Scanning device and method for intelligent medical image processing

Abstract

The invention relates to the technical field of medical images, in particular to a scanning device and method for intelligent medical image processing.The scanning device comprises scanning equipment and a scanning bed, the scanning bed comprises a bed frame, the scanning equipment is in signal connection with an image processing system, a first mattress is fixedly connected to the top of the bed frame, and hinge assemblies are fixedly connected to the two ends of the bottom of the first mattress; the other end of the hinge assembly is fixedly connected with a second mattress and a third mattress respectively; each of the first mattress, the second mattress and the third mattress is internally divided into a flexible layer and a sensing layer from top to bottom, a plurality of bulbs are arranged in each flexible layer, a lifting assembly and a sensing assembly are arranged in each sensing layer, the top of each lifting assembly is fixedly connected with the corresponding sensing assembly, and the sensing assemblies are electrically connected with the bulbs. The scanning device can rapidly acquire the body contour of the patient and then carry out positioning scanning. The system is used for realizing convenient positioning, efficient scanning and accurate image processing of the patient, so that the accuracy and efficiency of medical image diagnosis are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of medical imaging technology, in particular to a scanning device and method for intelligent medical image processing. BACKGROUND

[0002] The scanning device and method for intelligent medical image processing play a crucial role in modern medical diagnosis, especially in the diagnosis and postoperative evaluation of orthopedic diseases. For patients with fractures, traditional CT examination faces significant challenges: the severe pain at the fracture site makes it difficult for patients to adjust their body position independently, especially for patients with fractures in the spine, hip joint, long bones of the limbs, etc. During the process of lying on the examination bed and body position fixation, secondary pain is often caused by body position changes, and even the displacement of the fracture end may increase the risk of secondary injury. At the same time, the difficulty in changing body position makes it difficult for patients to maintain a stable posture during scanning, and slight movement can easily produce scanning artifacts, seriously affecting the clarity of key image information such as fracture lines and joint alignment, which may lead to deviations in the doctor's judgment of fracture type and the degree of intra-articular fracture displacement. In addition, in the postoperative review of fractures, if the body position of the two scans is inconsistent, the imaging comparison will lose its reference, making it difficult to accurately observe the details of the callus growth at the fracture end and the changes in the position of internal fixation, greatly reducing the clinical reference value of the review.

[0003] However, existing medical image scanning devices still have some significant defects in actual application, especially in patient positioning and scanning preparation. Most existing devices require patients to be placed in a specific position, which undoubtedly increases the difficulty of operation and the preparation time before scanning for patient groups that cannot move independently, such as severely injured fracture patients, paralyzed patients, or elderly patients. These patients often need the assistance of medical staff to complete positioning, not only consuming human resources, but also possibly leading to prolonged scanning procedures and affecting overall scanning efficiency. More seriously, due to patient movement or inaccurate positioning, the scanning results may be biased, affecting the doctor's accurate judgment of the patient's condition. In addition, traditional positioning methods cannot accurately adapt to the body position restrictions of fracture patients, leading to an increase in scanning error rate; image processing algorithms have difficulty effectively extracting true fracture features when faced with artifacts caused by body position changes, further affecting diagnostic accuracy. Therefore, developing an intelligent medical image scanning device and method specifically for orthopedic examination needs that can achieve painless positioning, stable body position maintenance, and precise body position replication for fracture patients is of great significance to improving the efficiency of orthopedic diagnosis and postoperative evaluation. SUMMARY

[0004] To solve the above problems, the present application provides a scanning device and method for intelligent medical image processing, which is used to realize the convenient positioning, efficient scanning and accurate image processing of patients, thereby significantly improving the accuracy and efficiency of medical image diagnosis.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A scanning device for intelligent medical image processing includes a scanning device and a scanning bed. The scanning bed includes a bed frame. The scanning device is signal-connected to an image processing system. A first mattress is fixedly connected to the top of the bed frame. Hinges are fixedly connected to both ends of the bottom of the first mattress. A second mattress and a third mattress are fixedly connected to the other end of the hinges respectively. The hinges are signal-connected to the image processing system. The hinges are driven by the image processing system to adjust the angle between the first mattress and the second mattress, as well as between the first mattress and the third mattress.

[0006] The first, second, and third mattresses are each divided into a flexible layer and a sensing layer from top to bottom. Each flexible layer contains several light bulbs, and each sensing layer contains a lifting component and a sensing component. The top of the lifting component is fixedly connected to the sensing component. The lifting component lifts the sensing component upwards based on the patient's weight estimated by the image processing system. The sensing component and the light bulbs are electrically connected. The sensing component is used to sense the pressure generated by the patient lying on the first, second, and third mattresses and change the current passing through the light bulb according to the pressure, so that the scanning device can quickly acquire the patient's body contour and perform a positioning scan.

[0007] The technical principles of the above solution are as follows:

[0008] By energizing the bulbs inside the first, second, and third mattresses to keep them constantly lit, the scanning bed initially functions as a folding chair. The patient's buttocks are placed on the first mattress, leaning against the second mattress, and their legs on the third mattress. The angle between the first, second, and third mattresses is adjusted via a hinge assembly. Once the angles of the first, second, and third mattresses are fixed, the pressure exerted by the patient on the flexible layers of the mattresses stabilizes. At this point, the flexible layers indent due to pressure from different parts of the patient. Based on the patient's weight estimated by the image processing system, the sensing component is dynamically raised to accommodate patients of different weights. The sensing component triggers a corresponding circuit based on pressure changes, causing the bulbs to disconnect. Because the bulbs are arranged in a certain pattern within the flexible layers, when a bulb disconnects due to the triggering of the sensing component, a dark area corresponding to the patient's body contour is formed on the flexible layer. The scanning device can quickly capture this dark area, thereby accurately acquiring the patient's body contour, and then perform a personalized and precise scan of the patient's affected area.

[0009] When a patient's body shifts during the scan, the pressure on the flexible layer changes, causing the sensor component within the flexible layer to press down, which in turn changes the state of the light bulb. The change in the light bulb's brightness repositions the patient's body outline and location. If the change in the light bulb's brightness is too large, it indicates that the patient's displacement is too great, and the image processing system issues a warning to prompt medical staff to assist the patient in adjusting their position.

[0010] The above approach has the following beneficial effects:

[0011] 1. This solution incorporates light bulbs and sensors inside the mattress. When a patient lies on the mattress, the sensors precisely trigger the corresponding light bulb to disconnect based on the patient's body contours and pressure distribution, creating a dark area on the flexible layer corresponding to the patient's body. The scanning device can then acquire the patient's body contours through changes in resistance, achieving precise positioning and accurate scanning. Furthermore, as the patient applies pressure to the first, second, and third mattresses, the flexible layer around the body indents, causing a decrease in brightness in the light bulbs and sensors. This peripheral area of ​​the light bulb is used as the outline of the patient's clothing, increasing the scanning device's margin of error. By comparing the current flowing through the light bulbs at the points where the patient's body is pressed down, the scanning accuracy is further improved.

[0012] 2. In this solution, if the patient's body moves slightly during the scanning process, the sensing component can immediately detect the change in pressure and trigger the change in the light bulb's on / off state. This allows the image processing system to acquire the patient's position and posture in real time and reposition the patient if necessary, ensuring the accuracy of the scan. If the patient's displacement is too large, the image processing system can issue a warning, prompting medical staff to assist the patient in adjusting their position, thereby avoiding deviations in the scan results.

[0013] 3. This solution uses a first, second, and third mattress to initially divide different parts of the patient's body, which can reduce the need for positioning and posture adjustments when lying flat, thus improving scanning efficiency. Furthermore, since some critically ill patients have limited mobility, the angles between the second and third mattresses and the first mattress can be adjusted in advance so that the patient can lean on the first, second, and third mattresses in a more comfortable position. The angles can then be adjusted using a hinge assembly, further enhancing patient comfort.

[0014] Furthermore, each hinge assembly includes several second bases and a first base fixedly connected to the bottom of the first mattress and symmetrically arranged. The second bases are respectively fixedly connected to the bottom of the second mattress and the third mattress. A first connecting block is hinged to one side of the first base, and a second connecting block is hinged between the first connecting block and the second base. A connecting rod is hinged to the other side of the first base, and a third connecting block is hinged to one end of the connecting rod. The third connecting block is hinged to the first connecting block and the second base. A telescopic guide rod is hinged between the third connecting block and the second connecting block. A tension spring is sleeved on the telescopic guide rod, and both ends of the tension spring are hinged to the third connecting block and the second connecting block, respectively. A drive assembly is fixedly connected to each first base. The drive assembly is used to drive the first connecting block and the connecting rod to rotate, thereby pulling or pushing the second connecting block and the third connecting block to drive the second base to move.

[0015] Beneficial effects: Through the multi-hinge connection and tension spring design, the hinge assembly allows for flexible movement of the second or third mattress relative to the first mattress while maintaining sufficient stability, ensuring safety and accuracy during the scanning process; the tension springs keep the entire hinge assembly stable when adjusting the angle between the second or third mattress and the first mattress, thereby improving the stability of the first, second, and third mattresses; the drive assembly can precisely control the rotation of the first connecting block and the connecting rod, thereby achieving precise adjustment of the angle between the second or third mattress and the first mattress, which helps to optimize the angle configuration according to the patient's body shape and scanning needs during the scanning process; by quickly adjusting the mattress angle, medical staff can locate the patient and prepare for the scan more quickly, thereby shortening the pre-scan preparation time and improving scanning efficiency.

[0016] Furthermore, each drive assembly includes a drive box fixedly connected to the side of the first base away from the first mattress. Each drive box has a symmetrically arranged motor fixedly connected to its inner sidewall. Each motor is connected to the image processing system signal. Each motor has a transmission rod fixedly connected to its output shaft. One end of the first connecting block and the connecting rod are coaxially fixedly connected to the transmission rod.

[0017] Beneficial effects: Through precise control of the motor, precise rotation control of the transmission rod, the first connecting block and the connecting rod can be achieved, which helps to fine-tune the angle between the second or third mattress and the first mattress according to the patient's body shape and scanning needs during the scanning process, ensuring the accuracy and comfort of the scan.

[0018] Furthermore, each lifting component includes a lifting plate and several electrically operated telescopic rods fixedly connected to the bottom wall of the sensing layer. The electrically operated telescopic rods are all connected to the image processing system signal. The output shaft of each electrically operated telescopic rod is coaxially fixedly connected to a slider. The slider is fixedly connected to one side of the lifting plate. A sliding groove is opened on the inner side wall of the sensing layer. The slider slides in the sliding groove. The top of the lifting plate is fixedly connected to the sensing component.

[0019] Beneficial effects: The height of the lifting plate can be adjusted to accommodate patients of different weights. When the patient is heavy, the flexible layer is pressed down a greater distance, and the displacement distance of the sensing component increases. At this time, the lifting height path controlled by the electric telescopic rod is short. When the patient is light, the flexible layer is pressed down a smaller distance, and the displacement distance of the sensing component decreases. At this time, the lifting height path controlled by the electric telescopic rod is long. This intelligent adjustment makes the degree of light bulb extinguishing and dimming of the sensing component more consistent, thus making it easier for the image processing system to judge the patient's body contour based on the degree of light bulb brightness.

[0020] Furthermore, each sensing component includes several telescopic components fixedly connected to the top of the lifting plate and sleeves fixedly connected to the bottom of the flexible layer corresponding to the telescopic components. Each telescopic component has a contact on one side of its top, and each sleeve has a slider fixedly connected to the side of the contact. The contact and the slider are slidably connected. The bulbs are connected in parallel to form a parallel circuit, and the parallel circuit is connected in series with a power supply. The positive and negative terminals of the power supply are electrically connected to the input and output terminals of the parallel power supply, respectively. The sliders are connected in series in the circuit of the bulbs and the power supply to form a closed loop. When the contact and the slider are in contact, the contact and the slider are connected in series.

[0021] When the top of the telescopic component contacts the inner top wall of the sleeve, it is compressed, which makes the contact always contact the slider on the inner side wall of the sleeve. The slider and the bulb are electrically connected to form a closed circuit. When the contact slides on the slider, the resistance value in the closed circuit increases as the contact slides from top to bottom, which reduces the current passing through the bulb and makes the bulb dimmer.

[0022] Beneficial effects: When the patient applies high pressure to the flexible layer, the telescopic component contacts the top wall of the sleeve after entering the sleeve and compresses the telescopic component, allowing it to adapt to the pressure change. The sliding plate acts as a sliding rheostat; when the contact slides on the sliding plate, it changes the resistance in the bulb circuit, causing a change in the current. The bulb's brightness depends on the current; a decrease in current causes the bulb to dim. When the contact slides to the top of the sliding plate, the resistance in the circuit reaches its maximum value. At this point, the current in the circuit decreases to a level insufficient to light the bulb, causing it to go out. The brightness of the bulb is used to quickly acquire the patient's body contour, thereby improving image scanning efficiency.

[0023] Furthermore, each telescopic component includes a telescopic rod fixedly connected to the top of the lifting plate, a movable block fixedly connected to the top of each telescopic rod, a contact fixedly connected to one side of the movable block, and a spring sleeved on the outside of each telescopic rod, with both ends of the spring fixedly connected to the bottom of the movable block and the top of the lifting plate, respectively.

[0024] Beneficial effects: The combined design of the telescopic rod and spring allows the telescopic component to accurately sense and respond to pressure changes when the patient lies on the mattress. When pressure increases, the telescopic rod compresses the spring downwards, causing the moving block and contacts to move; when the patient shifts, the pressure at a certain point decreases or disappears, and the spring's restoring force returns the telescopic rod to its original position, with the contacts also returning to their initial position. This process causes a change in current in the circuit, causing the light bulb to turn on and off, which is received by the image processing system. This allows medical personnel to quickly determine the patient's displacement and intervene in repositioning if necessary, improving the accuracy of image scanning.

[0025] Furthermore, the first mattress is used to place the patient's buttocks and surrounding area, while the second and third mattresses are used to place the patient's torso and lower limbs, respectively. Armrests are fixedly connected to both sides of the first mattress, and footrests are fixedly connected to the third mattress.

[0026] The foot pedal is equipped with an air bladder, and there is an air cavity between the lifting plate and the flexible layer. The air cavity is connected to the air bladder. When the lifting plate is raised, the air in the air cavity is compressed and discharged into the air bladder. After the air bladder expands, it is used to fix the patient's foot part located in the foot pedal.

[0027] Beneficial effects: The segmented design of the first, second, and third mattresses allows patients to directly adjust their lying position according to the position and shape of each mattress, eliminating the need for complex positioning adjustments while lying flat. This simplifies the positioning process and improves positioning efficiency. The pre-segmented mattress design helps ensure that patients maintain a stable posture during scanning or treatment, thereby improving scanning accuracy. Placing their hands on the handrails helps stabilize their hands, and airbags help stabilize their feet within the footrests, further enhancing scanning accuracy.

[0028] Furthermore, the image processing system includes a control module, a receiving module, a positioning module, a warning module, a processing module, and an output module;

[0029] The control module is used to control the operation of the motor, electric telescopic pole, and scanning equipment;

[0030] The receiving module is used to receive the light bulb brightness signal and then transmit it to the positioning module; the receiving module is also used to receive image information after the scanning device scans the patient's lesion.

[0031] The positioning module is used to obtain the patient's body contour based on the signal that the bulbs in the first, second, and third mattresses are constantly lit, and to further obtain the part of the patient's body that is pressing down on the mattress based on the signal that the bulbs in the first, second, and third mattresses are bright and off, and the current change of the bulbs, so as to obtain the patient's position in sequence.

[0032] The warning module is used to monitor patient displacement during scanning and, if necessary, issue an alert and notify medical staff or the patient to adjust their position to continue the scanning.

[0033] The processing module is used to process the patient lesion information obtained by the receiving module through a deep convolutional neural network to perform image analysis and identify the type and location of the patient's lesion;

[0034] The output module is used to output the image information processed by the processing module, so that it can be presented to doctors intuitively for medical image diagnosis.

[0035] Beneficial effects: The control module can control the operation of the motor, electric telescopic rod and scanning equipment to realize an automated and intelligent scanning process, which not only improves scanning efficiency, but also reduces human error and uncertainty;

[0036] The receiving module receives the light bulb's on / off signals and transmits them to the positioning module. The positioning module then uses the constant on / off signals of the light bulb inside the mattress to acquire the patient's body contours and pressure points, thereby determining the patient's position. This design enables real-time tracking and precise positioning of the patient, providing stability and accuracy for subsequent scanning and processing.

[0037] The warning module can issue a timely warning when the patient moves and notify medical staff or the patient to adjust their posture, which helps to ensure the accuracy of the scanning process and avoid scanning errors caused by patient movement.

[0038] The processing module employs deep convolutional neural network (DNN) image processing technology, enabling high-precision analysis of patient case images acquired by the receiving module. DNNs possess powerful feature extraction and pattern recognition capabilities, accurately extracting key information from complex images, such as the shape of lesion areas, outlining lesion boundaries, and determining the texture and pattern features of lesions. Through deep learning and training, the processing module can accurately identify and classify the type and location of patient lesions. This recognition capability is not limited to common lesion types but can also accurately diagnose complex or rare cases. This significantly improves the accuracy and efficiency of doctors' diagnoses, reducing the risk of misdiagnosis and missed diagnosis.

[0039] The output module can visually present the image information processed by the processing module to the doctor. This intuitive presentation helps doctors quickly understand the patient's condition and improves their diagnostic efficiency.

[0040] The entire image processing system achieves full-chain automation and intelligence from patient positioning, scanning, early warning, processing to output through the collaborative work of its various modules. This not only improves the overall performance and stability of the system, but also reduces the complexity and cost of manual operation.

[0041] Furthermore, the processing module includes a preprocessing unit, a segmentation unit, and a postprocessing unit;

[0042] The preprocessing unit is used to employ a global histogram equalization algorithm to improve image brightness, reduce the grayscale value interval, and equalize the grayscale values.

[0043] The segmentation unit is used to segment case images. It uses the upsampling and downsampling process of double convolution operation to directly connect the input and output. The 1×1 convolution connected to the original feature map in the original module is moved to the 1×1 convolution operation in the right path of the module to further increase the receptive field.

[0044] The post-processing unit is used to randomly initialize segmentation labels based on the MRF algorithm after the segmentation unit obtains the segmentation results of the case images, and then aggregates similar pixels in the segmented case image information to eliminate noise in the segmented case images.

[0045] Beneficial effects: Through the global histogram equalization algorithm, the preprocessing unit can significantly improve the brightness of the image and reduce the interval between image gray values, thereby achieving gray value equalization. This helps to improve the visual effect of the image, making the lesion area clearer and facilitating subsequent image analysis and processing. The preprocessing unit also includes an image denoising step, which can effectively remove noise and interference information in the image, helping to improve the image clarity and quality, and providing more accurate and reliable input for subsequent segmentation and recognition tasks.

[0046] The segmentation unit employs a dual convolution operation for upsampling and downsampling, enabling segmentation of case images. This segmentation method not only boasts high accuracy but also adapts to lesion regions of varying sizes and shapes, enhancing its flexibility and applicability. The segmentation unit moves the 1×1 convolution connected to the original feature map in the original module before the 1×1 convolution operation on the right path of the module, further increasing the receptive field and helping to capture more image information, thus improving segmentation accuracy and robustness. Furthermore, the segmentation unit directly connects the input and output, which helps retain more original image information, reduces information loss, and improves the quality of the segmentation results.

[0047] The post-processing unit helps eliminate noise in segmented case images, improving image smoothness and clarity. By aggregating similar pixels, the post-processing unit can effectively eliminate noise and isolated points in segmented case images, helping to improve the overall image quality and visual effect, and providing doctors with a more accurate and reliable basis for diagnosis.

[0048] A scanning method for intelligent medical image processing includes the following steps:

[0049] Step 1, Positioning and Fixation: Before the patient lies down, medical staff input the fracture site to be examined through the image processing system. The system provides a preset orthopedic positioning plan based on the examination site and automatically adjusts the initial angles of the first, second, and third mattresses. The bulbs inside the first, second, and third mattresses are powered on and kept constantly lit. The scanning bed is initially folded into a chair shape. The patient places their buttocks on the first mattress, hands on the armrests, and legs on the footrests. The motor is started through the control module according to the area to be scanned to adjust the angle between the first, second, and third mattresses.

[0050] Step two, sensing and positioning: The flexible layer indents due to pressure from different parts of the patient, thus pushing the sleeve downwards. Simultaneously, the control module adjusts the lifting height of the lifting plate based on the pre-input patient weight information. At this time, some movable blocks enter the lowered sleeve, causing the contact points to slide at different distances on the slider. As the sliding displacement between the contact points and the slider increases, the bulb gradually turns from bright to dark. The receiving module acquires the bulb's brightness information and converts it into a two-dimensional image. The constantly lit bulb represents the flexible layer that is not in contact with the patient, depicting the patient's body contour. The parts of the bulb that dim or turn off, as well as the parts of the bulb where the current changes, represent the patient's body position. Based on the patient's body contour map in the image processing system, the scanning device is controlled to scan the patient's body.

[0051] When the patient's body shifts during the scanning process, the system immediately corrects the patient's position by turning the bulb on and off and changing the current. It acquires the patient's displacement information and notifies medical staff to intervene and reposition the patient. When the warning module of the image processing system issues a warning, it notifies medical staff to assist the patient in adjusting the position.

[0052] Step 3, Scanning Processing: A specific area of ​​the patient to be examined is scanned using a scanning device at the corresponding level to obtain cross-sectional image information of the area to be examined and transmit it to the image processing system. The preprocessing unit preprocesses the image information and segments the image information based on a convolutional neural network algorithm to identify and label the lesion area. After the segmented image is processed by the post-processing unit, it is transmitted externally, and the doctor performs medical image diagnosis with the assistance of intelligent annotation in the image processing system.

[0053] Beneficial effects: By utilizing the indentation of the flexible layer caused by varying pressure at different parts of the patient, the sleeve is moved downwards, and the lifting plate is adjusted via an electric telescopic rod. Combined with pre-inputted patient weight information, the image processing system can intelligently generate a two-dimensional image of the patient's body contour and position based on the light bulb's on / off information. This method not only improves the accuracy of positioning but also allows for real-time monitoring of the patient's displacement during the scanning process, ensuring accuracy throughout the scan.

[0054] When a patient experiences slight displacement during the scan, the system can quickly detect and reposition the patient's body outline and location. If the displacement is too large, the system will issue a warning and notify medical staff to assist the patient in adjusting their position, thereby avoiding scanning errors caused by patient movement.

[0055] Cross-sectional images of the patient's lesion site are acquired using scanning equipment and transmitted to an image processing system for preprocessing, image segmentation, and post-processing. The system employs a convolutional neural network algorithm to segment and label the image information, accurately identifying lesion sites and generating images easily accessible for diagnosis. This method not only improves scanning efficiency but also reduces the complexity and cost of manual operation. Through the image processing system's preprocessing, segmentation, and post-processing, doctors can visually view images with labeled lesion sites, thereby improving diagnostic accuracy and reducing the risk of misdiagnosis and missed diagnosis. Simultaneously, the system's automation and intelligence significantly improve diagnostic efficiency, enabling doctors to provide patients with effective treatment plans more quickly. Attached Figure Description

[0056] Figure 1 This is an isometric view of a scanning bed in an embodiment of the intelligent medical image processing scanning device and method of the present invention;

[0057] Figure 2 This is an isometric schematic diagram of the hinge assembly in an embodiment of the intelligent medical image processing scanning device and method of the present invention;

[0058] Figure 3 This is a schematic cross-sectional view of the flexible layer and sensing layer under normal conditions in an embodiment of the intelligent medical image processing scanning device and method of the present invention.

[0059] Figure 4 This is a front cross-sectional view of the flexible layer and sensing layer under pressure in an embodiment of the intelligent medical image processing scanning device and method of the present invention.

[0060] Figure 5 This is a front cross-sectional view of the sensing components in an embodiment of the intelligent medical image processing scanning device and method of the present invention.

[0061] Figure 6 This is a schematic diagram of the method steps in an embodiment of the intelligent medical image processing scanning device and method of the present invention;

[0062] Figure 7 This is a schematic diagram of the system framework of an embodiment of the intelligent medical image processing scanning device and method of the present invention.

[0063] The reference numerals in the accompanying drawings of the instruction manual include: 1. Bed frame; 2. Second mattress; 3. Armrest frame; 4. First mattress; 5. Third mattress; 6. Foot pedal; 7. First base; 8. First connecting block; 9. Connecting rod; 10. Tension spring; 11. Third connecting block; 12. Second base; 13. Second connecting block; 14. Flexible layer; 15. Light bulb; 16. Sleeve; 17. Sensing layer; 18. Lifting plate; 19. Movable block; 20. Spring; 21. Telescopic rod; 22. Slider; 23. Electric telescopic rod; 24. Drive box; 25. Sliding plate; 26. Contact point. Detailed Implementation

[0064] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0066] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0067] The following detailed description illustrates the specific implementation method:

[0068] Example 1:

[0069] As attached Figure 1 , Figure 2 , Figure 3 , Figure 4 and Figure 5The following describes a smart medical image processing scanning device, comprising a scanning unit and a scanning bed. The scanning bed includes a bed frame 1. The scanning unit is signal-connected to an image processing system. A first mattress 4 is bolted to the top of the bed frame 1. Hinges are screwed to both ends of the bottom of the first mattress 4. A second mattress 2 and a third mattress 5 are screwed to the other ends of the hinges, respectively. The hinges are signal-connected to the image processing system and are driven by the image processing system to adjust the angles between the first mattress 4 and the second mattress 2, as well as between the first mattress 4 and the third mattress 5. The first mattress 4 is used to position the patient's buttocks and surrounding area, while the second mattress 2 and the third mattress 5 are used to position the patient. For the torso and lower limbs, armrests 3 are bolted to both sides of the first mattress 4, and footrests 6 are bolted to the third mattress 5. Through the segmented design of the first mattress 4, the second mattress 2, and the third mattress 5, the patient can directly adjust their lying position according to the position and shape of each mattress without having to make complicated positioning adjustments while lying flat, simplifying the positioning process and improving positioning efficiency. The preliminary segmented mattress design helps ensure that the patient maintains a stable posture during scanning or treatment, thereby improving scanning accuracy. By placing their limbs on the armrests 3 and footrests 6, the patient can fix their limbs and achieve preliminary positioning, thereby further improving the accuracy of the scan.

[0070] The first mattress 4, the second mattress 2, and the third mattress 5 are each divided into a flexible layer 14 and a sensing layer 17 from top to bottom. Each flexible layer 14 is equipped with several light bulbs 15, and each sensing layer 17 is equipped with a lifting component and a sensing component. The top of the lifting component is welded to the sensing component. The lifting component lifts the sensing component upward according to the patient's weight estimated by the image processing system. The sensing component and the light bulbs 15 are electrically connected. The sensing component is used to sense the pressure generated by the patient lying on the first mattress 4, the second mattress 2, and the third mattress 5 and change the current through the light bulb 15 according to the pressure, so that the scanning device can acquire the patient's body contour and perform positioning scanning.

[0071] Each hinge assembly includes several second bases 12 and a first base 7 symmetrically arranged and screwed to the bottom of the first mattress 4. The second bases 12 are screwed to the bottom of the second mattress 2 and the third mattress 5, respectively. A first connecting block 8 is hinged to one side of the first base 7, and a second connecting block 13 is hinged between the first connecting block 8 and the second base 12. A connecting rod 9 is hinged to the other side of the first base 7, and a third connecting block 11 is hinged to one end of the connecting rod 9. The third connecting block 11 is hinged to the first connecting block 8 and the second base 12, respectively. A telescopic guide rod is hinged between the third connecting block 11 and the second connecting block 13. A tension spring 10 is sleeved on the telescopic guide rod, and both ends of the tension spring 10 are hinged to the third connecting block 11 and the second connecting block 13, respectively. A drive assembly is screwed to each of the first bases 7. The drive assembly is used to drive the first connecting block 8 and the connecting rod 9 to rotate, thereby pulling or pushing the second connecting block 13 and the third connecting block 11 to move the second base 12.

[0072] Each drive assembly includes a drive box 24 screwed to the bottom of the first base 7. Symmetrically arranged motors are screwed to the inner wall of the drive box 24. Each motor is connected to the image processing system signal. A transmission rod is welded to the output shaft of each motor. One end of the first connecting block 8 and the connecting rod 9 are coaxially welded to the transmission rod. When the motor operates, the first connecting block 8 and the connecting rod 9 rotate with the motor. At this time, the first connecting block 8 and the connecting rod 9 simultaneously pull the third connecting block 11, and the first connecting block 8 pulls the second connecting block 13, thereby pulling the second base 12 and causing the angle of the second mattress 2 or the third mattress 5 to change. During this process, the tension spring 10 compresses as the angle between the first base 7 and the second base 12 decreases, ensuring that the first connecting block 8 and the second connecting block 13, and the first connecting block 8 and the connecting rod 9 and the third connecting block 11 remain stable. Furthermore, the tension spring 10 can adapt to this angle deformation, reducing wear between the hinge components.

[0073] Each lifting assembly includes a lifting plate 18 and several electrically operated telescopic rods 23 connected to the inner bottom wall of the sensing layer 17 by screws. The electrically operated telescopic rods 23 are all connected to the image processing system signal. The output shafts of the electrically operated telescopic rods 23 are all coaxially connected to sliders 22 by screws. The sliders 22 are all fixedly connected to one side of the lifting plate 18. The inner side wall of the sensing layer 17 has a sliding groove, and the sliders 22 slide in the sliding groove. The top of the lifting plate 18 is fixedly connected to the sensing assembly.

[0074] Each sensing component includes several telescopic components welded to the top of the lifting plate 18 and sleeves 16 bonded to the bottom of the flexible layer 14 corresponding to the telescopic components. Each telescopic component has a contact 26 on one side of its top. Each sleeve 16 has a slider 25 welded to the side of the sleeve near the contact 26. The contact 26 and the slider 25 are slidably connected. The bulbs 15 are connected in parallel to form a parallel circuit. The parallel circuit is connected in series with a power supply. The positive and negative terminals of the power supply are electrically connected to the input and output terminals of the parallel power supply, respectively. The sliders 25 are connected in series in the circuit of the bulbs 15 and the power supply to form a closed loop. When the contact 26 contacts the slider 25, the contact 26 and the slider 25 are connected in series.

[0075] When the top of the telescopic component contacts the inner top wall of the sleeve 16, it is compressed, which causes the contact 26 to always contact the slider 25 on the inner side wall of the sleeve 16. When the contact 26 slides on the slider 25, the resistance value in the closed circuit increases during the sliding process of the contact from top to bottom, which reduces the current value when passing through the bulb 15, and the bulb 15 dims.

[0076] Each telescopic assembly includes a telescopic rod 21 fixedly connected to the top of the lifting plate 18. A movable block 19 is welded to the top of each telescopic rod 21, and a contact point 26 is welded to one side of the movable block 19. A spring 20 is sleeved on the outside of each telescopic rod 21, and the two ends of the spring 20 are welded to the bottom of the movable block 19 and the top of the lifting plate 18, respectively.

[0077] The foot pedal 6 is equipped with an air bladder, and there is an air chamber between the lifting plate 18 and the flexible layer 14. The air chamber is connected to the air bladder, which can achieve precise fixation of the affected limb during orthopedic examinations: when the lifting plate 18 is raised, the air in the air chamber is compressed and discharged into the air bladder. After the air bladder expands, it is used to fix the patient's foot located in the foot pedal 6, preventing the limb from shaking due to involuntary muscle contraction during the scanning process, thereby reducing the probability of artifacts.

[0078] The specific implementation process is as follows: Before the patient lies down, medical staff input the fracture site to be examined through the image processing system. The system provides a preset orthopedic positioning plan based on the examination site and automatically adjusts the initial angles of the first mattress 4, the second mattress 2, and the third mattress 5. The bulbs 15 inside the first mattress 4, the second mattress 2, and the third mattress 5 are powered on and kept constantly lit. The scanning bed is initially folded into a chair shape. The patient places their hands on the armrests 3, which is convenient for patients with upper limb fractures to position their arms and avoid pain and positional shift caused by dangling arms. For example, a patient with a humerus fracture can adjust the armrests 3 to a 45° angle with the bed surface, allowing the affected limb to rest naturally, reducing pain and ensuring positional stability during the scan. With legs positioned on footrest 6, the motor is activated via the image processing system according to the area to be scanned, adjusting the angle between the first mattress 4, the second mattress 2, and the third mattress 5. The first mattress 4, the second mattress 2, and the third mattress 5 are used to initially segment different parts of the patient's body, reducing the need for posture adjustments when lying flat and improving scanning efficiency. Furthermore, since some critically ill patients have limited mobility, the angle between the second mattress 2 and the third mattress 5 and the first mattress 4 can be adjusted in advance, allowing the patient to lean comfortably against the first mattress 4, the second mattress 2, and the third mattress 5. The angle can then be adjusted via the hinge assembly, further enhancing patient comfort.

[0079] After the angles of the first mattress 4, the second mattress 2, and the third mattress 5 are adjusted to a fixed position, the pressure of the patient on the flexible layer 14 of the first mattress 4, the second mattress 2, and the third mattress 5 tends to stabilize. Based on the patient's weight estimated by the image processing system, the electric telescopic rod 23 is activated to lift the lifting plate 18. At this time, the flexible layer 14 is indented due to the pressure from different parts of the patient, pressing down on the flexible layer 14 and causing the sleeve 16 to move downward. During the lifting process of the lifting plate 18, some of the movable blocks 19 enter the interior of the downward-moving sleeve 16, causing the contact point 26 to slide at different distances on the slider 25. When the contact 26 slides from top to bottom on the slider 25, it changes the resistance value in the circuit of the bulb 15, and the current in the circuit decreases. The brightness of the bulb 15 depends on the amount of current passing through it. Therefore, the decrease in current will cause the bulb 15 to dim. When the contact 26 slides to the top of the slider 25, the resistance value connected to the circuit is at its maximum. At this time, the current in the circuit decreases to a level that is insufficient to light up the bulb 15, and the bulb 15 goes out. Based on the brightness of the bulb 15, the image processing system quickly captures the amount of current passing through the bulb 15, thereby quickly obtaining the patient's body contour and achieving accurate positioning.

[0080] Furthermore, due to the different weights of different patients, the electric telescopic rod 23 drives the lifting plate 18 to rise to different heights based on the pre-input patient weight information. When the patient is heavy, the flexible layer 14 is recessed, causing the sleeve 16 to move down a greater distance and the lifting plate 18 to rise a shorter path. When the patient is light, the flexible layer 14 is recessed, causing the sleeve 16 to move down a smaller distance and the lifting plate 18 to rise a longer path. This design ensures that regardless of the difference in patient weight, the lifting plate 18 can always send the movable block 19 on the telescopic rod 21 into the sleeve 16. Moreover, since the rising distance of the lifting plate 18 is inversely proportional to the patient's weight, the relative displacement distance between the contact point 26 and the slider 25 tends to be consistent regardless of the patient's weight. This reduces the judgment gap of the image processing system when acquiring image information and enables the patient's body contour to be drawn more quickly.

[0081] Furthermore, when the patient exerts pressure on the first mattress 4, the second mattress 2, and the third mattress 5, the flexible layer 14 around the body will also indent. At this time, the change in brightness between the bulb 15 and the sensing component is a decrease in brightness. This peripheral part of the bulb 15 is used as the outline of the patient's clothing to improve the error tolerance of the scanning device when scanning the patient's body. By comparing the magnitude of the current passing through the bulb 15 at the part of the patient's body that is pressed down, if the image processing system determines that the bulb 15 needs to be repositioned by medical personnel, the repositioning will be reset according to the subjective judgment of the medical personnel, further improving the accuracy of the scan.

[0082] Even minute displacements can lead to blurred images or diagnostic errors during scanning. During scanning, when the patient shifts, the pressure at a certain point decreases or disappears. The restoring force of spring 20 resets the telescopic rod 21, and contact 26 moves accordingly. This allows bulb 15 to quickly sense the current in the circuit and light up. Conversely, when the patient shifts and presses down on another flexible layer 14, bulb 15 dims or turns off. The image processing system can quickly detect these changes in bulb 15 and, by correcting the patient's position in real time, ensure that patient displacement information is acquired throughout the scanning process, allowing medical staff to intervene and reposition the patient. This keeps the patient's position static and accurate, thus improving the accuracy of the image scanning process. If the patient moves frequently during scanning, multiple scans may be necessary to obtain a clear image. Timely position correction reduces scan interruptions and repetitions caused by patient movement, significantly shortening scanning time and improving medical efficiency.

[0083] By utilizing historical scan data from the scanning equipment and the lamp status records from the image processing system, a mathematical model was established relating the number of lamps that lit up and went off during patient displacement to image artifacts. A correction algorithm was derived based on the correlation analysis between displacement and artifacts. This algorithm aims to compensate for or eliminate artifacts during the image reconstruction stage, thereby improving image quality. The image processing system displays a comparison of images before and after correction on the scanning interface, providing real-time feedback of the artifact correction results to medical staff. Based on this feedback, medical staff can directly apply the artifact-compensated or eliminated image results for diagnosis, improving diagnostic accuracy. If necessary, the patient's position or scanning parameters can be further adjusted to ensure that the final acquired image quality meets diagnostic requirements.

[0084] The scanning device scans a specific area of ​​the patient to be examined at a corresponding level, obtains cross-sectional image information of the area to be examined, and transmits it to the image processing system.

[0085] Example 2:

[0086] As attached Figure 7 As shown, the difference from Embodiment 1 is that the image processing system includes a control module, a receiving module, a positioning module, a warning module, a processing module, and an output module.

[0087] The control module is used to control the operation of the motor, the electric telescopic pole 23, and the scanning equipment.

[0088] The receiving module is used to receive the brightness signal of the bulb 15 and then transmit it to the positioning module; the receiving module is also used to receive the image information after the scanning device scans the patient's lesion.

[0089] The positioning module is used to obtain the patient's body contour based on the signal that the bulb 15 inside the first mattress 4, the second mattress 2, and the third mattress 5 is constantly lit. Based on the signal of the brightness and extinguishing of the bulb 15 inside the first mattress 4, the second mattress 2, and the third mattress 5, and the current change of the bulb 15, the module further obtains the part of the patient's body that is pressed down on the mattress, and thus obtains the patient's position.

[0090] The warning module is used to monitor patient displacement during scanning and, if necessary, issue an alert and notify medical staff or the patient to adjust their position to continue the scanning.

[0091] The processing module is used to process the patient lesion information obtained by the receiving module through a deep convolutional neural network to perform image analysis and identify the type and location of the patient's lesion.

[0092] The processing module includes a preprocessing unit, a segmentation unit, and a postprocessing unit.

[0093] The preprocessing unit is used to employ a global histogram equalization algorithm to improve image brightness, reduce the grayscale value interval, and equalize the grayscale values.

[0094] The segmentation unit is used to segment case images. It uses a double convolution operation for upsampling and downsampling, directly connecting the input and output. The 1×1 convolution connected to the original feature map in the original module is moved to the 1×1 convolution operation on the right path of the module to further increase the receptive field.

[0095] The post-processing unit is used to randomly initialize segmentation labels based on the MRF algorithm after the segmentation unit obtains the segmentation results of the case images, and then aggregates similar pixels in the segmented case image information to eliminate noise in the segmented case images.

[0096] The output module is used to output the image information processed by the processing module, so that it can be presented to doctors intuitively for medical image diagnosis.

[0097] The specific implementation process is as follows: When the patient lies on the first mattress 4, the control module can adjust the angle between the second mattress 2 and the third mattress 5 and the first mattress 4 by controlling the rotation angle of the first connecting block 8 and the connecting rod 9 on the motor according to the patient's body part to be detected, thus achieving preliminary positioning. In addition, the control module can also adjust the lifting plate 18 to different heights by controlling the electric telescopic rod 23 according to the pre-input patient weight data. The receiving module can receive the signal of the bulb 15 turning on and off and the current change through the bulb 15, and transmit it to the positioning module. The positioning module obtains the patient's body contour and pressing part based on the constant light and light-off signals of the bulb 15 in the mattress and the current change through the bulb 15, and determines the patient's position. The warning module can issue an early warning in time when the patient moves and notify medical staff or the patient to adjust their posture, which helps to ensure the accuracy of the scanning process and avoid scanning errors caused by patient movement.

[0098] After acquiring the patient's cross-sectional image information, the scanning device transmits the image information to the receiving module, and then to the processing module. The preprocessing unit performs histogram equalization within a rectangular region around the currently processed pixel. This increases local contrast and displays details in smooth areas by using a transformation function proportional to the cumulative distribution function around the pixel. Compared to traditional histogram equalization algorithms, the preprocessing unit restricts the histogram of each sub-block. Before calculating the transformation function and the cumulative distribution function around the pixel, it uses a predefined threshold to crop the histogram, limiting the magnification factor and the slope of both the transformation function and the cumulative distribution function around the pixel. This increases image contrast while suppressing image noise.

[0099] Traditional deep convolutional neural network image processing techniques are prone to gradient vanishing and gradient exploding problems. The segmentation unit uses a double convolution operation with upsampling and downsampling, directly connecting the input and output. The 1×1 convolution connected to the original feature map in the original module is moved before the 1×1 convolution operation on the right path of the module, further increasing the receptive field. Let d... (l -1) and d （l） Let d be the input and output of the l-th layer, respectively. （0） For input, d （L） For the output of the last layer L, and These represent the sizes of the input and output graphs for the l-th layer, respectively. and For convolution operations with kernels of 1×1 and 3×3 respectively, the expression for the convolutional layer of the segmentation unit from input to output is:

[0100] ;

[0101] in For channel stacking operations, + indicates an addition operation.

[0102] When segmenting image information, assume the input image is 512. A value of 512 is used in the subsampling layer with mean pooling, and the output value is... The formula for calculating the element (x, y) of the j-th feature map layer l in the subsampling layer is:

[0103] ;

[0104] Where s is the subsampling factor, and 0 ≤ x, y < 0. , Let be the bias element of the j-th output feature map of the l-th layer, and let the subsampling layer have the same number of output and input images.

[0105] After image segmentation by the segmentation unit, pixel rasterization occurs. The post-processing unit uses Markov random field (MRF) to post-process the segmentation results, thereby optimizing the pixel values ​​in the low-resolution image information and improving the image clarity before inputting it to the output module.

[0106] The output module visually presents the processed and annotated case images to the doctor, helping them quickly assess the case and improve efficiency. Furthermore, the output module generates a "positional baseline model" and stores it in the system database. During postoperative follow-up, the patient simply lies in the same position as in the initial scan. The positioning module automatically adjusts the hinge component angle by comparing the current on / off state of bulb 15 with the baseline model, ensuring the follow-up position matches the initial scan height, thus eliminating image contrast errors caused by positional differences. For example, during a follow-up examination of a patient after tibial fracture surgery, the system can adjust the third mattress 5 to the same flexion angle as in the initial scan based on stored positional data, ensuring complete correspondence between the tibial cross-sections in the two scans, facilitating the doctor's visual observation of callus growth.

[0107] Example 3:

[0108] As attached Figure 6 As shown, the difference from Embodiment 2 is that a scanning method for intelligent medical image processing includes the following steps:

[0109] Step 1, Positioning and Fixation: Before the patient lies down, medical staff input the fracture site to be examined through the image processing system. The system provides a preset orthopedic positioning plan based on the examination site and automatically adjusts the initial angles of the first mattress 4, the second mattress 2, and the third mattress 5. The bulbs 15 inside the first mattress 4, the second mattress 2, and the third mattress 5 are powered on and kept constantly lit. The scanning bed is initially folded into a chair shape. The patient places the first mattress 4, places their hands on the armrests 3, and places their legs on the footrests 6. The motor is started through the control module according to the area to be scanned to adjust the angle between the first mattress 4 and the second and third mattresses 5.

[0110] Step 2, Sensing and Positioning: The flexible layer 14 indents due to pressure from different parts of the patient, thereby pushing the sleeve 16 downward. At the same time, the control module adjusts the lifting height of the lifting plate 18 according to the pre-input patient weight information. At this time, some of the movable blocks 19 enter the lowered sleeve 16, causing the contact point 26 to slide at different distances on the slider 25. As the sliding displacement of the contact point 26 and the slider 25 increases, the bulb 15 gradually turns from bright to dark. At this time, the receiving module acquires the brightness information of the bulb 15 and converts it into a two-dimensional image. The flexible layer 14, which is always on, is used as the part of the flexible layer 14 that is not touched by the patient, to depict the patient's body contour. The parts of the bulb 15 that are dimmed or turned off, as well as the parts of the bulb 15 where the current changes, are used as the patient's body position. The scanning device is controlled to scan the patient's body according to the patient's body contour map in the image processing system.

[0111] When the patient's body shifts during the scan, the system immediately corrects the patient's position by monitoring the on / off state of bulb 15 and changes in current. It acquires the patient's displacement information and notifies medical staff to intervene and reposition the patient. When the warning module of the image processing system issues a warning, it notifies medical staff to assist the patient in adjusting their position.

[0112] Step 3, Scanning Processing: A specific area of ​​the patient to be examined is scanned using a scanning device at the corresponding level to obtain cross-sectional image information of the area to be examined and transmit it to the image processing system. The preprocessing unit preprocesses the image information and segments the image information based on a convolutional neural network algorithm to identify and label the lesion area. After the segmented image is processed by the post-processing unit, it is transmitted externally, and the doctor performs medical image diagnosis with the assistance of intelligent annotation in the image processing system.

[0113] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A scanning device for intelligent medical image processing, characterized in that, The scanning device includes a scanning device and a scanning bed. The scanning bed includes a bed frame. The scanning device is signal-connected to an image processing system. A first mattress (4) is fixedly connected to the top of the bed frame. Hinges are fixedly connected to both ends of the bottom of the first mattress (4). A second mattress (2) and a third mattress (5) are fixedly connected to the other end of the hinges respectively. The hinges are signal-connected to the image processing system. The hinges are driven by the image processing system to adjust the angle between the first mattress (4) and the second mattress (2), and between the first mattress (4) and the third mattress (5). The first mattress (4), the second mattress (2), and the third mattress (5) are each divided into a flexible layer (14) and a sensing layer (17) from top to bottom. Each flexible layer (14) is equipped with several light bulbs (15). Each sensing layer (17) is equipped with a lifting component and a sensing component. The top of the lifting component is fixedly connected to the sensing component. The lifting component lifts the sensing component upward according to the patient's weight estimated by the image processing system. The sensing component and the light bulbs (15) are electrically connected. The sensing component is used to sense the pressure generated by the patient lying on the first mattress (4), the second mattress (2), and the third mattress (5) and change the current through the light bulbs (15) according to the pressure, so that the scanning device can acquire the patient's body contour and perform positioning scanning.

2. The intelligent medical image processing scanning device according to claim 1, characterized in that, Each hinge assembly includes several second bases (12) and a first base (7) fixedly connected to the bottom of the first mattress (4) and symmetrically arranged. The second bases (12) are fixedly connected to the bottom of the second mattress (2) and the third mattress (5) respectively. A first connecting block (8) is hinged to one side of the first base (7), and a second connecting block (13) is hinged between the first connecting block (8) and the second base (12). A connecting rod (9) is hinged to the other side of the first base (7), and a third connecting block (11) is hinged to one end of the connecting rod (9). The third connecting block (11) is divided into... The third connecting block (11) is hinged to the first connecting block (8) and the second base (12). A telescopic guide rod is hinged between the third connecting block (11) and the second connecting block (13). A tension spring (10) is sleeved on the telescopic guide rod. The two ends of the tension spring (10) are respectively hinged to the third connecting block (11) and the second connecting block (13). A drive assembly is fixedly connected to the first base (7). The drive assembly is used to drive the first connecting block (8) and the connecting rod (9) to rotate, thereby pulling or pushing the second connecting block (13) and the third connecting block (11) to drive the second base (12) to move.

3. The intelligent medical image processing scanning device according to claim 2, characterized in that, Each drive assembly includes a drive box (24) fixedly connected to the side of the first base (7) away from the first mattress (4). Each drive box (24) has a symmetrically arranged motor fixedly connected to its inner sidewall. Each motor is connected to the image processing system signal. Each motor has a transmission rod fixedly connected to its output shaft. One end of the first connecting block (8) and the connecting rod (9) are coaxially fixedly connected to the transmission rod.

4. The intelligent medical image processing scanning device according to claim 3, characterized in that, Each lifting assembly includes a lifting plate (18) and several electric telescopic rods (23) fixedly connected to the bottom wall of the sensing layer (17). The electric telescopic rods (23) are all connected to the signal of the image processing system. The output shaft of each electric telescopic rod (23) is coaxially fixedly connected to a slider (22). The sliders (22) are all fixedly connected to one side of the lifting plate (18). The inner side wall of the sensing layer (17) has a sliding groove. The sliders (22) slide in the sliding groove. The top of the lifting plate (18) is fixedly connected to the sensing assembly.

5. The intelligent medical image processing scanning device according to claim 4, characterized in that, Each sensing component includes several telescopic components fixedly connected to the top of the lifting plate (18) and sleeves (16) fixedly connected to the bottom of the flexible layer (14) and corresponding to the telescopic components. Each telescopic component has a contact (26) on one side of its top. Each sleeve (16) has a slider (25) fixedly connected to the side of the contact (26). The contact (26) and the slider (25) are slidably connected. The bulbs (15) are connected in parallel to form a parallel circuit. The parallel circuit is connected in series with a power supply. The positive and negative terminals of the power supply are electrically connected to the input and output terminals of the parallel power supply, respectively. The sliders (25) are connected in series in the circuit of the bulbs (15) and the power supply to form a closed loop. When the contact (26) contacts the slider (25), the contact (26) and the slider (25) are connected in series. When the top of the telescopic component contacts the inner top wall of the sleeve (16), it is compressed, which causes the contact (26) to always contact the slider (25) on the inner side wall of the sleeve (16). The slider (25) is electrically connected to the bulb (15) to form a closed circuit. When the contact (26) slides on the slider (25), the current flowing from the contact (26) through the slider (25) increases the resistance value in the closed circuit during the sliding process of the contact from top to bottom, which reduces the current value when passing through the bulb (15) and makes the bulb (15) dimmer.

6. The intelligent medical image processing scanning device according to claim 5, characterized in that, All telescopic components include telescopic rods (21) fixedly connected to the top of the lifting plate (18), with movable blocks (19) fixedly connected to the top of each telescopic rod (21), and contact points (26) fixedly connected to one side of the movable blocks (19). Springs (20) are sleeved on the outside of each telescopic rod (21), with the two ends of the springs (20) fixedly connected to the bottom of the movable blocks (19) and the top of the lifting plate (18) respectively.

7. The intelligent medical image processing scanning device according to claim 6, characterized in that, The first mattress (4) is used to place the patient's buttocks and surrounding area. The second mattress (2) and the third mattress (5) are used to place the patient's torso and lower limbs, respectively. The first mattress (4) is fixedly connected to armrests (3) on both sides, and the third mattress (5) is fixedly connected to foot pedals (6). An air bladder is provided inside the foot pedal (6), and an air cavity is provided between the lifting plate (18) and the flexible layer (14). The air cavity is connected to the air bladder. When the lifting plate (18) is lifted upward, the air in the air cavity is compressed and discharged into the air bladder. After the air bladder expands, it is used to fix the patient's foot part located inside the foot pedal (6).

8. The intelligent medical image processing scanning device according to claim 7, characterized in that, The image processing system includes a control module, a receiving module, a positioning module, a warning module, a processing module, and an output module; The control module is used to control the operation of the motor, the electric telescopic pole (23), and the scanning equipment; The receiving module is used to receive the brightness signal of the bulb (15) and then transmit it to the positioning module; the receiving module is also used to receive the image information after the scanning device scans the lesion of the patient; The positioning module is used to obtain the patient's body contour based on the signal that the bulbs (15) inside the first mattress (4), the second mattress (2) and the third mattress (5) are constantly lit. Based on the signal of the brightness and extinguishing of the bulbs (15) inside the first mattress (4), the second mattress (2) and the third mattress (5) and the current change of the bulbs (15), the module further obtains the part of the patient's body when pressing down on the mattress, and obtains the patient's position in sequence. The warning module is used to monitor patient displacement during scanning and, if necessary, issue an alert and notify medical staff or the patient to adjust their position to continue the scanning. The processing module is used to process the patient lesion information obtained by the receiving module through a deep convolutional neural network to perform image analysis and identify the type and location of the patient's lesion; The output module is used to output the image information processed by the processing module, so that it can be presented to doctors intuitively for medical image diagnosis.

9. The intelligent medical image processing scanning device according to claim 8, characterized in that, The processing module includes a preprocessing unit, a segmentation unit, and a postprocessing unit; The preprocessing unit is used to employ a global histogram equalization algorithm to improve image brightness, reduce the grayscale value interval, and equalize the grayscale values. The segmentation unit is used to segment case images. It uses the upsampling and downsampling process of double convolution operation to directly connect the input and output. The 1×1 convolution connected to the original feature map in the original module is moved to the 1×1 convolution operation in the right path of the module to further increase the receptive field. The post-processing unit is used to randomly initialize segmentation labels based on the MRF algorithm after the segmentation unit obtains the segmentation results of the case image, and then aggregates similar pixels in the segmented case image information to eliminate noise in the segmented case image.

10. A scanning method for intelligent medical image processing, based on the intelligent medical image processing scanning device according to any one of claims 1-9, characterized in that, Includes the following steps: Step 1, Positioning and Fixation: Before the patient lies down, the medical staff inputs the fracture site to be examined through the image processing system. The system provides a preset orthopedic positioning plan according to the examination site and automatically adjusts the initial angles of the first mattress (4), the second mattress (2), and the third mattress (5). The bulbs (15) inside the first mattress (4), the second mattress (2), and the third mattress (5) are powered on to keep the bulbs (15) constantly lit. The scanning bed is initially folded like a chair. The patient places his buttocks on the first mattress (4), his hands on the armrests (3), and his legs on the foot pedals (6). The motor is started through the control module according to the area to be scanned to adjust the angle between the first mattress (4) and the second mattress (2) and the third mattress (5). Step 2, Sensing and Positioning: The flexible layer (14) is indented due to pressure from different parts of the patient, which in turn pushes the sleeve (16) downward. At the same time, the control module adjusts the lifting height of the lifting plate (18) according to the pre-input patient weight information. At this time, some of the movable blocks (19) enter the lowered sleeve (16) and make the contact point (26) slide at different distances on the slider (25). When the sliding displacement between the contact point (26) and the slider (25) increases, the bulb (15) gradually turns from bright to dark. At this time, the receiving module obtains the brightness information of the bulb (15) and converts it into a two-dimensional image. The flexible layer (14) with the bulb (15) that is always on is used as the part of the flexible layer (14) that the patient has not touched. The part of the bulb (15) that has dimmed or turned off and the part of the bulb (15) with the change in current are used as the patient's body position. The scanning device is controlled to scan the patient's body according to the patient's body contour map in the image processing system. When the patient's body shifts during the scanning process, the position of the patient is corrected in real time by the light bulb (15) turning on and off and the current change, the patient's displacement information is obtained and the medical staff are notified to intervene and reposition the patient. When the warning module issues a warning, it notifies the medical staff to assist the patient in adjusting the position. Step 3, Scanning Processing: A specific area of ​​the patient to be examined is scanned using a scanning device at the corresponding level to obtain cross-sectional image information of the area to be examined and transmit it to the image processing system. The preprocessing unit preprocesses the image information and segments the image information based on a convolutional neural network algorithm to identify and label the lesion area. After the segmented image is processed by the post-processing unit, it is transmitted externally, and the doctor performs medical image diagnosis with the assistance of intelligent annotation in the image processing system.

Images