Real-time monitoring and intelligent optimizing device for image quality of medical imaging equipment

Abstract

The invention discloses an image quality real-time monitoring and intelligent optimization device for medical imaging equipment, which belongs to the technical field of medical imaging equipment and comprises a hardware state analysis module, an image quality analysis module, an environment state analysis module, an exposure electrical state analysis module, an exposure environment analysis module and an exposure time optimization module. According to the method, the multi-dimensional state model is constructed, the equipment hardware state, the image quality, the environmental factors and the electrical parameters are comprehensively evaluated, the width-speed adaptation degree model is introduced, and finally the target exposure time is adaptively output through the exposure time optimization model. According to the method, the problems of single monitoring dimension and disjunction with an optimization link in the prior art are solved, closed-loop control from multi-dimensional state sensing to exposure parameter intelligent optimization is realized, and the radiation dose can be effectively reduced while the image quality is ensured.

Description

Technical Field

[0001] This invention belongs to the field of medical imaging equipment technology, and in particular relates to a device for real-time monitoring and intelligent optimization of image quality in medical imaging equipment. Background Technology

[0002] Medical imaging equipment, such as CT and DR, is a core tool in modern medical diagnosis, and its imaging quality directly affects the accurate detection and diagnosis of diseases. In actual clinical applications, various factors, including equipment hardware status, working environment, and exposure parameters, dynamically influence the final image quality. Therefore, how to monitor image quality in real time and intelligently optimize exposure parameters to minimize radiation dose while ensuring image quality has become an important research and development direction in the field of medical imaging equipment technology.

[0003] Currently, existing technologies for monitoring image quality are mostly limited to post-processing analysis of the image itself, such as evaluating it by calculating single indicators like noise and contrast. Regarding parameter optimization, most devices still rely on preset exposure protocols or manual adjustments by operators, lacking a systematic approach. While some research has attempted to introduce hardware status monitoring, it often provides isolated warnings, failing to deeply integrate and analyze the multi-dimensional states of hardware, environment, electrical systems, and image quality, thus preventing truly adaptive intelligent optimization.

[0004] The main shortcomings of existing technologies are: first, the lack of a comprehensive monitoring system, which fails to simultaneously consider the coupling relationship between hardware degradation, environmental interference, electrical fluctuations and image deterioration; second, the disconnect between monitoring and optimization, with most warnings being passive rather than active adjustments, and the inability to output optimal exposure parameters in real time; and third, the optimization model considering only a few factors, ignoring the adaptability of parameters such as collimation width and scanning bed movement speed in the actual environment, resulting in insufficient accuracy and robustness of optimization suggestions. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a real-time monitoring and intelligent optimization device for image quality in medical imaging equipment, thus solving the aforementioned problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a real-time image quality monitoring and intelligent optimization device for medical imaging equipment, comprising an exposure time optimization system, including:

[0007] The hardware status analysis module constructs a hardware status model based on X-ray tube temperature, anode rotation speed, and filter grid position deviation (the distance between the current filter grid position and the ideal filter grid position) and outputs hardware status coefficients.

[0008] The image quality analysis module constructs an image quality model based on the standard deviation of image noise, image resolution, and the maximum signal intensity difference within the image region, and outputs image quality coefficients.

[0009] The environmental state analysis module constructs an environmental state model based on environmental vibration amplitude, environmental temperature, and environmental dust concentration, and outputs environmental state coefficients.

[0010] The exposure electrical condition analysis module constructs an exposure electrical condition model based on tube voltage and tube current, and outputs exposure electrical condition coefficients.

[0011] The exposure environment analysis module constructs a width-speed adaptation model based on the exposure collimation and scanning bed movement speed in the exposure parameters under the environmental state coefficient, hardware state coefficient, and exposure electrical state coefficient, and outputs the width-speed adaptation.

[0012] The exposure time optimization module constructs an exposure time optimization model based on the image quality coefficient, width-speed adaptation, and baseline exposure time, and outputs the target exposure time.

[0013] Based on the above technical solutions, the present invention also provides the following optional technical solutions:

[0014] A further technical solution: The exposure time optimization model is expressed as follows:

[0015]

[0016] in, Indicates the target exposure time. This represents the current image quality coefficient. Indicates the baseline exposure time. This indicates the current width-speed adaptation. This indicates the influence factor of width-speed adaptation. For a minimal constant (e.g.) () is used to ensure that the denominator is not zero.

[0017] Further technical solution: Based on environmental state coefficients, hardware state coefficients, and exposure electrical state coefficients, the steps for constructing a width-speed adaptation model and outputting the width-speed adaptation degree in the exposure parameters, including collimation and scanning bed movement speed, are as follows:

[0018] Import the collimation width into the formula The collimation width adaptation is obtained from the data. Indicates the collimation width. Indicates the ideal collimation width. This indicates the permissible deviation from the collimation width value. This represents the width sensitivity coefficient;

[0019] Import the scanning bed movement speed into the formula The scanning bed movement speed adaptation is obtained in the process, among which, Indicates the scanning bed movement speed. This represents the ideal scanning bed travel speed. This indicates the permissible deviation from the ideal scanning bed traverse speed value. This represents the sensitivity coefficient of the scanning bed movement speed;

[0020] The current collimation width adaptation, current scanning bed movement speed adaptation, current environmental state coefficient, current hardware state coefficient, and current exposure electrical state coefficient are imported into the width-speed adaptation model to output the current width-speed adaptation. The width-speed adaptation model is expressed as follows:

[0021]

[0022]

[0023]

[0024] in, This indicates the current width-speed adaptation. Represents the actual state vector. Represents the ideal state vector. Represents the transpose of a vector. Indicates the current hardware status coefficient. This represents the current environmental state coefficient. Indicates the current exposed electrical condition coefficient. Indicates the current collimation width fit. This indicates the current scanning bed movement speed adaptation. Furthermore, the larger the value, the better the compatibility between the collimation width and the scanning bed movement speed under the current conditions.

[0025] Further technical solution: The steps for constructing an exposure electrical state model based on tube voltage and tube current and outputting exposure electrical state coefficients are as follows:

[0026] After taking the absolute difference between the tube voltage and tube current and the corresponding ideal values, the tube voltage index and tube current index are obtained by comparing the two with the corresponding allowable deviation from the ideal value.

[0027] The current tube voltage index and the current tube current index are imported into the exposure electrical state model to output the current exposure electrical state coefficient. The exposure electrical state model is expressed as follows:

[0028]

[0029] in, Indicates the current exposed electrical condition coefficient. Indicates the current tube voltage index. Indicates the current index of the tube. Represents the weight coefficient and The Furthermore, the higher the value, the better the electrical performance during exposure.

[0030] Further technical solution: The steps for constructing an environmental state model and outputting environmental state coefficients based on environmental vibration amplitude, environmental temperature, and environmental dust concentration are as follows:

[0031] The vibration index and dust concentration index are obtained by comparing the environmental vibration amplitude and environmental dust concentration with their maximum allowable values, respectively.

[0032] The ambient temperature index is obtained by comparing the absolute difference between the ambient temperature and the standard ambient temperature with the allowable deviation from the standard ambient temperature.

[0033] The current vibration index, current ambient temperature index, and current dust concentration index are imported into the environmental state model to obtain the current environmental state coefficient. The environmental state model is expressed as follows:

[0034]

[0035] in, This represents the current environmental state coefficient. Indicates the current vibration index. This indicates the current ambient temperature index. This indicates the current dust concentration index. Represents the vibration sensitivity coefficient. Indicates the environmental temperature sensitivity coefficient. The dust concentration sensitivity coefficient is represented by the following. The higher the value, the better the environmental condition.

[0036] Further technical solution: The steps for constructing an image quality model and outputting image quality coefficients based on image noise standard deviation, image resolution, and maximum signal intensity difference within the image region are as follows:

[0037] The noise index, resolution index, and signal intensity difference index are obtained by performing max-min normalization on the image noise standard deviation, image resolution, and maximum signal intensity difference within the image region.

[0038] The current noise index, current resolution index, and current signal strength difference index are imported into a preset image state model to output the current image state coefficients. The image state model is represented as follows:

[0039]

[0040] in, Indicates the current image state coefficient. Indicates the current noise level. Indicates the current resolution index. This indicates the current signal strength index. Represents the weight coefficient and ;

[0041] The current image state coefficients are imported into the image quality model to obtain the current image quality coefficients. The image quality model is represented as follows:

[0042]

[0043] in, This represents the current image quality coefficient. Represents the image quality sensitivity coefficient. Indicates the current image state coefficient. The threshold representing the image state coefficient, the The larger the value, the better the image quality.

[0044] Further technical solution: The steps for constructing a hardware state model and outputting hardware state coefficients based on X-ray tube temperature, anode rotation speed, and filter grid position deviation (the distance between the current filter grid position and the ideal filter grid position) are as follows:

[0045] The position deviation of the filter grid is obtained by performing maximum-min normalization;

[0046] The X-ray tube temperature index is obtained by processing the absolute difference between the X-ray tube temperature and the ideal temperature and then performing maximum-min normalization.

[0047] Import the anode speed into the formula Obtain the anode speed index, where, Indicates the anode rotation speed. Indicates the ideal anode rotation speed. Indicates the maximum permissible anode speed. Minimum permissible anode speed, the ;

[0048] The current X-ray tube temperature index, current position deviation index, and current anode speed index are imported into the hardware state model to obtain the current hardware state coefficients. The hardware state model is represented as follows:

[0049]

[0050] in, Indicates the current hardware status coefficient. This indicates the current X-ray tube temperature index. This indicates the current anode speed index. This indicates that the current position deviates from the index. Represents the weight coefficient and The The higher the value, the better the hardware condition.

[0051] The medical imaging equipment image quality real-time monitoring and intelligent optimization device adopts the aforementioned medical imaging equipment image quality real-time monitoring and intelligent optimization device.

[0052] This invention provides a device for real-time monitoring and intelligent optimization of image quality in medical imaging equipment, which has the following advantages compared with the prior art:

[0053] 1. This invention constructs multiple models, including hardware status, image quality, environmental status, exposure electrical status, and width-speed adaptability. For the first time, it deeply integrates and collaboratively analyzes the device hardware health, working environment, electrical parameters, scanning parameters, and image quality, comprehensively reflecting the overall status of the system and providing a precise data foundation for intelligent optimization.

[0054] 2. This invention can dynamically and adaptively adjust the target exposure time based on the real-time calculated image quality coefficient and width-speed adaptability through an exposure time optimization model, realizing a complete closed loop from "state perception" to "decision optimization", effectively reducing reliance on human experience;

[0055] 3. This invention quantifies the impact of various factors through precise models, and can intelligently recommend the lowest possible exposure time while ensuring that the image quality meets diagnostic requirements. This helps to implement the ALARA (reasonably low) principle, reduce the radiation dose to patients and equipment, and has both clinical and safety value. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating the exposure time optimization system of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0058] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0059] Please see Figure 1 According to one embodiment of the present invention, a real-time monitoring and intelligent optimization device for image quality of medical imaging equipment includes an exposure time optimization system, comprising:

[0060] The hardware status analysis module constructs a hardware status model based on X-ray tube temperature, anode rotation speed, and filter grid position deviation (the distance between the current filter grid position and the ideal filter grid position) and outputs hardware status coefficients.

[0061] The image quality analysis module constructs an image quality model based on the standard deviation of image noise, image resolution, and the maximum signal intensity difference within the image region, and outputs image quality coefficients.

[0062] The environmental state analysis module constructs an environmental state model based on environmental vibration amplitude, environmental temperature, and environmental dust concentration, and outputs environmental state coefficients.

[0063] The exposure electrical condition analysis module constructs an exposure electrical condition model based on tube voltage and tube current, and outputs exposure electrical condition coefficients.

[0064] The exposure environment analysis module constructs a width-speed adaptation model based on the exposure collimation and scanning bed movement speed in the exposure parameters under the environmental state coefficient, hardware state coefficient, and exposure electrical state coefficient, and outputs the width-speed adaptation.

[0065] The exposure time optimization module constructs an exposure time optimization model based on the image quality coefficient, width-speed adaptation, and baseline exposure time, and outputs the target exposure time.

[0066] This application enables dynamic adjustment of exposure time parameters when equipment hardware performance degrades, environmental interference intensifies, or electrical parameters fluctuate. When the filter grid position shifts due to mechanical wear, the hardware state coefficient decreases, triggering a recalculation of the adaptation, and image quality loss is compensated by optimizing the exposure time. When dust concentration increases, leading to decreased heat dissipation efficiency, changes in the environmental state coefficient prompt the system to increase exposure time redundancy in advance, preventing image noise aggravation caused by X-ray tube overheating. Through the collaborative analysis and real-time feedback of multi-dimensional state parameters, a balance between minimizing radiation dose and maintaining image quality stability is achieved.

[0067] Preferably, the steps for constructing a hardware state model and outputting hardware state coefficients based on X-ray tube temperature, anode rotation speed, and filter grid position deviation (the distance between the current filter grid position and the ideal filter grid position) are as follows:

[0068] The position deviation of the filter grid is obtained by performing maximum-min normalization;

[0069] The X-ray tube temperature index is obtained by processing the absolute difference between the X-ray tube temperature and the ideal temperature and then performing maximum-min normalization.

[0070] Import the anode speed into the formula Obtain the anode speed index, where, Indicates the anode rotation speed. Indicates the ideal anode rotation speed. Indicates the maximum permissible anode speed. Minimum permissible anode speed, the ;

[0071] The current X-ray tube temperature index, current position deviation index, and current anode speed index are imported into the hardware state model to obtain the current hardware state coefficients. The hardware state model is represented as follows:

[0072]

[0073] in, Indicates the current hardware status coefficient. This indicates the current X-ray tube temperature index. This indicates the current anode speed index. This indicates that the current position deviates from the index. Represents the weight coefficient and The The higher the value, the better the hardware condition.

[0074] The X-ray tube temperature index refers to the distance between the current position of the X-ray tube and the ideal position. This can be measured in real time using a laser rangefinder or position encoder. Normalization converts the physical displacement deviation into a standardized index, eliminating the influence of different dimensions on the evaluation. The tube temperature index is the absolute difference between the actual temperature and the ideal temperature of the X-ray tube. This can be achieved by collecting temperature data using thermocouples or infrared temperature measurement modules. Normalization reflects the degree of temperature deviation, avoiding evaluation bias caused by fluctuations in the absolute temperature value. The anode rotation speed index refers to the deviation between the actual rotation speed and the ideal rotation speed of the anode. This can be measured using a rotation speed sensor or Hall effect sensor. The rotation speed deviation is quantified using a formula and constrained within an allowable range to ensure the physical rationality of the calculation results. The hardware state model is a mathematical model that uses a weighted linear combination of the three types of hardware parameter indices. Normalized weight coefficients are used to adjust the evaluation weights of different parameters, forming a comprehensive hardware state coefficient to achieve dynamic quantitative evaluation of the hardware state.

[0075] Specifically, firstly, filter grid position deviation data is collected. After maximum-minimum normalization, a position deviation index ranging from 0 to 1 is obtained, with the index closer to 0 indicating a larger position deviation. Simultaneously, the X-ray tube temperature is measured, and the absolute difference between it and the ideal temperature is calculated. This is then normalized to convert it into an X-ray tube temperature index, with the index closer to 0 indicating a more severe temperature deviation. For the anode rotation speed, a rotation speed index is calculated based on the ratio of the difference between the actual and ideal rotation speeds to the allowable rotation speed range; the closer this index is to 1, the better the rotation speed condition. These three indices are input into a hardware status model and weighted and summed using preset weighting coefficients to output a comprehensive hardware status coefficient. This coefficient reflects the real-time operating status of the core components of the equipment. When the hardware status coefficient decreases, it indicates problems such as X-ray tube overheating, filter grid misalignment, or abnormal anode rotation speed, providing a quantitative basis for subsequent exposure parameter optimization.

[0076] Compared to existing technologies, traditional methods only provide threshold alarms for single hardware parameters, such as monitoring whether the X-ray tube temperature exceeds limits, but they lack a multi-parameter fusion evaluation model. Existing technologies also lack quantitative analysis of the dynamic offset of the filter grid position and do not consider the cumulative impact of anode rotation speed deviation on image quality. This solution, through normalization and weighted combination, transforms hardware parameters with different dimensions and physical meanings into a unified evaluation system, achieving multi-dimensional dynamic quantification of hardware status and providing standardized input parameters that can be fused for image quality optimization.

[0077] Through the above technical solution, this application solves the problem of isolated hardware status monitoring and its lack of linkage with image quality optimization. By dynamically quantifying and evaluating the combined effects of X-ray tube temperature, filter grid position, and anode rotation speed, the system can perceive the impact of hardware degradation on image quality in real time. This solution directly links the hardware status evaluation results to the exposure parameter optimization model. When a decrease in the hardware status coefficient is detected, parameters such as exposure time are automatically adjusted to compensate, thereby maintaining image quality stability.

[0078] Preferably, the step of constructing an image quality model based on the image noise standard deviation, image resolution, and maximum signal intensity difference within the image region to output the image quality coefficient is as follows:

[0079] The noise index, resolution index, and signal intensity difference index are obtained by performing max-min normalization on the image noise standard deviation, image resolution, and maximum signal intensity difference within the image region.

[0080] The current noise index, current resolution index, and current signal strength difference index are imported into a preset image state model to output the current image state coefficients. The image state model is represented as follows:

[0081]

[0082] in, Indicates the current image state coefficient. Indicates the current noise level. Indicates the current resolution index. This indicates the current signal strength index. Represents the weight coefficient and ;

[0083] The current image state coefficients are imported into the image quality model to obtain the current image quality coefficients. The image quality model is represented as follows:

[0084]

[0085] in, This represents the current image quality coefficient. Represents the image quality sensitivity coefficient. Indicates the current image state coefficient. The threshold representing the image state coefficient, the The larger the value, the better the image quality.

[0086] Among them, the noise index is a quantitative value reflecting the noise level of the image, calculated by normalizing the noise standard deviation, and is used to characterize image sharpness. The resolution index is a quantitative value reflecting the image's ability to resolve details, calculated by normalizing the resolution value, and is used to characterize image sharpness. The signal intensity difference index is a quantitative value reflecting the signal uniformity of an image region, calculated by normalizing the maximum signal intensity difference, and is used to characterize image contrast. The image state model is a linear weighted model that integrates multiple indices, using weight coefficients... The contribution of each indicator is dynamically adjusted. The image quality model refers to a non-linear function that maps image state coefficients to quality scores, specifically implemented using a sigmoid function and sensitivity coefficients. Control the rate of change in quality scores.

[0087] Specifically, the noise standard deviation, resolution, and maximum signal strength difference are first normalized, transforming them into comparable standardized indices. These three indices are then input into the image state model and summed using preset weighting coefficients. The noise index and signal strength difference index are calculated inversely to reflect the characteristic that smaller values ​​are better. Finally, a sigmoid function is used to convert the comprehensive score into an image quality coefficient. When the image state coefficient exceeds a threshold... When the quality coefficient rapidly approaches 1, it accurately reflects the image quality compliance status. This process establishes a collaborative evaluation mechanism for noise suppression, detail preservation, and signal uniformity through multi-dimensional index fusion and nonlinear mapping.

[0088] Compared to existing technologies, current methods typically use only a single noise or resolution metric for evaluation, failing to simultaneously reflect the impact of signal uniformity on diagnostic value. This proposed solution, however, effectively detects abnormal signal fluctuations within the image region by introducing the maximum signal intensity difference metric; it overcomes the shortcomings of traditional methods that ignore the interrelationships between metrics by establishing a three-metric weighted model; and it employs a sigmoid function mapping mechanism to provide clear qualification boundaries for the quality evaluation results, making it more aligned with clinical diagnostic needs compared to linear scoring methods.

[0089] Through the above technical solution, this application can comprehensively evaluate three key dimensions of image quality: noise level, detail resolution, and signal uniformity, thus solving the problem of misjudgment caused by single-index evaluation. By dynamically adjusting the importance of each index through weighting coefficients, it adapts to the differentiated image characteristic requirements of different inspection sites. Using a nonlinear mapping function, it accurately classifies image quality levels, providing a reliable quantitative basis for exposure parameter optimization and effectively avoiding overexposure or substandard image quality caused by inaccurate evaluation.

[0090] Preferably, the steps for constructing an environmental state model and outputting environmental state coefficients based on environmental vibration amplitude, environmental temperature, and environmental dust concentration are as follows:

[0091] The vibration index and dust concentration index are obtained by comparing the environmental vibration amplitude and environmental dust concentration with their maximum allowable values, respectively.

[0092] The ambient temperature index is obtained by comparing the absolute difference between the ambient temperature and the standard ambient temperature with the allowable deviation from the standard ambient temperature.

[0093] The current vibration index, current ambient temperature index, and current dust concentration index are imported into the environmental state model to obtain the current environmental state coefficient. The environmental state model is expressed as follows:

[0094]

[0095] in, This represents the current environmental state coefficient. Indicates the current vibration index. This indicates the current ambient temperature index. This indicates the current dust concentration index. Represents the vibration sensitivity coefficient. Indicates the environmental temperature sensitivity coefficient. The dust concentration sensitivity coefficient is represented by the following. The higher the value, the better the environmental condition.

[0096] Among these parameters, environmental vibration amplitude refers to the intensity of mechanical vibration in the environment in which the equipment operates. This can be achieved by using an accelerometer to collect vibration signals and calculate their effective value. This parameter reflects the degree of interference of external vibration on the stability of the imaging equipment. Ambient temperature refers to the real-time temperature value of the equipment's operating environment. This can be achieved by using a temperature sensor for periodic sampling. This parameter is used to assess the impact of temperature fluctuations on the equipment's electronic components and mechanical structure. Ambient dust concentration refers to the content of suspended particulate matter per unit volume of air. This can be achieved by using a laser dust sensor for real-time monitoring. This parameter is used to determine the risk of dust contamination to the equipment's heat dissipation system and optical components. The vibration index quantifies the degree to which the current vibration level approaches the safety threshold. The dust concentration index characterizes the current dust pollution state. The ambient temperature index reflects the degree to which the temperature deviates from ideal operating conditions. The exponential function model, by introducing a weighted sum of squared terms and exponential operations, can amplify the negative impact of abnormal environmental parameters while maintaining the nonlinear coupling relationship between the parameters.

[0097] Specifically, this technical solution uses sensors to collect real-time data on environmental vibration amplitude, temperature, and dust concentration. Vibration amplitude data, after being calculated as effective values, is compared to a preset maximum permissible vibration amplitude. When the measured value approaches the permissible threshold, the vibration index approaches 1, indicating that the vibration interference has reached a critical state. Dust concentration data is directly compared to a preset maximum permissible concentration. When the concentration exceeds the permissible value, the dust concentration index is limited to 1, ensuring the index range is controllable. Temperature data, after calculating the absolute difference from the standard temperature, is then normalized to the permissible temperature deviation, allowing the temperature index to linearly reflect the degree of temperature deviation. These three indices are input into the environmental state model, which calculates the weighted sum of the squared terms of each index and then maps it using an exponential function to generate the environmental state coefficient. The squared term processing strengthens the influence weight of parameters exceeding the standard; for example, when the vibration index reaches 0.8, its squared term contribution will increase by 28% compared to linear processing. The decay characteristic of the exponential function ensures that when any parameter approaches a critical value, the environmental state coefficient will rapidly decrease, thereby triggering the exposure parameter adjustment mechanism. The weighting coefficients can be dynamically configured according to the equipment type. For example, the vibration sensitivity coefficient can be increased for precision CT equipment, while the weighting of dust concentration can be enhanced for mobile DR equipment.

[0098] Compared to existing technologies, traditional methods typically monitor only a single environmental parameter or employ linear weighted evaluation, failing to accurately reflect the coupled effects of multiple factors. For example, some systems only monitor ambient temperature and set fixed threshold alarms, neglecting the synergistic effects of vibration and dust. This solution, by establishing a nonlinear exponential model, can not only comprehensively evaluate the combined effects of vibration, temperature, and dust, but also amplify the negative effects of abnormal parameters through squared terms. Compared to the simple threshold comparison method in existing technologies, the environmental state coefficient output by this model can continuously characterize the degree of environmental condition, providing a more refined control basis for subsequent parameter optimization. Furthermore, in existing technologies, environmental parameter monitoring and exposure control are independent, while this solution, through quantified environmental state coefficients, achieves closed-loop linkage control between environmental disturbances and exposure parameter adjustments.

[0099] Through the above technical solution, this application can quantitatively assess the impact of complex environmental factors on imaging equipment in real time, and accurately identify interference sources such as excessive vibration, abnormal temperature, and dust accumulation. When the environmental state coefficient is lower than a set threshold, the system can automatically trigger an exposure parameter compensation mechanism, such as adjusting the exposure time to compensate for image blur caused by vibration. This solution effectively solves the problem of disconnect between environmental monitoring and image quality control in traditional methods, realizes dynamic compensation for environmental interference, and avoids excessive radiation dose while ensuring image quality.

[0100] Preferably, the step of constructing the exposure electrical state model based on tube voltage and tube current and outputting the exposure electrical state coefficients is as follows:

[0101] After taking the absolute difference between the tube voltage and tube current and the corresponding ideal values, the tube voltage index and tube current index are obtained by comparing the two with the corresponding allowable deviation from the ideal value.

[0102] The current tube voltage index and the current tube current index are imported into the exposure electrical state model to output the current exposure electrical state coefficient. The exposure electrical state model is expressed as follows:

[0103]

[0104] in, Indicates the current exposed electrical condition coefficient. Indicates the current tube voltage index. Indicates the current index of the tube. Represents the weight coefficient and The Furthermore, the higher the value, the better the electrical performance during exposure.

[0105] The tube voltage index quantifies the degree to which the tube voltage deviates from the ideal state. The tube current index quantifies the degree to which the tube current deviates from the ideal state. The allowable deviation from the ideal value refers to the maximum physical quantity that the tube voltage or tube current is allowed to deviate from the ideal state, which can be obtained from the technical specifications provided by the equipment manufacturer or through experimental calibration. This parameter provides a benchmark for normalization. The weighting coefficient is a parameter used to adjust the influence of the tube voltage index and tube current index on the final electrical state coefficient. It can be obtained from empirical values ​​or through machine learning optimization. This parameter enables the model to adapt to the differences in the influence of tube voltage and tube current on electrical performance in different devices.

[0106] Specifically, during the electrical condition assessment process, the measured data of the current tube voltage and current are first acquired in real time. The absolute difference between the measured value and the preset ideal value is calculated; for example, if the ideal tube voltage is 120kV and the measured value is 118kV, the absolute difference is 2kV. This difference is then compared to the allowable deviation; for example, if the allowable deviation is 5kV, the tube voltage exponent is 2 / 5 = 0.4. Similarly, the tube current exponent is calculated, and the two exponents are substituted into an exponential function model for nonlinear fusion. The effect of larger deviations is amplified through squaring; for example, when the tube voltage exponent is 0.8, its squared term will be amplified to 0.64, significantly higher than the linear superposition effect. The weighting coefficient is configured according to the equipment characteristics; for example, in equipment emphasizing tube voltage stability, a weighting factor can be set. =0.7, =0.3. Finally, a normalized electrical state coefficient is output through an exponential function. This coefficient decreases exponentially with the increase of the deviation of the tube voltage or tube current, accurately reflecting the real-time state of electrical performance.

[0107] Compared to existing technologies, traditional methods only monitor whether the tube voltage or tube current exceeds a threshold range, failing to establish a quantitative evaluation model. Existing technologies commonly employ fixed threshold alarm mechanisms, such as triggering an alarm when the tube voltage deviates by more than ±10%, but they cannot distinguish the gradient of the impact of different deviations on electrical performance. This solution, through a normalized exponent and exponential function model, achieves continuous quantitative evaluation of the degree of deviation, accurately reflecting the cumulative impact of small deviations on system performance. Furthermore, existing technologies treat tube voltage and tube current monitoring data in isolation, while this solution establishes a synergistic effect model of their impact on electrical performance using a weighted sum of squares, which better reflects the characteristics of multi-parameter coupling in real physical systems.

[0108] Through the above technical solution, this application effectively solves the problem of inaccurate electrical performance monitoring caused by the lack of dynamic quantitative evaluation. By converting physical parameter deviations into standardized exponents, precise quantification of the deviations in tube voltage and current is achieved. Utilizing an exponential function model to fuse the influence of multiple parameters, it can sensitively reflect the real-time fluctuations in electrical performance. The introduction of weighting coefficients allows the model to optimize parameters for different equipment types, improving the system's adaptability. The final output electrical state coefficient provides a reliable quantitative basis for subsequent exposure time optimization, ensuring timely adjustment of exposure parameters to maintain image quality stability during electrical performance fluctuations.

[0109] Preferably, the step of constructing the width-speed adaptation model and outputting the width-speed adaptation degree based on the collimation and scanning bed movement speed in the exposure parameters under the environmental state coefficient, hardware state coefficient, and exposure electrical state coefficient is as follows:

[0110] Import the collimation width into the formula The collimation width adaptation is obtained from the data. Indicates the collimation width. Indicates the ideal collimation width. This indicates the permissible deviation from the collimation width value. This represents the width sensitivity coefficient;

[0111] Import the scanning bed movement speed into the formula The scanning bed movement speed adaptation is obtained in the process, among which, Indicates the scanning bed movement speed. This represents the ideal scanning bed travel speed. This indicates the permissible deviation from the ideal scanning bed traverse speed value. This represents the sensitivity coefficient of the scanning bed movement speed;

[0112] The current collimation width adaptation, current scanning bed movement speed adaptation, current environmental state coefficient, current hardware state coefficient, and current exposure electrical state coefficient are imported into the width-speed adaptation model to output the current width-speed adaptation. The width-speed adaptation model is expressed as follows:

[0113]

[0114]

[0115]

[0116] in, This indicates the current width-speed adaptation. Represents the actual state vector. Represents the ideal state vector. Represents the transpose of a vector. Indicates the current hardware status coefficient. This represents the current environmental state coefficient. Indicates the current exposed electrical condition coefficient. Indicates the current collimation width fit. This indicates the current scanning bed movement speed adaptation. Furthermore, the larger the value, the better the compatibility between the collimation width and the scanning bed movement speed under the current conditions.

[0117] Collimation width adaptation refers to the deviation between the actual collimation width and the ideal value, calculated using a Gaussian function. Specifically, an exponential decay function can be used, and the allowable deviation and sensitivity coefficient can be adjusted within a tolerance range based on different equipment models. Scanning bed speed adaptation refers to the deviation between the actual scanning bed speed and the ideal value, calculated using a Gaussian function. This can be achieved by adjusting the scanning bed speed sensitivity coefficient to adapt to the accuracy requirements of different scanning modes. Cosine similarity is a measure of the directional consistency between the actual and ideal state vectors in multidimensional space. It is achieved by the ratio of the vector dot product to the product of their magnitudes, and normalization eliminates the influence of dimensional differences in the parameters on the evaluation results.

[0118] Specifically, the fit between collimation width and scanning bed speed is calculated using a Gaussian function. When the actual parameters deviate from the ideal values, the fit decreases non-linearly. The actual state vector is composed of hardware state coefficients, environmental state coefficients, exposure electrical state coefficients, and the fit of two geometric parameters. The ideal state vector is set as an all-1 vector. By calculating the cosine similarity between the two, when each component of the actual state vector approaches 1, the fit A approaches 1, indicating that under conditions of stable hardware performance, minimal environmental interference, and accurate electrical parameters, the matching degree between collimation width and scanning bed speed is optimal. This model achieves dynamic collaborative evaluation of geometric parameters and the overall system state by integrating equipment operating status and environmental factors.

[0119] Compared to existing technologies, traditional methods only rely on fixed thresholds to determine the compatibility of collimation width and scanning bed movement speed, neglecting the impact of factors such as equipment hardware degradation and environmental vibration on parameter matching. This solution introduces multi-dimensional state collaborative analysis, incorporating dynamic factors such as filter grid offset caused by environmental vibration and anode speed fluctuations due to X-ray tube temperature changes into the evaluation system. This allows the parameter fit calculation to reflect the comprehensive impact under real-world working conditions. Existing methods that isolate the evaluation of geometric parameters cannot solve the fit problem under the coupled effects of multiple factors, while this solution achieves a normalized evaluation of multi-dimensional state vectors through cosine similarity calculation.

[0120] Through the above technical solution, this application can dynamically adjust the compatibility evaluation criteria between the collimation width and the scanning bed movement speed according to the real-time operating status of the equipment and environmental conditions, avoiding parameter mismatch caused by hardware performance degradation or environmental interference. During CT scanning, when environmental vibration causes the filter grid position to shift, the model can automatically reduce the width compatibility score and synchronously reflect changes in other state parameters through cosine similarity calculation, thereby providing accurate input parameters for exposure time optimization. This solution effectively solves the problem of parameter compatibility evaluation distortion under complex working conditions in traditional methods, improving the robustness and accuracy of exposure parameter optimization.

[0121] Preferably, the exposure time optimization model is expressed as:

[0122]

[0123] in, Indicates the target exposure time. This represents the current image quality coefficient. Indicates the baseline exposure time. This indicates the current width-speed adaptation. This indicates the influence factor of width-speed adaptation. For a minimal constant (e.g.) () is used to ensure that the denominator is not zero.

[0124] Among them, the target exposure time refers to the optimal exposure duration calculated based on real-time monitoring parameters, used to replace the preset fixed value. The image quality coefficient is a quantitative indicator reflecting the current imaging quality, used to characterize the negative feedback requirement of image quality on exposure time. The baseline exposure time refers to the factory preset or historically optimal exposure duration, serving as a benchmark reference value for dynamic adjustment. Width-speed fit refers to the degree of matching between the collimation width and the scanning bed movement speed parameters in the current environment. Specifically, it can be calculated using a cosine similarity algorithm to determine the closeness between the actual state vector and the ideal vector, used to quantify the impact of parameter fit on exposure efficiency. The minimum constant is a mathematical correction term used to prevent the denominator from being zero; specifically, it can be used as... The magnitude of the numerical values ​​is ensured by using floating-point arithmetic to guarantee computational stability.

[0125] Specifically, the model couples the current image quality coefficient with the current width-velocity fit using a nonlinear function. When the image quality coefficient is high, the denominator increases, leading to a shorter target exposure time and thus reducing radiation dose. When the width-velocity fit decreases, the correction magnitude for the exposure time is adjusted by an influence factor α. A baseline exposure time serves as the calculation benchmark, maintaining the inherent characteristics of the equipment while achieving dynamic optimization. The introduction of a minimal constant ensures that the model can still output effective values ​​under extreme conditions. During model operation, real-time acquired hardware status, environmental parameters, and image quality data are processed by each sub-model and input into the optimization formula to generate a target exposure time adapted to the current operating conditions.

[0126] Compared with existing technologies, traditional methods only make linear adjustments based on image quality or electrical parameters, without considering the impact of the compatibility between collimation width and scanning bed movement speed on exposure efficiency, and lack a multi-parameter coupled calculation mechanism. This scheme introduces width-speed compatibility as an independent variable, combined with the dynamic feedback of the image quality coefficient, to establish a nonlinear optimization model, effectively solving the problem of radiation dose and image quality imbalance caused by single-factor adjustment.

[0127] Through the above technical solution, this application achieves dynamic and precise adjustment of exposure time, automatically reducing radiation dose based on real-time operating conditions while ensuring image quality meets diagnostic requirements. The model avoids computational anomalies through mathematical correction terms, ensuring stable system operation, while preserving the inherent parameter characteristics of the equipment to prevent image quality fluctuations caused by parameter mutations.

[0128] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0129] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A device for real-time monitoring and intelligent optimization of image quality of medical imaging equipment, characterized in that, The exposure time optimization system comprises: a hardware state analysis module, which constructs a hardware state model based on an X-ray tube temperature, an anode rotating speed, and a filter position deviation, and outputs a hardware state coefficient; an image quality analysis module, which constructs an image quality model based on an image noise standard deviation, an image resolution, and a maximum signal intensity difference in an image area, and outputs an image quality coefficient; an environment state analysis module, which constructs an environment state model based on an environment vibration amplitude, an environment temperature, and an environment dust concentration, and outputs an environment state coefficient; an exposure electrical state analysis module, which constructs an exposure electrical state model based on a tube voltage and a tube current, and outputs an exposure electrical state coefficient; an exposure environment analysis module, which constructs a width-speed adaptation model based on the environment state coefficient, the hardware state coefficient, and the exposure electrical state coefficient, and outputs a width-speed adaptation degree, wherein the width-speed adaptation model is based on an exposure collimation and a scanning bed moving speed in exposure parameters; an exposure time optimization module, which constructs an exposure time optimization model based on the image quality coefficient, the width-speed adaptation degree, and a reference exposure time, and outputs a target exposure time.

2. The medical imaging device image quality real-time monitoring and intelligent optimization apparatus of claim 1, wherein, The exposure time optimization model is represented as: wherein denotes a target exposure time, denotes a current image quality coefficient, denotes a reference exposure time, denotes a current width-velocity adaptation, denotes a width-velocity adaptation influence factor, is a very small constant (e.g. ) for ensuring that the denominator is not zero.

3. The medical imaging device image quality real-time monitoring and intelligent optimization apparatus of claim 2, wherein, The step of constructing the width-speed adaptation model based on the collimation and the scanning bed moving speed in the exposure parameters under the environment state coefficient, the hardware state coefficient, and the exposure electrical state coefficient to output the width-speed adaptation degree is: The collimation width is introduced into the formula wherein the collimation width fitness is obtained in the middle, denotes the collimation width, denotes the ideal collimation width, denotes the allowed deviation collimation width value, denotes the width sensitivity coefficient; The scan bed movement speed is introduced into the formula The scan bed movement speed adaptation is obtained in the middle, wherein, The scan bed movement speed is represented, The ideal scan bed movement speed is represented, The allowed deviation from the ideal scan bed movement speed value is represented, The scan bed movement speed sensitivity coefficient is represented; The current collimation width adaptation degree, the current scanning bed moving speed adaptation degree, the current environment state coefficient, the current hardware state coefficient, and the current exposure electrical state coefficient are introduced into the width-speed adaptation model to output the current width-speed adaptation degree, wherein the width-speed adaptation model (cosine similarity formula) is represented as: wherein, represents a current width-velocity fitness, represents an actual state vector, represents an ideal state vector, represents the transpose of a vector, represents a current hardware state coefficient, represents a current environment state coefficient, represents a current exposure electrical state coefficient, represents a current collimation width fitness, represents a current scan bed movement velocity fitness, said and the greater the value the better the fitness of the collimation width and scan bed movement velocity under the current conditions.

4. The medical imaging device image quality real-time monitoring and intelligent optimization apparatus of claim 3, wherein, The step of constructing the exposure electrical state model based on the tube voltage and the tube current to output the exposure electrical state coefficient is: The absolute difference values of the tube voltage and the tube current from corresponding ideal values are taken, and the absolute difference values are processed by ratio with corresponding allowed deviation ideal values to obtain a tube voltage index and a tube current index; The current tube voltage index and the current tube current index are introduced into the exposure electrical state model to output the current exposure electrical state coefficient, wherein the exposure electrical state model is represented as: wherein represents the current exposure electrical state coefficient, represents the current tube voltage index, represents the current tube current index, represents the weight coefficient and , said and the greater the value the better the exposure electrical performance.

5. The medical imaging device image quality real-time monitoring and intelligent optimization apparatus of claim 3, wherein, The step of constructing the environment state model based on the environment vibration amplitude, the environment temperature, and the environment dust concentration to output the environment state coefficient is: The environment vibration amplitude and the environment dust concentration are processed by ratio with maximum allowed values to obtain a vibration index and a dust concentration index; The absolute difference value of the environment temperature from a standard environment temperature is processed by ratio with an allowed deviation standard environment temperature value to obtain an environment temperature index; The current vibration index, the current environment temperature index, and the current dust concentration index are introduced into the environment state model to obtain the current environment state coefficient, wherein the environment state model is represented as: wherein, represents the current environmental state coefficient, represents the current vibration index, represents the current environmental temperature index, represents the current dust concentration index, represents the vibration sensitivity coefficient, represents the environmental temperature sensitivity coefficient, represents the dust concentration sensitivity coefficient, and the and the greater the value the better the environmental state.

6. The medical imaging device image quality real-time monitoring and intelligent optimization apparatus of claim 3, wherein, The step of constructing the image quality model based on the image noise standard deviation, the image resolution, and the maximum signal intensity difference in the image area to output the image quality coefficient is: The image noise standard deviation, the image resolution, and the maximum signal intensity difference in the image area are processed by maximum-minimum normalization to obtain a noise index, a resolution index, and a signal intensity difference index; The current noise index, the current resolution index and the current signal strength difference index are introduced into a preset image state model to output a current image state coefficient, and the image state model is represented as: wherein represents a current image state coefficient, represents a current noise index, represents a current resolution index, represents a current signal strength index, represents a weight coefficient and ; The current image state coefficient is introduced into an image quality model to obtain a current image quality coefficient, and the image quality model is represented as: wherein, represents a current image quality coefficient, represents an image quality sensitivity coefficient, represents a current image state coefficient, represents an image state coefficient threshold, said and the greater the value the better the image quality.

7. The medical imaging device image quality real-time monitoring and intelligent optimization apparatus of claim 3, wherein, The steps of constructing a hardware state model based on the X-ray tube temperature, the anode rotating speed and the filter position deviation (the distance between the current filter position and the ideal filter position) to output a hardware state coefficient are as follows: The filter position deviation is subjected to maximum-minimum normalization processing to obtain a position deviation index; The X-ray tube temperature and the ideal temperature are subjected to absolute difference value processing and then subjected to maximum-minimum normalization processing to obtain a tube temperature index; Anode speed is introduced into the equation Anode speed index is obtained, wherein, Anode speed is represented, Ideal anode speed is represented, Maximum allowed anode speed is represented, Minimum allowed anode speed is represented, which ; The current tube temperature index, the current position deviation index and the current anode rotating speed index are introduced into a hardware state model to obtain a current hardware state coefficient, and the hardware state model is represented as: wherein, represents the current hardware status coefficient, represents the current tube temperature index, represents the current anode speed index, represents the current position deviation index, represents the weight coefficient and , said and the greater the value the better the hardware status.

Images