receiving array G, a pressure sensor I and AR display glasses K; the handle-type host A is integrated with a photoacoustic blood vessel and tissue diagnosis system; The replaceable probe B is a conical puncture probe C and / or a flat head scanning probe D; The pressure sensor I is provided with an elastic feedback unit J, and the elastic feedback unit J is integrated with the photoacoustic guided puncture system and the pressure injury diagnosis system according to any one of claims 1-2.

4. The apparatus of claim 3, wherein, The laser control module E is provided with a dual-wavelength laser source F, and the wavelengths on the dual-wavelength laser source F are 532nm and 850nm.

5. The apparatus of claim 3, wherein, The ultrasonic receiving array G is provided with a 128-element CMUT array, which is arranged in a ring shape on a probe base.

6. The apparatus of claim 3, wherein, The AR display glasses K are provided with a Wi-Fi 6 wireless transmission module, and the AR display glasses K project the blood vessel direction / injury area to the operator's field of view.

7. The apparatus of claim 3, wherein, The pressure injury diagnosis system comprises: a laser excitation module configured to emit 850nm laser pulses for collagen imaging; a photoacoustic detection module configured to receive photoacoustic signals generated by biological tissues and reconstruct collagen imaging images therefrom; an image processing module configured to calculate a hemoglobin / collagen concentration ratio R based on the collagen imaging images and generate an injury area heat map; an elastic decision module configured to obtain tissue elasticity information in real time, combine the hemoglobin / collagen concentration ratio R and the injury area heat map output by the image processing module with a tissue injury diffusion coefficient and a pressure injury index, and output an elastic feedback decision. The display and report module is configured to mark the injury diffusion risk area according to the tissue injury diffusion coefficient and the pressure injury index, prompt different color warnings, and automatically generate a diagnosis report.

8. The apparatus of claim 7, wherein, The calculation of the tissue injury diffusion coefficient is as follows: wherein denotes the Euclidean norm of the collagen concentration gradient, ; denotes the rate of change of hemoglobin concentration, , denotes the time step, ; When DDI>2.5, it is marked as a high-risk area of injury diffusion.

9. The apparatus of claim 7, wherein, The calculation of the pressure injury index is as follows: PDI=R×(tissue hardness / 10) wherein R represents the hemoglobin / collagen concentration ratio: , represents the average hemoglobin concentration in the selected area; represents the average collagen concentration in the same area; , represents the calibration factor for the ex vivo porcine tissue, ; When PDI>0.8, a high risk is prompted, and a red warning is prompted; When 0.4≤PDI≤0.8, a medium risk is prompted, and a yellow warning is prompted; When PDI<0.4, a low risk is prompted, and no warning is prompted.

Images