Intelligent control method and system for magnetic resonance imaging equipment

Abstract

The invention discloses an intelligent control method and system for magnetic resonance imaging equipment, and relates to the technical field of magnetic resonance imaging control, and the system reduces the demand of manual intervention through a whole-course automatic parameter adjustment and quality evaluation mechanism. Therefore, the efficiency of the scanning process is improved, and the system can be ensured to automatically optimize the acquisition parameters Po, including the scanning time To, the resolution Ro and the signal intensity So, according to actual conditions under different patients and scanning conditions, so that the image quality and the diagnosis accuracy are improved. The automatic process enables the operation to be simpler and more convenient, and human errors are also reduced. The traditional image quality standard is concerned, and the quality evaluation is carried out through the noise level Nr and the definition Cr to ensure that the image quality is more comprehensive and meticulous; the system combines a feedback mechanism and a continuous optimization module, continuously obtains image quality evaluation results and feeds back and adjusts acquisition parameters, and gradually optimizes a scanning process.

Description

Technical Field

[0001] The present invention relates to the technical field of magnetic resonance imaging control, and specifically to an intelligent control method and system for a magnetic resonance imaging device. Background Art

[0002] Magnetic resonance imaging is a non-invasive and non-radiative medical imaging technology, which is widely used in the field of clinical diagnosis, especially playing an important role in the early screening and precise positioning of diseases in the nervous system, musculoskeletal system, cardiovascular system, etc. With the development of medical imaging technology, MRI technology has also gradually moved towards the goals of high resolution, fast imaging, and higher quality images. However, traditional MRI control systems mainly rely on preset scanning parameters, such as scanning time, resolution, and signal intensity, etc., to perform image acquisition.

[0003] Currently, most MRI control systems still rely on fixed parameter settings for scanning. Although these parameters can provide basic image quality under standardized conditions, when facing different patients, different physical conditions, and different scanning parts, the situation of unsatisfactory acquired image quality often occurs. For example, an overly long scanning time may cause discomfort to the patient. Especially when long-term immobility is required, the patient is prone to anxiety and restlessness, and may even interrupt the examination due to inability to cooperate; when the resolution is insufficient, the image details may be unclear, affecting the early detection and diagnosis of diseases; insufficient signal intensity may lead to excessive image noise, resulting in low image quality and affecting the doctor's diagnostic accuracy. Therefore, how to adjust the acquisition parameters in real time according to the actual situation to ensure the optimization of image quality has become a major challenge in the current application of MRI technology.

[0004] The occurrence of this phenomenon mainly stems from the limitations of traditional MRI control systems, which cannot dynamically adjust scanning parameters according to real-time image quality feedback. Since parameter settings are usually based on fixed standards or empirical values, they cannot effectively cope with the individual differences of patients and the changes in the scanning environment during actual operation. As a result, patients may experience an uncomfortable scanning process, and even misdiagnosis or missed diagnosis may occur due to image quality problems. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides an intelligent control method and system for a magnetic resonance imaging device, which solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent control system for a magnetic resonance imaging device includes a data acquisition and preliminary analysis module, a deep learning model analysis module, a parameter adaptive adjustment module, an image resampling module, an image quality verification and quality assessment module, and a feedback mechanism and continuous optimization module; The data acquisition and preliminary analysis module acquires the original image data Ir from the magnetic resonance imaging device and conducts a preliminary quality assessment to obtain preliminary assessment indicators, including the noise level Nr and the clarity Cr; The deep learning model analysis module uses a convolutional neural network model and combines the noise level Nr and the clarity Cr to perform real-time analysis on the original image data Ir, predicts the image quality score Qe and the occurrence probability PI of potential problems of the original image data Ir, and infers the acquisition parameters Po; The parameter adaptive adjustment module uses a convolutional neural network model to automatically adjust the acquisition parameters Po of the magnetic resonance imaging device to obtain the adjusted acquisition parameters Pon; The image resampling module re-acquires the original image data Ir according to the adjusted acquisition parameters Pon to obtain the image data Iro; The image quality verification and quality assessment module verifies the quality of the image data Iro to obtain the quality assessment result Qo; The feedback mechanism and continuous optimization module judges the image quality according to the obtained quality assessment result Qo. When the image quality does not meet the standard, it returns to the deep learning model analysis module to continue adjusting the acquisition parameters Po.

[0007] Preferably, the data acquisition and preliminary analysis module includes an image data acquisition unit and a preliminary quality assessment unit; The image data acquisition unit is responsible for acquiring the original image data Ir from the magnetic resonance imaging device; The preliminary quality assessment unit conducts a preliminary assessment of the noise assessment and clarity assessment on the acquired original image data Ir to obtain the noise level Nr and the clarity Cr; The noise level Nr is obtained through the following formula: By calculating the noise level of the image, it can reflect the external interference and hardware problems during the image acquisition process; the noise level Nr represents the noise level in the image; noise usually manifests as the difference between pixel values. By calculating the difference between the image pixel and its neighboring pixel and taking the average, the noise index of the image can be obtained; if the noise level is high, it indicates that the image may be interfered and needs to be optimized in the subsequent acquisition process.

[0008] ; In the formula, n represents the number of pixels in the original image data Ir, Iri represents the value of the i-th pixel in the original image data Ir, and vIri represents the average value of the neighboring pixels of the i-th pixel in the original image data Ir; The clarity Cr is obtained through the following formula: The clarity Cr calculates the clarity of the image; the clearer the image, the greater the difference between the pixel value and the surrounding pixels, resulting in a larger numerator part and a smaller denominator difference. The higher the clarity Cr value, the better the clarity of the image; ; In the formula, μ represents the average value of all pixel values in the original image data Ir.

[0009] Preferably, the deep learning model analysis module includes a feature extraction and analysis unit and a parameter speculation unit; The feature extraction and analysis unit inputs the obtained original image data Ir, noise level Nr, and clarity Cr into the convolutional neural network model, and performs feature extraction on the original image data Ir through the convolutional layer in the convolutional neural network model, and calculates and obtains the image quality score Qe and the occurrence probability PI of potential problems; Among them, the extracted features include edge features, texture features, and shape features; The steps for obtaining the image quality score Qe are as follows: S1. Input the original image data Ir, noise level Nr, and clarity Cr into the convolutional layer of the convolutional neural network model, and extract features. Let the output of the p-th layer of the convolutional layer be F(p), and the obtained formula is: ; In the formula, Conv represents the convolutional layer operation function, W(p) represents the convolutional kernel of the p-th layer, b(p) represents the bias term, and F(p) represents the feature map of the p-th layer; S2. Perform a dimensionality reduction operation on the feature map F(p) of the p-th layer through the pooling layer to obtain the pooled feature map P(p) of the p-th layer; the pooled feature map P(p) of the p-th layer is obtained through the following formula: ; In the formula, Pool represents the pooling layer operation function; S3. Combine the extracted features through the fully connected layer, and calculate and obtain the image quality score Qe; The image quality score Qe is obtained through the following formula: ; In the formula, f represents the ReLU activation function, W(fc) represents the weight value of the fully connected layer, b(fc) represents the bias term of the fully connected layer, and fc represents the fully connected layer in the convolutional neural network model; The occurrence probability PI of potential problems is obtained through the following formula: ; In the formula, represents the Sigmoid activation function. PI represents the occurrence probability of potential problems, represents whether there are potential problems in the image, including the probability of artifacts and noise, and the value is between [0,1].

[0010] Preferably, the parameter inference unit calculates and obtains the acquisition parameter Po based on the acquired image quality score Qe and the occurrence probability PI of potential problems, where the acquisition parameter Po includes the scanning time To, the resolution Ro, and the signal strength So; The scanning time To is obtained through the following formula: ; In the formula, represents the time adjustment factor; PI represents the occurrence probability of potential problems. If the system identifies serious motion artifacts or hardware problems, the occurrence probability PI of potential problems will be relatively large. If the image quality is poor or there are problems, the system will automatically increase the scanning time; The resolution Ro is obtained through the following formula: ; In the formula, represents the resolution adjustment factor; if the image clarity is low, or the impact of potential problems is small, the system will automatically increase the resolution; The signal strength So is obtained through the following formula: ; In the formula, represents the signal adjustment factor; if the noise is high, the system will infer and increase the signal strength to improve the image quality.

[0011] Preferably, the parameter adaptive adjustment module adjusts the acquisition parameter Po of the magnetic resonance imaging device, combines it with the image quality score Qe(t) at time t and the occurrence probability PI(t) of potential problems at time t, and calculates and obtains the adjusted acquisition parameter Pon, including the adjusted scanning time Ton, the adjusted resolution Ron, and the adjusted signal strength Son; The adjusted scanning time Ton is obtained through the following formula: ; In the formula, ΔT represents the time adjustment amount; The time adjustment amount ΔT is obtained through the following formula; ; In the formula, represents the preset weight value of the occurrence probability PI(t) of potential problems at time t; The adjusted resolution Ron is obtained through the following formula: ; In the formula, ΔR represents the image quality adjustment amount; The image quality adjustment amount ΔR is obtained through the following formula; ; Wherein, C represents a constant; The adjusted signal strength Son is obtained through the following formula: ; Wherein, ΔS represents the signal strength adjustment amount; The signal strength adjustment amount ΔS is obtained through the following formula; ; Wherein, B 1 represents a constant, and B 2 represents a real number. It represents the influence weight of the noise level and potential problems on the signal strength.

[0012] Preferably, the image resampling module includes an image interpolation unit and a new image acquisition unit; The image interpolation unit adjusts the size and quality of the original image data Ir according to the adjusted scanning time Ton and the adjusted resolution Ron by using interpolation technology; the purpose of the interpolation operation is to generate a new image based on the original image according to the adjusted scanning time Ton and the adjusted resolution Ron to improve the accuracy and quality of the image; The size of the original image data Ir is adjusted by the adjusted resolution Ron. The size of the original image data Ir is (Wr, Hr). Through the adjustment of the adjusted resolution Ron, the new image size obtained is (Wro, Hro); Wherein, Wro = Ron * Wr, Hro = Ron * Hr; For the pixel points in the original image data Ir, the adjusted image data Irn(x, y) is obtained through bicubic interpolation; The details of the adjusted image data Irn(x, y) are adjusted by the adjusted scanning time Ton. Specifically, the details of the interpolated adjusted image data Irn(x, y) are enhanced through a filter. The enhancement formula is as follows: The adjustment of the scanning time can be controlled by reconstructing the details during the interpolation process of the image data; a longer scanning time usually means more data acquisition points, thus bringing higher image accuracy; for this effect, the details of the interpolated image can be smoothed and enhanced; ; Wherein, G represents the enhancement coefficient, and Filter represents the smoothing filter.

[0013] Preferably, the new image acquisition unit adjusts the brightness and contrast of the adjusted image data Irn(x, y) through the adjusted signal strength Son to obtain the image data Iro; The change in signal strength directly affects the brightness and contrast of the image; when the adjusted signal strength Son increases, the brightness of the image will increase; while when the signal strength decreases, the image may require contrast enhancement to compensate for the low-signal condition; The image data Iro is obtained through the following formula: ; In the formula, k 1 represents the brightness gain factor, and k 2 represents the contrast adjustment parameter; usually, the contrast is increased when the adjusted signal strength Son is low to maintain the visibility of the image details.

[0014] Preferably, the image quality verification and quality assessment module analyzes the noise characteristics, sharpness characteristics, and contrast characteristics of the image data Iro to obtain the noise assessment SNR(Iro), sharpness assessment C(Iro), and contrast assessment CD(Iro), evaluates the quality of the image data Iro, and obtains the quality assessment result Qo; The noise assessment SNR(Iro) is obtained through the following formula: ; In the formula, μIro represents the mean of the pixel values of the image data Iro, and σIro represents the standard deviation of the pixel values of the image data Iro; The sharpness assessment C(Iro) is obtained through the following formula: ; In the formula, N represents the total number of pixels of the image data Iro, represents the gradient value at the pixel point (i, j); The contrast assessment CD(Iro) is obtained through the following formula: ; In the formula, max(Iro) represents the peak value of the gray values in the image data Iro, and min(Iro) represents the valley value of the gray values in the image data Iro; The quality assessment result Qo is obtained through the following formula: ; In the formula, respectively represent the preset weight values of the noise assessment SNR(Iro), sharpness assessment C(Iro), and contrast assessment CD(Iro), and .

[0015] Preferably, the feedback mechanism and continuous optimization module compare the obtained quality assessment result Qo with the preset quality standard threshold TQo to judge the image quality and obtain the image quality status; The image quality status is obtained through the following formula: When the quality evaluation result Qo ≥ the quality standard threshold TQo, the image quality meets the standard and there is no need to adjust the acquisition parameters; When the quality evaluation result Qo < the quality standard threshold TQo, the image quality does not meet the standard, and it is necessary to adjust the acquisition parameters to generate a parameter adjustment feedback signal; The feedback signal brings the quality evaluation result Qo back to the deep learning model analysis module; the acquisition parameter Po is adjusted through the convolutional neural network model.

[0016] An intelligent control method for a magnetic resonance imaging device, comprising the following steps: Step 1: The data acquisition and preliminary analysis module acquires the original image data Ir from the magnetic resonance imaging device and conducts a preliminary quality evaluation to obtain preliminary evaluation indicators, including the noise level Nr and the clarity Cr; Step 2: The deep learning model analysis module conducts real-time analysis on the original image data Ir by using a convolutional neural network model and combining the noise level Nr and the clarity Cr, predicts the image quality score Qe and the occurrence probability PI of potential problems of the original image data Ir, and infers the acquisition parameter Po; Step 3: The parameter adaptive adjustment module automatically adjusts the acquisition parameter Po of the magnetic resonance imaging device by using a convolutional neural network model to obtain the adjusted acquisition parameter Pon; Step 4: The image resampling module re-acquires the original image data Ir according to the adjusted acquisition parameter Pon to obtain the image data Iro; Step 5: The image quality verification and quality evaluation module conducts quality verification on the image data Iro to obtain the quality evaluation result Qo; Step 6: The feedback mechanism and continuous optimization module judges the image quality according to the obtained quality evaluation result Qo. When the image quality does not meet the standard, it returns to the deep learning model analysis module to continue adjusting the acquisition parameter Po.

[0017] The present invention provides an intelligent control method and system for a magnetic resonance imaging device, having the following beneficial effects: (1) During the operation of the system, according to the real-time acquired image data and its quality evaluation indicators, the acquisition parameter Po including the scanning time To, the resolution Ro and the signal intensity So is automatically adjusted to ensure the maximization of the image quality of each scan. This intelligent dynamic adjustment not only eliminates the need for manual intervention, but also optimizes the parameter settings for different patients and scanning parts, thereby effectively improving the diagnostic effect.

[0018] Traditional MRI systems often result in overly long scan times or poor image quality due to the use of fixed scanning parameters, causing unnecessary physical discomfort and psychological stress to patients. By automatically adjusting parameters such as scan time and resolution, the system can optimize the scanning process according to actual needs, reduce unnecessary scan time, relieve patients' discomfort, and reduce patients' anxiety.

[0019] (2) By introducing a data acquisition and preliminary analysis module, the system can conduct a preliminary assessment of the noise level and clarity of the original image data, ensuring real-time monitoring of image quality during the acquisition process. This real-time image quality assessment can provide more accurate data support for subsequent deep learning model analysis, thereby precisely adjusting the acquisition parameters to ensure the best image quality. This innovation effectively avoids the problem that fixed parameter settings cannot adapt to different patients and scanning environments.

[0020] The deep learning model analysis module extracts features of image data through convolutional neural networks, including edge features, texture features, and shape features, which are crucial for the evaluation of image quality. The convolutional layer can automatically extract the detailed features of the image, and through the combination of feature maps and pooling layers, precisely analyze the quality of the image. Through this efficient feature extraction, the system can more comprehensively evaluate the image quality and further improve the diagnostic accuracy of the image.

[0021] (3) The system reduces the need for manual intervention through a fully automated parameter adjustment and quality assessment mechanism. This not only improves the efficiency of the scanning process but also ensures that under different patients and scanning conditions, the system can automatically optimize the acquisition parameters Po, such as scan time To, resolution Ro, and signal intensity So, thereby improving the image quality and diagnostic accuracy. The automated process makes the operation more convenient and reduces the occurrence of human errors. It not only focuses on traditional image quality standards but also conducts quality assessment through the noise level Nr and clarity Cr to ensure more comprehensive and detailed image quality; the system combines a feedback mechanism with a continuous optimization module, gradually optimizing the scanning process by continuously obtaining image quality assessment results and feedback-adjusting the acquisition parameters.

[0022] (4) The system intelligently infers the acquisition parameters Po by combining the image quality score Qe and the probability of potential problems PI. This parameter inference mechanism based on real-time image quality feedback can precisely adjust the acquisition parameters dynamically according to the specific scanning environment and patients' needs, thereby avoiding the problem of fixed parameters in traditional MRI systems that cannot be adjusted in real time.

[0023] The parameter adaptive adjustment module ensures the best image quality for each scan by fine-tuning the acquired parameter Po. By flexibly adjusting the scan time To, resolution Ro, and signal strength So, the system can adapt to different clinical needs, ensuring clear and noise-free diagnostic images while avoiding patient discomfort caused by overly long scan times.

[0024] In actual operation, the system dynamically adjusts the scan time To, resolution Ro, and signal strength So according to the occurrence probability PI of potential problems and the change in the image quality score Qe. For example, when the occurrence probability PI of potential problems is high, the system automatically increases the scan time To and resolution Ro to ensure image quality; conversely, it reduces the scan time To to optimize the efficiency of the diagnostic process. Brief Description of the Drawings

[0025] Figure 1 It is a schematic diagram of the block diagram process of the intelligent control system of a magnetic resonance imaging device according to the present invention; Figure 2 It is a schematic diagram of the steps of the intelligent control method of a magnetic resonance imaging device according to the present invention; Figure 3 It is a schematic diagram of the judgment process of the quality assessment result according to the present invention; Figure 4 It is a schematic diagram of the steps for obtaining the image quality score according to the present invention; Figure 5 It is a line schematic diagram of the statistical data chart according to the present invention. Detailed Embodiments

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. Embodiment 1

[0027] The present invention provides an intelligent control system for a magnetic resonance imaging device. Please refer to Figure 1 , which includes a data acquisition and preliminary analysis module, a deep learning model analysis module, a parameter adaptive adjustment module, an image resampling module, an image quality verification and quality assessment module, and a feedback mechanism and continuous optimization module; The data acquisition and preliminary analysis module acquires the original image data Ir from the magnetic resonance imaging device and conducts a preliminary quality assessment to obtain preliminary assessment indicators, including the noise level Nr and clarity Cr; The deep learning model analysis module uses a convolutional neural network model and combines the noise level Nr and clarity Cr to analyze the original image data Ir in real time, predict the image quality score Qe and the probability of occurrence of potential problems PI of the original image data Ir, and infer the acquisition parameters Po; The parameter adaptive adjustment module automatically adjusts the acquisition parameter Po of the magnetic resonance imaging device by using a convolutional neural network model to obtain the adjusted acquisition parameter Pon; The image resampling module re-acquires the original image data Ir according to the adjusted acquisition parameter Pon to obtain the image data Iro; The image quality verification and quality assessment module verifies the quality of the image data Iro and obtains the quality assessment result Qo; The feedback mechanism and continuous optimization module judges the image quality based on the obtained quality assessment result Qo. When the image quality does not meet the standard, it returns to the deep learning model analysis module to continue adjusting the acquisition parameter Po.

[0028] In this embodiment, by introducing a deep learning model, the system can automatically adjust the acquisition parameters Po including scanning time To, resolution Ro and signal strength So according to the real-time acquired image data and its quality evaluation index, to ensure the maximum image quality of each scan. This intelligent dynamic adjustment not only eliminates the need for manual intervention, but also optimizes parameter settings for different patients and scanning parts, thereby effectively improving the diagnostic effect.

[0029] Traditional MRI systems often use fixed scanning parameters, which often result in long scanning times or poor image quality, causing unnecessary physical discomfort and psychological stress to patients. By automatically adjusting parameters such as scanning time and resolution, the system can optimize the scanning process according to actual needs, reduce unnecessary scanning time, relieve patient discomfort, and reduce patient anxiety.

[0030] The system predicts potential image quality problems by analyzing the original image data in real time, combined with indicators such as noise level and image clarity, and automatically adjusts the acquisition parameter Po. This intelligent optimization can effectively avoid image quality problems caused by fixed parameter settings, such as image blur and excessive noise, improve image quality, and thus improve doctors' ability to diagnose and accurately locate diseases early. The intelligent control system can ensure the best quality of each image acquisition by adjusting the scanning parameters in real time, thereby reducing repeated examinations caused by unqualified image quality and reducing medical costs.

[0031] The system introduces a feedback mechanism and a continuous optimization module. By continuously obtaining the results of image quality assessment and optimizing the acquisition parameters according to the assessment results, the system's ability of self-learning and continuous optimization enables it to gradually adapt to the needs of different patients and scanning scenarios during long-term use, thereby providing personalized diagnostic support and further improving the system's intelligence level and adaptability. Through automated and intelligent control, the system reduces the intervention of doctors or operators in parameter settings, saving the time and effort of manual adjustment. Doctors can focus more on the diagnostic work, while the system can ensure the efficiency and image quality of each scan, improving the work efficiency of the entire MRI examination process. Embodiment 2

[0032] This embodiment is an explanatory description based on Embodiment 1. Please refer to Figure 1 and Figure 4 , specifically: The data acquisition and preliminary analysis module includes an image data acquisition unit and a preliminary quality assessment unit; The image data acquisition unit is responsible for acquiring the original image data Ir from the magnetic resonance imaging device; The preliminary quality assessment unit conducts a preliminary assessment of the acquired original image data Ir for noise assessment and clarity assessment, obtaining the noise level Nr and the clarity Cr; The noise level Nr is obtained through the following formula: ; In the formula, n represents the number of pixels in the original image data Ir, Iri represents the value of the i-th pixel in the original image data Ir, and vIri represents the average value of the neighboring pixels of the i-th pixel in the original image data Ir; The clarity Cr is obtained through the following formula: ; In the formula, μ represents the average value of all pixel values in the original image data Ir.

[0033] The deep learning model analysis module includes a feature extraction and analysis unit and a parameter speculation unit; The feature extraction and analysis unit inputs the acquired original image data Ir, noise level Nr, and clarity Cr into the convolutional neural network model, and extracts features from the original image data Ir through the convolutional layer in the convolutional neural network model, and calculates and obtains the image quality score Qe and the occurrence probability PI of potential problems; Among them, the extracted features include edge features, texture features, and shape features; The steps for obtaining the image quality score Qe are: S1. Input the original image data Ir, noise level Nr, and clarity Cr into the convolutional layer of the convolutional neural network model, extract features, and set the output of the p-th layer of the convolutional layer as F(p). The acquisition formula is: ; In the formula, Conv represents the convolutional layer operation function, W(p) represents the convolutional kernel of the p-th layer, b(p) represents the bias term, and F(p) represents the feature map of the p-th layer; S2. Perform a dimensionality reduction operation on the feature map F(p) of the p-th layer through the pooling layer to obtain the pooled feature map P(p) of the p-th layer; the pooled feature map P(p) of the p-th layer is obtained through the following formula: ; In the formula, Pool represents the pooling layer operation function; S3. Combine the extracted features through the fully connected layer and calculate to obtain the image quality score Qe; The image quality score Qe is obtained through the following formula: ; In the formula, f represents the ReLU activation function, W(fc) represents the weight value of the fully connected layer, b(fc) represents the bias term of the fully connected layer, and fc represents the fully connected layer in the convolutional neural network model; The occurrence probability PI of potential problems is obtained through the following formula: ; In the formula, represents the Sigmoid activation function.

[0034] In this embodiment, by introducing the data acquisition and preliminary analysis module, the system can perform a preliminary assessment of the noise level and clarity of the original image data to ensure real-time monitoring of the image quality during the acquisition process. This real-time image quality assessment can provide more accurate data support for the subsequent deep learning model analysis, so as to precisely adjust the acquisition parameters to ensure the best image quality. This innovation effectively avoids the problem that fixed parameter settings cannot adapt to different patients and scanning environments.

[0035] The deep learning model analysis module extracts features of the image data through the convolutional neural network, including edge features, texture features, and shape features, which are crucial for the evaluation of image quality. The convolutional layer can automatically extract the detailed features of the image, and through the combination of the feature map and the pooling layer, accurately analyze the quality of the image. Through this efficient feature extraction, the system can more comprehensively evaluate the image quality and further improve the diagnostic accuracy of the image.

[0036] The system extracts and analyzes features from the original image through a convolutional neural network, combines indicators such as noise level and clarity, and calculates the image quality score Qe to achieve a comprehensive evaluation of the image quality. The calculation process of this score, through the operations of convolutional layers and pooling layers, and by combining features through fully connected layers, can accurately quantify the image quality. This mechanism enables the image quality to be optimized according to specific diagnostic requirements, avoiding the defects of unstable quality and dependence on manual adjustment in traditional MRI systems.

[0037] By calculating the probability PI of potential problems occurring through a deep learning model, the system can real-time identify possible problems in the image data, such as excessive noise and blurred images, and give early warnings. Through this intelligent prediction mechanism, the system can detect potential problems in a timely manner during the image acquisition process and intervene to avoid misdiagnosis or missed diagnosis caused by substandard image quality.

[0038] The system reduces the need for manual intervention through a fully automated parameter adjustment and quality evaluation mechanism. This not only improves the efficiency of the scanning process but also ensures that under different patients and scanning conditions, the system can automatically optimize the acquisition parameters Po according to the actual situation, such as the scanning time To, resolution Ro, and signal intensity So, thereby improving the image quality and diagnostic accuracy. The automated process makes the operation more convenient and reduces the occurrence of human errors.

[0039] The system not only focuses on traditional image quality standards but also conducts quality evaluation through the noise level Nr and clarity Cr to ensure more comprehensive and detailed image quality; the system combines a feedback mechanism with a continuous optimization module. By continuously obtaining the image quality evaluation results and feedback-adjusting the acquisition parameters, it gradually optimizes the scanning process. This method based on real-time feedback and continuous optimization enables the MRI scanning system to have the ability of self-learning and continuous improvement, continuously improving its performance during long-term use to ensure that the effect of each scan can be maximally optimized. Embodiment 3

[0040] This embodiment is an explanatory description carried out in Embodiment 2. Please refer to Figure 1 and Figure 3 Specifically: The parameter speculation unit calculates and obtains the acquisition parameter Po according to the obtained image quality score Qe and the occurrence probability PI of potential problems, where the acquisition parameter Po includes the scanning time To, resolution Ro, and signal intensity So; The scanning time To is obtained through the following formula: ; In the formula, represents the time adjustment factor; The resolution Ro is obtained through the following formula: ; In the formula, represents the resolution adjustment factor; The signal intensity So is obtained through the following formula: ; In the formula, represents the signal adjustment factor.

[0041] The parameter adaptive adjustment module adjusts the acquisition parameter Po of the magnetic resonance imaging device, combines it with the image quality score Qe(t) at time t and the occurrence probability PI(t) of potential problems at time t, and calculates and obtains the adjusted acquisition parameter Pon, including the adjusted scan time Ton, the adjusted resolution Ron, and the adjusted signal intensity Son; The adjusted scan time Ton is obtained through the following formula: ; In the formula, ΔT represents the time adjustment amount; The time adjustment amount ΔT is obtained through the following formula; ; In the formula, represents the preset weight value of the occurrence probability PI(t) of potential problems at time t; The adjusted resolution Ron is obtained through the following formula: ; In the formula, ΔR represents the image quality adjustment amount; The image quality adjustment amount ΔR is obtained through the following formula; ; In the formula, C represents a constant; The adjusted signal intensity Son is obtained through the following formula: ; In the formula, ΔS represents the signal intensity adjustment amount; The signal intensity adjustment amount ΔS is obtained through the following formula; ; In the formula, B 1 represents a constant, B 2 represents a real number.

[0042] In this embodiment, the system intelligently infers the acquisition parameter Po by combining the image quality score Qe and the occurrence probability PI of potential problems. This parameter inference mechanism based on real-time image quality feedback can accurately adjust the acquisition parameters dynamically according to the specific scanning environment and the needs of patients, thus avoiding the problem of fixed parameters that cannot be adjusted in real time in traditional MRI systems.

[0043] The parameter adaptive adjustment module ensures the best image quality for each scan by fine-tuning the acquisition parameter Po. Through flexible adjustment of the scan time To, resolution Ro and signal strength So, the system can adapt to different clinical needs, ensure clear and noise-free diagnostic images, and avoid patient discomfort caused by excessive scan time.

[0044] In actual operation, the system dynamically adjusts the scan time To, resolution Ro, and signal strength So according to the changes in the probability of occurrence of potential problems PI and the image quality score Qe. For example, when the probability of occurrence of potential problems PI is high, the system will automatically increase the scan time To and resolution Ro to ensure image quality, otherwise it will reduce the scan time To, thereby optimizing the efficiency of the diagnostic process.

[0045] By fine-tuning the time To, resolution Ro, and signal strength So, the system can significantly reduce unnecessary waiting time during the scanning process while ensuring image quality. Excessive scanning time not only increases the patient's physical burden, but may also affect the efficiency of the overall scanning process. The optimization adjustment of the system can effectively improve scanning efficiency, reduce patient waiting time and operator burden.

[0046] The system precisely controls the signal strength So and resolution Ro, so that each image acquisition can be performed at the most appropriate signal and resolution, avoiding noise problems caused by too low signal strength and image overexposure problems caused by too high signal strength. At the same time, adjusting the resolution Ro can also optimize according to the required level of detail, ensuring that the image has high-resolution details without wasting scanning time due to too high resolution.

[0047] The system eliminates the need for manual intervention by automatically adjusting and inferring the acquisition parameters Po. Operators no longer need to manually adjust acquisition parameters under different patients and scanning conditions, which not only improves operational efficiency but also reduces the incidence of human errors. The system's adaptive adjustment is based on the actual image quality score Qe and the probability of potential problems occurring PI, and can be fine-tuned based on real-time feedback. Each adjustment, including the time adjustment ΔT, resolution adjustment ΔR, and signal strength adjustment ΔS, is calculated based on precise algorithms to ensure that each adjustment can effectively improve image quality and avoid quality degradation or waste of resources caused by blind adjustments. Example 4

[0048] This embodiment is explained in Example 3, please refer to Figure 1 , Figure 3 and Figure 5 ,Specifically: the image resampling module includes an image interpolation unit and a new image acquisition unit; The image interpolation unit adjusts the size and quality of the original image data Ir according to the adjusted scanning time Ton and the adjusted resolution Ron using interpolation techniques; Adjust the size of the original image data Ir through the adjusted resolution Ron. The size of the original image data Ir is (Wr, Hr). After adjustment through the adjusted resolution Ron, the new image size obtained is (Wro, Hro); Among them, Wro = Ron * Wr, Hro = Ron * Hr; For the pixel points in the original image data Ir, calculate through the bicubic interpolation method to obtain the adjusted image data Irn(x, y); Adjust the details of the adjusted image data Irn(x, y) through the adjusted scanning time Ton. Specifically, enhance the details of the interpolated adjusted image data Irn(x, y) through a filter. The enhancement formula is as follows: ; In the formula, G represents the enhancement coefficient, and Filter represents the smoothing filter.

[0049] The new image acquisition unit adjusts the brightness and contrast of the adjusted image data Irn(x, y) through the adjusted signal strength Son to obtain the image data Iro; The image data Iro is obtained through the following formula: ; In the formula, k 1 represents the brightness gain factor, and k 2 represents the contrast adjustment parameter.

[0050] The image quality verification and quality assessment module analyzes the noise characteristics, sharpness characteristics, and contrast characteristics of the image data Iro to obtain the noise assessment SNR(Iro), sharpness assessment C(Iro), and contrast assessment CD(Iro), evaluates the quality of the image data Iro, and obtains the quality assessment result Qo; The noise assessment SNR(Iro) is obtained through the following formula: ; In the formula, μIro represents the mean of the pixel values of the image data Iro, and σIro represents the standard deviation of the pixel values of the image data Iro; The sharpness assessment C(Iro) is obtained through the following formula: ; In the formula, N represents the total number of pixels of the image data Iro, represents the gradient value at the pixel point (i, j); The contrast evaluation CD(Iro) is obtained by the following formula: ; Where max(Iro) represents the peak value of the gray value in the image data Iro, and min(Iro) represents the valley value of the gray value in the image data Iro; The quality evaluation result Qo is obtained by the following formula: ; Where respectively represent the preset weight values of the noise evaluation SNR(Iro), the sharpness evaluation C(Iro), and the contrast evaluation CD(Iro), and .

[0051] The feedback mechanism and continuous optimization module compare the obtained quality evaluation result Qo with the preset quality standard threshold TQo to judge the image quality and obtain the image quality status; The image quality status is obtained by the following formula: When the quality evaluation result Qo ≥ the quality standard threshold TQo, the image quality meets the standard and there is no need to adjust the acquisition parameters; When the quality evaluation result Qo < the quality standard threshold TQo, the image quality does not meet the standard, and it is necessary to adjust the acquisition parameters to generate a parameter adjustment feedback signal; The feedback signal brings the quality evaluation result Qo back to the deep learning model analysis module; the acquisition parameter Po is adjusted through the convolutional neural network model.

[0052] Specific example: The preset quality standard threshold TQo = 10.5; The obtained noise evaluation SNR(Iro) = 8.00; The sharpness evaluation C(Iro) = 25; The contrast evaluation CD(Iro) = 0.80; Preset are 0.4, 0.3, and 0.3 respectively; Calculate the quality evaluation result Qo: ; The quality evaluation result Qo = 10.94 > the quality standard threshold TQo = 10.5, indicating that the image quality meets the standard.

[0053] Table 1 ; In this embodiment, the image resampling module adaptively interpolates and adjusts the original image data Ir using the adjusted scan time Ton and the adjusted resolution Ron. By adjusting the resolution to modify the image size and combining with bicubic interpolation for detail optimization, both the size and quality of the image can be precisely matched to the current acquisition requirements. This process can flexibly adjust the image size while ensuring image quality, meeting the needs in different clinical scenarios.

[0054] Based on the image interpolation, the image detail enhancement operation enhances the details of the image through a smoothing filter and an enhancement coefficient G. This detail enhancement process optimizes the image quality, enabling the maintenance or even improvement of image clarity while adjusting the image size. For some clinical applications with high detail requirements, this optimization is particularly important, providing more refined and clear image data.

[0055] The brightness and contrast of the image are dynamically adjusted by the adjusted signal intensity Son. The brightness gain factor k1 and the contrast adjustment parameter k2 control the display effect of the image, ensuring that both the contrast and brightness of the image can match the actual scanning conditions, avoiding image distortion or the inability to clearly display important information due to inappropriate brightness or contrast. The image quality verification and quality assessment module evaluates the image quality from multiple dimensions including noise assessment SNR(Iro), clarity assessment C(Iro), and contrast assessment CD(Iro). This comprehensive quality assessment method can comprehensively measure various quality indicators of the image, ensuring accurate and reliable assessment results. This is crucial for clinical diagnosis, helping doctors quickly evaluate whether the image meets the standards and make better judgments accordingly.

[0056] In the image quality assessment, the precise assessment of noise and clarity is calculated through the pixel mean, standard deviation, and gradient value in the formula. This enables the system to accurately identify the noise level in the image and automatically adjust parameters to improve clarity, avoiding low-quality images that may affect diagnosis. In this way, the improvement of image quality not only depends on parameter adjustment but can also be automatically optimized through intelligent algorithms. The feedback mechanism and continuous optimization module automatically determine whether the image meets the standards by comparing the quality assessment results with the preset quality standard threshold TQo and generate a feedback signal based on the assessment results. If the image quality does not meet the standard, the system automatically returns to the deep learning model analysis module to continue adjusting the acquisition parameters Po. This continuous optimization feedback mechanism ensures that the best image quality can be achieved through parameter fine-tuning after each scan.

[0057] The system can optimize image quality based on real-time feedback to ensure that the final image results always meet the preset quality standard threshold TQo. This process helps eliminate image quality problems caused by unsatisfactory scanning conditions, ensures that the final image meets high standards of clinical diagnosis, and avoids misdiagnosis caused by quality problems.

[0058] Precise image acquisition and quality adjustment not only improves the accuracy of diagnosis, but also significantly improves diagnostic efficiency by reducing repeated scanning time. For patients, this also means a more efficient and comfortable diagnostic experience, avoiding repeated scanning processes caused by substandard image quality. Through automated image acquisition optimization and continuous feedback mechanisms, the system can automatically adjust parameters according to real-time image quality, avoid unnecessary repeated scans, maximize the use efficiency of magnetic resonance imaging equipment, avoid waste of equipment resources, and thus improve the operational efficiency of the hospital. Example 5

[0059] An intelligent control method for magnetic resonance imaging equipment, please refer to Figure 2 , specifically: including the following steps: Step 1: The data acquisition and preliminary analysis module acquires the original image data Ir from the magnetic resonance imaging device and performs a preliminary quality assessment to obtain preliminary evaluation indicators, including the noise level Nr and the clarity Cr; Step 2: The deep learning model analysis module uses a convolutional neural network model and combines the noise level Nr and the clarity Cr to perform real-time analysis on the original image data Ir, predict the image quality score Qe and the probability of occurrence of potential problems PI of the original image data Ir, and infer the acquisition parameter Po; Step 3: The parameter adaptive adjustment module automatically adjusts the acquisition parameter Po of the magnetic resonance imaging device by using the convolutional neural network model to obtain the adjusted acquisition parameter Pon; Step 4: The image resampling module re-acquires the original image data Ir according to the adjusted acquisition parameter Pon to obtain the image data Iro; Step 5: The image quality verification and quality assessment module verifies the quality of the image data Iro and obtains the quality assessment result Qo; Step 6: The feedback mechanism and continuous optimization module judges the image quality based on the obtained quality assessment result Qo. When the image quality does not meet the standard, it returns to the deep learning model analysis module to continue adjusting the acquisition parameter Po.

[0060] In this embodiment, an automated quality assessment and acquisition parameter adjustment process is achieved. After each acquisition, the system can analyze the image quality in real time and automatically adjust the acquisition parameters Po, including the scanning time To, resolution Ro, and signal intensity So, reducing the need for manual intervention and optimizing the operation process. By integrating deep learning algorithms, the system can perform real-time quality assessment on the original image data, automatically predict the occurrence probability PI of potential problems, and infer the acquisition parameters. This enables the MRI device to dynamically adjust according to the actual image quality during each scan, ensuring that the final image meets the standards and avoiding misdiagnosis or rescan caused by low-quality images.

[0061] The image resampling module re-acquires the image according to the acquisition parameters Pon after real-time adjustment to ensure effective adjustment of the image size, quality, and details. During the adjustment process, the deep learning model is used to intelligently predict and dynamically optimize the parameters to maximize the clarity and details of the image, reduce the complexity of manual adjustment, and enhance the diagnostic value of the image.

[0062] Continuous optimization and feedback mechanism: The feedback mechanism and continuous optimization function of this method enable the system to adjust the acquisition parameters based on the quality assessment results after each image acquisition. This adaptive optimization method ensures the quality improvement after each image acquisition and avoids repeated scans caused by inappropriate parameters. This not only improves the image quality but also reduces the time and waste of device resources required for repeated scans due to unqualified image quality.

[0063] The data acquisition and preliminary analysis module evaluates the noise level Nr and clarity Cr to ensure that the deep learning model analysis module analyzes based on accurate preliminary data. This step helps the system to refine the analysis of noise and clarity problems in the image and dynamically adjust the parameters according to the occurrence probability of these problems to further optimize the image quality. Through deep learning analysis and adaptive adjustment, the system can automatically optimize the acquisition parameters based on real-time feedback without manual intervention. This eliminates the need to waste time on cumbersome adjustment of scanning settings after each scan and avoids repeated scans due to non-compliant image quality, improving the overall diagnostic efficiency and shortening the patient's waiting time.

[0064] Through intelligent control and automatic optimization, the scanning efficiency of the device has been significantly improved. Repeated scans caused by improper parameter adjustment are avoided, and the utilization rate and efficiency of the device are enhanced. This not only saves time but also improves the production capacity of the device, thereby reducing the operating costs. The design of the feedback mechanism and continuous optimization module enables this intelligent control method to have the ability of continuous self-optimization. The quality assessment results after each scan provide new feedback to the system, and the system will dynamically adjust the acquisition parameters based on this feedback, gradually improving the image quality and ensuring that each scan can achieve the best results.

[0065] Through this intelligent control method, the magnetic resonance imaging device can more precisely adapt to the needs of different patients and different scanning sites, and optimize the image quality in real time. This provides doctors with higher-quality imaging data, thereby improving the accuracy and reliability of clinical diagnosis and enhancing the ability to support the early detection and diagnosis of diseases.

[0066] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent control system for a magnetic resonance imaging device, characterized in that: It includes data acquisition and preliminary analysis module, deep learning model analysis module, parameter adaptive adjustment module, image resampling module, image quality verification and quality assessment module, and feedback mechanism and continuous optimization module; The data acquisition and preliminary analysis module acquires the original image data Ir from the magnetic resonance imaging device, and performs preliminary quality assessment to obtain preliminary assessment indicators, including noise level Nr and clarity Cr; The deep learning model analysis module uses a convolutional neural network model and combines the noise level Nr and clarity Cr to analyze the original image data Ir in real time, predict the image quality score Qe and the probability of occurrence of potential problems PI of the original image data Ir, and infer the acquisition parameters Po; The parameter adaptive adjustment module automatically adjusts the acquisition parameter Po of the magnetic resonance imaging device by using a convolutional neural network model to obtain the adjusted acquisition parameter Pon; The image resampling module re-acquires the original image data Ir according to the adjusted acquisition parameter Pon to obtain the image data Iro; The image quality verification and quality assessment module verifies the quality of the image data Iro and obtains the quality assessment result Qo; The feedback mechanism and continuous optimization module judges the image quality based on the obtained quality assessment result Qo. When the image quality does not meet the standard, it returns to the deep learning model analysis module to continue adjusting the acquisition parameter Po.

2. An intelligent control system for a magnetic resonance imaging device according to claim 1, characterized in that: The data acquisition and preliminary analysis module includes an image data acquisition unit and a preliminary quality assessment unit; The image data acquisition unit is responsible for acquiring the original image data Ir from the magnetic resonance imaging device; The preliminary quality assessment unit performs preliminary assessment of noise assessment and clarity assessment on the collected original image data Ir to obtain the noise level Nr and clarity Cr; The noise level Nr is obtained by the following formula: ； Where n represents the number of pixels in the original image data Ir, Iri represents the value of the i-th pixel in the original image data Ir, and vIri represents the average value of the neighborhood pixels of the i-th pixel in the original image data Ir; The clarity Cr is obtained by the following formula: ； Wherein, μ represents the average value of all pixel values ​​in the original image data Ir.

3. An intelligent control system for a magnetic resonance imaging device according to claim 1, characterized in that: The deep learning model analysis module includes a feature extraction and analysis unit and a parameter inference unit; The feature extraction and analysis unit inputs the acquired original image data Ir, noise level Nr and clarity Cr into the convolutional neural network model, extracts features of the original image data Ir through the convolutional layer in the convolutional neural network model, and calculates and obtains the image quality score Qe and the probability of occurrence of potential problems PI; Among them, the extracted features include edge features, texture features and shape features; The steps for obtaining the image quality score Qe are: S1. Input the original image data Ir, noise level Nr and clarity Cr into the convolutional layer of the convolutional neural network model and extract features. Suppose the output of the pth layer of the convolutional layer is F(p), and the formula is: ； In the formula, Conv represents the convolution layer operation function, W(p) represents the convolution kernel of the p-th layer, b(p) represents the bias term, and F(p) represents the feature map of the p-th layer; S2. Perform dimensionality reduction operation on the feature map F(p) of the p-th layer through the pooling layer to obtain the feature map P(p) after pooling of the p-th layer. The feature map P(p) after pooling of the p-th layer is obtained by the following formula: ； In the formula, Pool represents the pooling layer operation function; S3, combining the extracted features through the fully connected layer to calculate the image quality score Qe; The image quality score Qe is obtained by the following formula: ； Where f represents the ReLU activation function, W(fc) represents the weight value of the fully connected layer, b(fc) represents the bias term of the fully connected layer, and fc represents the fully connected layer in the convolutional neural network model; The probability of occurrence of potential problems PI is obtained by the following formula: ； In the formula, Represents the Sigmoid activation function.

4. An intelligent control system for magnetic resonance imaging equipment according to claim 3, characterized in that: The parameter inference unit calculates and obtains the acquisition parameter Po according to the acquired image quality score Qe and the probability of occurrence of the potential problem PI, wherein the acquisition parameter Po includes the scanning time To, the resolution Ro and the signal strength So; The scanning time To is obtained by the following formula: ； In the formula, represents the time adjustment factor; The resolution Ro is obtained by the following formula: ； In the formula, represents the resolution adjustment factor; The signal strength So is obtained by the following formula: ； In the formula, Indicates the signal adjustment factor.

5. An intelligent control system for magnetic resonance imaging equipment according to claim 4, characterized in that: The parameter adaptive adjustment module adjusts the acquisition parameter Po of the magnetic resonance imaging device, and combines it with the image quality score Qe(t) at time t and the probability of occurrence of potential problems PI(t) at time t to calculate and obtain the adjusted acquisition parameter Pon, including the adjusted scanning time Ton, the adjusted resolution Ron and the adjusted signal strength Son; The adjusted scanning time Ton is obtained by the following formula: ； In the formula, ΔT represents the time adjustment amount; The time adjustment ΔT is obtained by the following formula: ； In the formula, The preset weight value representing the probability of occurrence of a potential problem at time t, PI(t); The adjusted resolution Ron is obtained by the following formula: ； Where ΔR represents the image quality adjustment amount; The image quality adjustment amount ΔR is obtained by the following formula; ； In the formula, C represents a constant; The adjusted signal strength Son is obtained by the following formula: ； In the formula, ΔS represents the signal strength adjustment amount; The signal strength adjustment amount ΔS is obtained by the following formula; ； In the formula, B1 represents a constant and B2 represents a real number.

6. An intelligent control system for a magnetic resonance imaging device according to claim 5, characterized in that: The image resampling module includes an image interpolation unit and a new image acquisition unit; The image interpolation unit adjusts the size and quality of the original image data Ir using interpolation technology according to the adjusted scanning time Ton and the adjusted resolution Ron; The size of the original image data Ir is adjusted by adjusting the resolution Ron. The size of the original image data Ir is (Wr, Hr). By adjusting the resolution Ron, a new image size is obtained as (Wro, Hro). Among them, Wro=Ron*Wr, Hro=Ron*Hr; For the pixel points in the original image data Ir, the bicubic interpolation method is used to calculate and obtain the adjusted image data Irn (x, y); The details of the adjusted image data Irn (x, y) are adjusted by adjusting the scan time Ton. Specifically, the details of the interpolated adjusted image data Irn (x, y) are enhanced by a filter. The enhancement formula is as follows: ； Where G represents the enhancement coefficient and Filter represents the smoothing filter.

7. An intelligent control system for a magnetic resonance imaging device according to claim 6, characterized in that: The new image acquisition unit adjusts the brightness and contrast of the adjusted image data Irn (x, y) by the adjusted signal strength Son to acquire the image data Iro; The image data Iro is obtained by the following formula: ； Where k1 represents the brightness gain factor, and k2 represents the contrast adjustment parameter.

8. An intelligent control system for a magnetic resonance imaging device according to claim 1, characterized in that: The image quality verification and quality assessment module analyzes the noise characteristics, clarity characteristics and contrast characteristics of the image data Iro, obtains the noise assessment SNR (Iro), clarity assessment C (Iro) and contrast assessment CD (Iro), evaluates the quality of the image data Iro, and obtains the quality assessment result Qo; The noise evaluation SNR (Iro) is obtained by the following formula: ； Where, μIro represents the mean value of the pixel value of the image data Iro, and σIro represents the standard deviation of the pixel value of the image data Iro; The clarity evaluation C(Iro) is obtained by the following formula: ； Where N represents the total pixels of the image data Iro, Represents the gradient value at pixel point (i, j); The contrast evaluation CD (Iro) is obtained by the following formula: ； Wherein, max(Iro) represents the peak value of the grayscale value in the image data Iro, and min(Iro) represents the valley value of the grayscale value in the image data Iro; The quality assessment result Qo is obtained by the following formula: ； In the formula, Respectively represent the preset weight values ​​of noise evaluation SNR (Iro), clarity evaluation C (Iro) and contrast evaluation CD (Iro), and .

9. An intelligent control system for a magnetic resonance imaging device according to claim 8, characterized in that: The feedback mechanism and continuous optimization module compares the obtained quality assessment result Qo with the preset quality standard threshold TQo, judges the image quality, and obtains the image quality status; The image quality status is obtained by the following formula: When the quality assessment result Qo≥quality standard threshold TQo, the image quality meets the standard and there is no need to adjust the acquisition parameters; When the quality assessment result Qo is less than the quality standard threshold TQo, the image quality does not meet the standard, and the acquisition parameters need to be adjusted to generate a parameter adjustment feedback signal; The feedback signal brings the quality assessment result Qo back to the deep learning model analysis module; the acquisition parameter Po is adjusted through the convolutional neural network model.

10. An intelligent control method for a magnetic resonance imaging device, applied to an intelligent control system for a magnetic resonance imaging device according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: The data acquisition and preliminary analysis module acquires the original image data Ir from the magnetic resonance imaging device and performs a preliminary quality assessment to obtain preliminary evaluation indicators, including the noise level Nr and the clarity Cr; Step 2: The deep learning model analysis module uses a convolutional neural network model and combines the noise level Nr and the clarity Cr to perform real-time analysis on the original image data Ir, predict the image quality score Qe and the probability of occurrence of potential problems PI of the original image data Ir, and infer the acquisition parameter Po; Step 3: The parameter adaptive adjustment module automatically adjusts the acquisition parameter Po of the magnetic resonance imaging device by using the convolutional neural network model to obtain the adjusted acquisition parameter Pon; Step 4: The image resampling module re-acquires the original image data Ir according to the adjusted acquisition parameter Pon to obtain the image data Iro; Step 5: The image quality verification and quality assessment module verifies the quality of the image data Iro and obtains the quality assessment result Qo; Step 6: The feedback mechanism and continuous optimization module judges the image quality based on the obtained quality assessment result Qo. When the image quality does not meet the standard, it returns to the deep learning model analysis module to continue adjusting the acquisition parameter Po.

Images