77 results about "Small blood vessel" patented technology

The invention discloses a retinal blood vessel image segmentation method based on dynamic weight distribution and index guide optimization. Comprising the steps that a to-be-segmented retinal blood vessel image is input into an LM-TransUNet model, an image segmentation visualization result is obtained, the LM-TransUNet model is obtained based on training of a training set, and the training set comprises the retinal blood vessel image and a corresponding true value; the process of training the LM-TransUNet model based on the training set comprises the following steps: predicting a retinal blood vessel image through an OfficientNet and ResNet hybrid encoder model, calculating segmentation loss and performance index loss according to a prediction result and a truth value, adaptively fusing multi-scale features by using a dynamic weight distribution mechanism, and adjusting network parameters in real time through an index guide optimization module, so as to achieve the purpose of improving the performance of the retinal blood vessel image. And a new prediction result is obtained for iteration until convergence, and a final model is obtained. According to the method, the accuracy of retinal blood vessel image segmentation can be remarkably improved, and particularly, the method is excellent in segmentation of complex blood vessel structures and small blood vessels.

The invention relates to the technical field of image modeling, in particular to a three-dimensional modeling method and system for medical image data, and the method comprises the steps: carrying out the preprocessing of the obtained two-dimensional or three-dimensional medical image data, extracting a blood vessel region based on the image texture and gray features, and generating a preliminary blood vessel structure segmentation image; in combination with a blood flow simulation or real acquisition means, obtaining dynamic feature data from the vascular structure segmentation map, and mapping the dynamic feature data to a corresponding vascular structure position to form a vascular distribution map with blood flow constraint; and inputting the blood vessel distribution diagram into a three-dimensional modeling module, expanding and complementing small blood vessel branches by utilizing blood flow constraint, dynamically correcting the form of a main blood vessel, and generating a three-dimensional blood vessel structure model with real fluid physiological characteristics. The method can be widely applied to medical research and clinical aid decision-making scenes such as preoperative simulation, individualized intervention path planning and blood flow reconstruction analysis, and has good popularization and application prospects and engineering practical value.

The invention belongs to the technical field of biological medicine, and particularly relates to application of corynebacterium casei in preparation of medicine for preventing and treating lung injury. By constructing a radon inhalation-induced lung injury mouse model, after exposure at different radon levels, the weight of a mouse is reduced, the lung coefficient is increased, the tidal volume of the lung, the ventilation volume per minute and the maximum inspiration flow are reduced, the respiratory function is inhibited, the alveolar cavity of the mouse is abnormally expanded, the pulmonary septum is thickened and fractured, and more fibrin exudation is accompanied; the traditional Chinese medicine composition can be used for effectively inhibiting pulmonary fibrosis and lung injury due to obvious inflammatory infiltration beside bronchial tubes, extravasated blood in small blood vessels and lung tissue collagen deposition, and can be used for gavage of corynebacterium casei to effectively inhibit the above phenomena. Therefore, the corynebacterium casei can be used for preventing and treating lung injury, especially lung injury caused by radon exposure.

The invention provides a covered stent. The covered stent comprises a cutting stent and a covering film partially covering the cutting stent. The cutting support comprises a plurality of wave rings which are sequentially connected in the axial direction of the wave rings, and each wave ring comprises a plurality of foundation structures which are sequentially connected in the circumferential direction. The plurality of wave rings comprise z-shaped wave rings and double-n-shaped wave rings, and the z-shaped wave rings and the double-n-shaped wave rings are alternately arranged in the axial direction of the cutting support; the basic structure in the z-shaped wave ring is a first basic structure; the basic structure in the double-n-shaped wave ring is a second basic structure, and the second basic structure is a closed grid and is formed by butt joint of two n-shaped wave rods in the circumferential direction in a face-to-face mode. The covered stent has enough supporting performance, flexibility and adherence, internal leakage can be prevented, the thickness of the stent can be well controlled, and the covered stent can be conveyed and enter smaller blood vessels.

Some embodiments of a rotational atherectomy device can remove (partially or completely) stenotic lesions in blood vessels by rotating one or more abrasive elements in an orbital path to abrade and breakdown the lesion. In particular implementations, multiple abrasive elements are arranged along a distal portion of a drive shaft with an improved configuration so as to facilitate both efficient navigation into smaller blood vessels below the ankle or in the heart and effective orbital paths for abrading stenotic material in such smaller vessels.

Some embodiments of a rotational atherectomy device can remove (partially or completely) stenotic lesions in blood vessels by rotating one or more abrasive elements in an orbital path to abrade and breakdown the lesion. In particular implementations, multiple abrasive elements are arranged along a distal portion of a drive shaft with an improved configuration so as to facilitate both efficient navigation into smaller blood vessels below the ankle or in the heart and effective orbital paths for abrading stenotic material in such smaller vessels.

The invention relates to the technical field of blood vessel image segmentation, and provides a fundus blood vessel image accurate segmentation method based on a U-Net model, aiming at solving the problem that the existing blood vessel segmentation method based on deep learning is difficult to effectively balance global context and local details. In order to solve the problem of inaccurate blood vessel structure segmentation caused by factors such as poor image quality and low contrast ratio, a method of combining a wavelet and a visual state space model and an adaptive dynamic fusion module is provided to improve the integrity of a slender blood vessel and the structural segmentation accuracy of a complex blood vessel network in a fundus blood vessel image. By integrating multiple efficient feature extraction mechanisms, accurate recognition of small blood vessels and the whole blood vessel structure is achieved, the accuracy and robustness of blood vessel segmentation in complex medical images are remarkably improved, and the method can be widely applied to analysis of eye blood vessels such as retinopathy and hypertensive fundus lesion and research of biomarkers related to the blood vessels.

Some embodiments of a rotational atherectomy device can remove (partially or completely) stenotic lesions in blood vessels by rotating one or more abrasive elements in an orbital path to abrade and breakdown the lesion. In particular implementations, multiple abrasive elements are arranged along a distal portion of a drive shaft with an improved configuration so as to facilitate both efficient navigation into smaller blood vessels below the ankle or in the heart and effective orbital paths for abrading stenotic material in such smaller vessels.

The invention belongs to the field of medical devices, and particularly relates to a soft robot thrombus removing device which comprises a robot body, a minimally invasive thrombus taking head, an external magnetic field control device and an imaging display device. The minimally invasive thrombectomy head comprises a front section thrombectomy part, a middle section positioning part and a tail section control part; a plurality of through hole structures are arranged in the thrombus extraction component; a rotating structure, a suction component and a clamping structure are respectively arranged in a plurality of through hole structures arranged in the thrombus extraction component; a magnetic navigation sensor is arranged in a middle section positioning component of the minimally invasive thrombus extraction head; a control unit electrically connected with the rotating structure, the suction part and the clamping structure is arranged in the control part at the tail section of the minimally invasive thrombus extraction head; the robot body is transported to the position close to the diseased region through the catheter and the puncture needle sheath. The device has the advantages that the device is specially designed for treating vascular embolism, is particularly suitable for removing thrombus of small blood vessels such as cerebral vessels, and has extremely high flexibility and operation precision.

The invention discloses a multi-mode whole-body blood vessel magnetic resonance imaging method and device, and the method comprises the steps: displaying prompt information for carrying out the cascade wrapping of a whole-body magnetic resonance coil and the input of a contrast agent for a target object, and enabling the contrast agent to be polysaccharide superparamagnetic iron oxide; setting a signal mark point on the constructed whole body positioning image and configuring a multi-modal imaging sequence; emitting a magnetic resonance radio frequency pulse, and receiving a coil electric signal transmitted by the whole body magnetic resonance coil; determining the signal intensity of each set signal mark point according to the coil electric signal; when the signal intensity is larger than a preset threshold value, whole-body blood vessel scanning is conducted on the target object through the multi-mode imaging sequence; and performing image reconstruction according to the scanned target image sequence to generate a whole body blood vessel image. By adopting the embodiment of the invention, one-stop imaging of large and small blood vessels of the whole body can be realized at one time by utilizing a scanning mode of cascading wrapping of the whole body magnetic resonance coil, so that the scanning time is greatly shortened, and the blood vessel imaging efficiency is improved.

A personalized whole-brain small artery angiogenesis method based on multiple constraints, belonging to the field of mathematical modeling algorithm development, includes the following steps: constructing a personalized three-dimensional model based on real head and carotid CTA images to determine the starting position of blood vessel growth; recording the geometric parameters (such as outlet area, outlet blood vessel length, etc.) of each blood vessel model based on the personalized model; using the Brainstorm standard head model three-dimensional point cloud file, uniformly sampling it as the ending position of blood vessel growth; based on multiple constraints, using the simulated annealing algorithm to optimize the geometric parameters of small blood vessels and calculate the spatial positions of bifurcations of blood vessels at all levels after meeting the constraints. The personalized whole-brain small artery angiogenesis real model of the present invention improves the accuracy for numerical simulation and hemodynamic calculation.

The present invention belongs to the field of biomedicine technology, and in particular to the application of Corynebacterium casei in the preparation of medicines for preventing and treating lung injury. The present invention constructs a mouse model of lung injury caused by radon inhalation and finds that after exposure to different radon levels, the mice lose weight, increase the lung coefficient, reduce the lung tidal volume, minute ventilation and maximum inspiratory flow, and their respiratory function is inhibited. The alveolar cavity of the mice is abnormally expanded, the lung septum is thickened and fractured, accompanied by more fibrin exudation, obvious inflammatory infiltration appears near the bronchus, congestion occurs in small blood vessels, and collagen deposition in lung tissue can lead to the occurrence of pulmonary fibrosis, causing lung injury. Oral administration of Corynebacterium casei can effectively inhibit the above-mentioned phenomenon. Therefore, Corynebacterium casei can be used for the prevention and treatment of lung injury, especially lung injury caused by radon exposure.

The present application belongs to the field of medical image segmentation, and particularly relates to a retinal blood vessel image segmentation method. In view of the problems of low segmentation accuracy, insufficient segmentation ability of small blood vessels at the edge of eyeball, fracture at the blood vessel branch, and excessive interference of image noise in the existing retinal blood vessel image segmentation, the method comprises the steps of retinal image preprocessing and establishment of a retinal blood vessel segmentation model, wherein the preprocessing comprises converting a color retinal image into a gray image by giving different weights to the RGB three channels of the color retinal image; using a normalized and contrast-limited adaptive histogram equalization method to improve the image; using a local adaptive gamma change algorithm to adjust the retinal image; using translation, rotation, and noise increase to expand the data set; and the model establishment comprises feature extraction, feature fusion, and retinal blood vessel image segmentation.

Devices, systems, and methods for treatment of the subcutaneous tissues of various types. Non-invasive ultrasound energy is used for planning, monitoring and treating the adipose tissue, small blood vessels, and benign tumors of the subcutaneous tissue. While high intensity focal ultrasound (HIFU) has been used to treat many areas of the body including cancer lesions typically using ablation / high heat energy, lower intensity volumetric ultrasound (LIVU) techniques are used to denature the subcutaneous layer. Various safety features may be included to further enhance the treatment of the subcutaneous tissue using LIVU techniques.

The invention discloses a cerebral cortex penetrator small blood vessel magnetic resonance imaging method based on an inflow enhancement effect. The method comprises the following steps: modeling a penetrator small blood vessel to generate a blood vessel TOF (Time of Flight) image; calculating the blood vessel-grey matter contrast ratio of the blood vessel TOF image under different FAs and different TRs, and optimizing the FAs and the TRs according to the blood vessel-tissue contrast ratio; obtaining a structural image of the in-vivo brain tissue, performing cortex reconstruction, and calculating an angle of a surface normal of each cortex vertex relative to the B0 field to obtain a cortex angle graph; adjusting a single-layer blood vessel scanning plane angle according to the cortex angle diagram; and according to the optimized FA and TR and the single-layer blood vessel scanning plane angle determined in the step S4, carrying out magnetic resonance imaging on the cerebral cortex perforation small blood vessel. The invention further discloses a cerebral cortex puncturing small blood vessel magnetic resonance imaging device based on the inflow enhancement effect. According to the method and the device, magnetic resonance imaging of the cerebral cortex perforation small blood vessel can be realized under the non-invasive and contrast-agent-free conditions.

The invention relates to a deep learning fundus blood flow imaging method based on OCT amplitude and phase information, and belongs to the field of fundus blood flow imaging. An artificial neural network is used for fusing multi-modal information from OCT, multiple labels from OCTA are used for optimizing the network, and finally high-precision reconstruction of a blood flow image is achieved. Compared with the prior art, the method has the following outstanding advantages that firstly, phase information which is not fully utilized in a traditional method is creatively integrated, and the microvessel recognition sensitivity is remarkably improved through an amplitude-phase bimodal feature collaborative learning mechanism; secondly, a blood flow texture-blood flow position double-label supervised learning strategy is creatively provided, blood vessel position labels are introduced in a function loss stage, the network is assisted to accurately position the blood flow position, the characterization capacity of the network on capillary and other small blood vessel structures is effectively enhanced, and blood flow microcirculation visualization is achieved; and thirdly, blood flow information can be reconstructed by using single OCT scanning data.

The invention provides an OCTA retinal vessel segmentation method for improving U-Net based on adjacent feature extraction, and belongs to the field of medical image processing. According to the method, on the basis of a U-Net network, feature maps of adjacent layers are combined through a Proximal Feature Merge (PFM) module, and the capability of segmenting small blood vessels is remarkably improved; meanwhile, a shallow feature complementation module (SFC) is introduced, feature maps of different levels of an encoder are combined at the deepest level, and connectivity of segmented blood vessels is effectively guaranteed through a series of information recovery; in order to adapt to small blood vessels and refined edge information, a BottleUnext module is further added in the method, the resolution of the feature map is gradually amplified, then complex features are extracted, and finally the resolution of the feature map is recovered. Experiments show that compared with an existing method, the method shows more advanced performance in an OCTA retinal blood vessel task, and accurate segmentation of the retinal blood vessel is achieved.

A deep learning-based automatic segmentation method for diabetic retinopathy fundus images includes the following steps: S1 resizing the diabetic retinopathy fundus image to 512*512, S2 starting image preprocessing, S3 passing the preprocessed image into a network model for node feature initialization, S4 feature transfer and fusion, S5 each node passing features to the upper layer node, segmenting the vascular structure in the image through deconvolution / interpolation / depooling methods, and when the last node of the first layer completes feature fusion, the resolution is restored to the same size as the original input image. S6 After the vascular structure segmentation is completed, a grayscale image of the vascular structure segmentation is output through a single layer of convolution. The method provides automatic segmentation of fundus vascular structures for medical assistance systems, solving the problems of slow image reading, unclear vascular ends, and incomplete segmentation caused by the easy loss of new small blood vessels. The present invention has the advantages of automatic extraction of fundus blood vessels, low computational resource consumption, and high segmentation accuracy.

The present invention relates to a retinal vessel segmentation method and system. The method comprises the following steps: obtaining a retinal image dataset; processing the retinal image dataset of the present invention to obtain a processed retinal image dataset; constructing a bidirectional ConvLSTM residual UNet model; training the bidirectional ConvLSTM residual UNet model of the present invention to obtain a trained bidirectional ConvLSTM residual UNet model; and inputting the processed retinal image dataset of the present invention into the trained bidirectional ConvLSTM residual UNet model of the present invention to obtain a retinal vessel segmentation result. The method can distinguish small blood vessels from backgrounds in areas such as edges and low-contrast regions, thereby resolving overfitting and segmenting more complete blood vessels.

The present application belongs to the field of medical image segmentation of computer vision, and particularly relates to a retinal blood vessel segmentation method based on UNet and Transformer fusion, which has the following characteristics: step 1, pre-processing the image to be trained to obtain a pre-processed image; step 2, inputting the pre-processed image into a retinal blood vessel segmentation model based on UNet and Transformer fusion to obtain a weight file, the model comprising an encoder, a decoder and a fusion attention mechanism, the encoder comprising a multi-flow cascaded convolution layer, a plurality of pooling layers and a plurality of residual modules, each convolution layer using a residual module, and the pooling layer being arranged between two convolution layer units; the decoder comprising a plurality of improved residual modules based on extended convolution, a plurality of up-sampling modules and a deconvolution layer, the up-sampling module being arranged between two adjacent improved residual modules based on extended convolution; the fusion attention mechanism taking the output of the pooling layer in the encoder and the output of the adjacent flow pooling layer as low-level feature input and high-level feature input, respectively, and the high-level feature input of the fusion attention mechanism at the third layer being a feature map formed by the Transformer module corresponding to the encoder pooling layer; step 3, loading the weight file and inputting a test fundus image into the model to obtain a retinal blood vessel segmentation result. In addition, the retinal blood vessel segmentation model of the present application is more sensitive to small blood vessels, and the segmentation accuracy is improved.

The invention belongs to the technical field of medical devices, and discloses an intelligent thyroid postoperation compression hemostasis device which comprises a pressure acquisition module, a main control module, a compression module, a fixing module, a pressure adjusting module, a monitoring module, a sterilization module and a display module. By means of reasonable design and accurate pressure adjustment, the compression hemostasis device can enable small blood vessels of a surgical wound to be closed rapidly, postoperative bleeding is effectively reduced, the risk of hematoma formation is reduced, and wound healing is promoted. Due to the application of the pressure adjusting module, the medical staff can accurately control the compression pressure according to the specific condition of the patient, the adverse effect caused by too large or too small pressure is avoided, and the hemostasis effect and the safety of the patient are improved. The reliable fixing module ensures the stability of the compression hemostasis device in the moving process of the patient, the continuity of the compression hemostasis effect is ensured, and complications caused by device displacement are reduced.

The utility model discloses a covered stent capable of being used for treating aortic aneurysm diseases, which relates to the technical field of medical instruments and comprises a self-expanding stent, a clamping piece is fixedly connected to the waist line position of an expanding portion of the self-expanding stent, and the clamping piece is made of super-elastic memory alloy. A mode of combining radial force anchoring and structure anchoring is adopted, a part of an anchoring piece is changed into a clamping piece from a traditional barb, the stent moves downwards after arriving at a designed position, the anchoring piece is unfolded and fixed on small blood vessels on two sides, meanwhile, the upper end of the clamping piece is self-expanded and unfolded, double anchoring is carried out, displacement of the stent is avoided, and the stability of the stent is improved. A self-expanding stent is adopted for supporting and an elastic clamping piece and a native blood vessel are clamped for fixing, additional damage is avoided, meanwhile, double-layer polyurethane films are adopted for covering, the anti-leakage capacity is improved while the flexibility is guaranteed, the blood vessel can be prevented from being damaged, and stable support is provided.

The present application relates to a method, device, equipment and storage medium for enhancing blood vessel images. The method includes: obtaining a blood vessel segmentation image corresponding to an original blood vessel image and a blood vessel segmentation probability map corresponding to the original blood vessel image; determining target blood vessel points to be enhanced in the original blood vessel image according to the eigenvalue of each blood vessel point in the blood vessel segmentation image and a preset threshold; enhancing the eigenvalue of the target blood vessel points according to the eigenvalue of the target blood vessel points and the blood vessel segmentation probability map to obtain an enhanced blood vessel image. By using this method, small blood vessels and the ends of blood vessels can be clearly displayed.

The invention discloses a children tracheal intubation navigation device for difficult airways, the intubation navigation device comprises a catheter, the tip section of the catheter is sequentially provided with an SMA driving assembly and a piezoelectric ceramic driving assembly along a first direction, the SMA driving assembly can drive the tip section of the catheter to macroscopically rotate at a large angle, and the piezoelectric ceramic driving assembly can drive the tip section of the catheter to rotate at a large angle. The piezoelectric ceramic driving assembly can drive the sharp section of the guide pipe to rotate in a precise fine adjustment mode, and when the SMA driving assembly drives the sharp section of the guide pipe to rotate, the piezoelectric ceramic driving assembly is started to restrain vibration. The SMA driving assembly provides large-stroke and high-torque coarse adjustment and is suitable for passing through glottis, carina and other scenes needing large-angle steering, the piezoelectric ceramic driving assembly provides high-precision and high-response-speed fine adjustment, and avoiding of tiny blood vessels and tissues is achieved; the SMA driving assembly is prone to generating mechanical oscillation during starting and stopping, the sharp section of the catheter can shake to scratch fragile airway tissue, at the moment, the piezoelectric ceramic driving assembly actively restrains vibration, and the stability and safety of operation are greatly improved.

The invention discloses a retinal vessel segmentation method of a multi-scale residual U-Net model based on multi-tense features, and belongs to the field of computer vision and medical image processing. According to the method, double residual convolution (DResConv) is adopted to remarkably enhance the context feature extraction capability and improve the segmentation performance of small blood vessels under a complex background. In addition, a mixed downsampling block (MDB) combining an attention mechanism (SKNet) and a pooling operation is adopted, so that the detail loss is effectively reduced, and abundant blood vessel information is reserved at the same time. And finally, channel and space attention are introduced through a multi-scale multi-time fusion module (MFM), a richer multi-scale feature map is generated, and multi-scale feature expression at different moments is effectively realized. The method is more excellent in color fundus retina blood vessel segmentation tasks.

The invention discloses a lightweight OCTA retinal vessel segmentation method based on a full-convolution full-resolution network. According to the method, an encoder and two decoders are adopted; the encoder is combined with the thought of cyclic convolution and a ConvNeXt convolution module of a full-resolution design to extract a full-resolution feature map under the condition of not carrying out down-sampling so as to reserve information of small blood vessels to the maximum extent; dynamic snakelike convolution is introduced into the main decoder for self-adaptive fitting of the slender tubular shape of the blood vessel; the secondary decoder is mainly composed of a depth separable convolution and a point convolution so as to provide an additional supervision signal for the encoder in a training stage; according to the method, good balance between the segmentation performance and the calculation efficiency can be realized, the segmentation performance equivalent to the segmentation performance in the prior art can be realized only by using lightweight calculation cost, and the method has application potential in a computing resource limited scene.

The invention discloses a hemostasis device for general surgery operation, relates to the technical field of hemostasis instruments, and provides the following scheme that the hemostasis device comprises a hemostasis forceps instrument, an upper-layer hemostasis dressing assembly is arranged on the front side of an outer ring of the hemostasis forceps instrument, and a lower-layer hemostasis dressing assembly is arranged on the surface of the front side of the upper-layer hemostasis dressing assembly; plug connectors are fixed to the upper area and the lower area of an inner cavity of the upper-layer hemostatic dressing assembly, and a low-temperature manufacturing part is arranged on the end face of the rear side of the upper-layer hemostatic dressing assembly. The dressing capable of being connected with the hemostatic forceps in a clamping mode is designed, the wound end of a small blood vessel in a wound is blocked and stopped by means of the dressing, the dressing can be attached to the skin around the wound, the hemostatic forceps can be supported, the hemostatic forceps clamping a large blood vessel can be kept in a hemostatic posture, and the dressing is designed to be of a multi-layer structure, so that the dressing is convenient to use. A refrigeration structure is arranged in the upper layer, the temperature of blood vessels around a wound is reduced in a heat exchange mode, and the bleeding amount is reduced through cold contraction of the blood vessels.

The invention relates to a vascular intervention radio-frequency pulse ablation operation system, and relates to the technical field of medical instruments. The vascular intervention radio frequency pulse ablation operation system comprises an outer sleeve, an inner tube, a guide wire and a pulse radio frequency generating element, the outer sleeve is arranged on the inner tube in a sliding and sleeving mode, and the guide wire penetrates through the inner tube in a sliding mode; an operating handle is fixedly arranged at one end of the outer sleeve; one end of the inner tube is sleeved with an elastic balloon capable of extending out of the other end of the outer sleeve, and the surface of the elastic balloon is coated with a silver paste electrode electrically connected with the pulse radio frequency generation element; a deflection guide head is arranged at the end, close to the elastic balloon, of the guide wire and can deflect and swing when encountering resistance. Through the design of the deflection guide head, the guide wire can deflect and swing when encountering resistance, the navigation capacity of the guide wire in a complex small blood vessel is improved, and the risk of damage to the inner wall of the blood vessel is reduced. The silver paste electrode coated on the surface of the elastic balloon realizes efficient and uniform ablation of plaques, and the treatment effect is improved.

The application discloses a method for training a segmentation model for cerebral vascular segmentation and related products. The method comprises: obtaining a three-dimensional gray image and a corresponding three-dimensional blood vessel segmentation mask based on a three-dimensional angiogram; generating a three-dimensional gray entropy feature map according to the three-dimensional gray image; generating a three-dimensional label entropy map according to the three-dimensional blood vessel segmentation mask; performing feature fusion on the three-dimensional gray image and the three-dimensional gray entropy feature map to obtain fused features; inputting the fused features into the segmentation model for blood vessel segmentation and calculating a weighted loss based on the three-dimensional label entropy map; and training the segmentation model for cerebral vascular segmentation based on the weighted loss. The scheme of the application can improve the recognition ability of the model for blood vessel boundaries, small blood vessels and complex structure regions, effectively reduce the blood vessel adhesion phenomenon, and enhance the robustness and clinical applicability of the segmentation result.

The present invention belongs to the technical field of biomedicine, and specifically relates to the application of 9-hydroxy pentadecanoic acid in the preparation of a drug for preventing and treating lung injury. By establishing a mouse model of radon inhalation-induced lung injury, the present invention found that after exposure to different radon levels, the body weight of the mice decreased, the lung coefficient increased, the tidal volume, minute ventilation volume and maximum inspiratory flow of the lungs decreased, the respiratory function was inhibited, the alveolar cavities of the mice were abnormally dilated, the lung septum was thickened and fractured, accompanied by more fibrin exudation, obvious inflammatory infiltration appeared beside the bronchi, congestion occurred in the small blood vessels, and collagen deposition occurred in the lung tissue, which could lead to the occurrence of pulmonary fibrosis and cause lung injury. By intragastric administration of 9-hydroxy pentadecanoic acid, the above phenomena can be effectively inhibited. Therefore, 9-hydroxy pentadecanoic acid can be used for the prevention and treatment of lung injury, especially for the prevention and treatment of lung injury caused by radon exposure.