Positioning correction method, positioning correction device, and storage medium

Abstract

The application relates to a positioning correction method, a positioning correction device and a storage medium, wherein the positioning correction method comprises the following steps: acquiring a to-be-detected image; wherein the to-be-detected image is an image generated after positioning failure of a magnetic resonance object in magnetic resonance scanning; performing abnormal positioning analysis on the to-be-detected image to determine an abnormal positioning cause of the magnetic resonance object; and determining and outputting positioning correction information based on the abnormal positioning cause. The positioning correction method can analyze the positioning failure cause, provide a positioning correction scheme, facilitate a user to successfully complete automatic positioning based on the positioning correction information, and thus reduce the operation complexity of the magnetic resonance scanning.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of magnetic resonance scanning, and in particular, to a positioning correction method, a positioning correction device, and a storage medium. BACKGROUND

[0002] In magnetic resonance scanning, a user can accurately identify a positioning structure by means of automatic positioning technology, and then quickly select a scanning region of the magnetic resonance scanning. In the related art, after automatic positioning fails, the user can only be prompted that the automatic positioning is poor, and cannot be provided with an analysis of a cause of the automatic positioning failure and a solution. Therefore, the user can only complete the scanning by means of manual positioning after the automatic positioning fails in the current magnetic resonance scanning, which leads to high operation complexity of the magnetic resonance scanning.

[0003] Currently, there is no effective solution to the problem of high operation complexity of the magnetic resonance scanning in the related art. SUMMARY

[0004] The present application provides a positioning correction method, a positioning correction device, and a storage medium to solve the problem of high operation complexity of the magnetic resonance scanning in the related art.

[0005] In a first aspect, the present application provides a positioning correction method, comprising:

[0006] obtaining a to-be-detected image; wherein the to-be-detected image is an image generated after a positioning failure of a magnetic resonance object in magnetic resonance scanning;

[0007] performing abnormal positioning analysis on the to-be-detected image to determine an abnormal positioning cause for the magnetic resonance object;

[0008] determining and outputting positioning correction information based on the abnormal positioning cause.

[0009] In some embodiments, the abnormal positioning analysis includes at least one of the following: artifact occlusion analysis, positioning deviation analysis, and scanning posture abnormality analysis.

[0010] In some embodiments, the performing abnormal positioning analysis on the to-be-detected image to determine an abnormal positioning cause for the magnetic resonance object comprises:

[0011] performing target structure detection and artifact detection on the to-be-detected image by using a trained complete target detection model to obtain a target structure detection result and an artifact detection result;

[0012] performing artifact occlusion analysis on the to-be-detected image according to the target structure detection result and the artifact detection result to determine whether the abnormal positioning cause includes artifact occlusion.

[0013] perform positioning deviation analysis on the to-be-detected image according to the target structure detection result to determine whether the abnormal positioning cause includes positioning deviation.

[0014] In some embodiments, the abnormal positioning analysis on the to-be-detected image to determine the abnormal positioning cause of the magnetic resonance subject further includes:

[0015] performing rotation amount prediction on the to-be-detected image by using the trained complete regression model to obtain a rotation amount of the to-be-detected image in a preset direction;

[0016] performing scan posture abnormality analysis on the to-be-detected image based on the rotation amount of the to-be-detected image in the preset direction to determine whether the abnormal positioning cause includes scan posture abnormality.

[0017] In some embodiments, the artifact occlusion analysis on the to-be-detected image according to the target structure detection result and the artifact detection result to determine whether the abnormal positioning cause includes artifact occlusion includes:

[0018] in a case where the target structure detection result and the artifact detection result indicate that an overlap ratio of a target structure and an artifact in the to-be-detected image reaches a preset overlap threshold, determining that the abnormal positioning cause of the magnetic resonance subject includes artifact occlusion;

[0019] the positioning deviation analysis on the to-be-detected image according to the target structure detection result to determine whether the abnormal positioning cause includes positioning deviation includes:

[0020] in a case where the target structure detection result indicates that a deviation distance of the target structure relative to a center of the to-be-detected image reaches a preset distance threshold, determining that the abnormal positioning cause includes positioning deviation.

[0021] In some embodiments, the scan posture abnormality analysis on the to-be-detected image based on the rotation amount of the to-be-detected image in the preset direction to determine whether the abnormal positioning cause includes scan posture abnormality includes:

[0022] in a case where the rotation amount of the to-be-detected image in the preset direction reaches a preset rotation amount threshold, determining that the scan posture of the magnetic resonance subject is abnormal.

[0023] In some embodiments, the determination and output of the positioning correction information based on the abnormal positioning cause includes:

[0024] In a case where the abnormal positioning cause includes a positioning deviation, a positioning deviation amount occurring after the magnetic resonance object is positioned in the image to be detected is calculated;

[0025] Positioning correction information is output according to the positioning deviation amount.

[0026] In some embodiments, the determining and outputting of the positioning correction information based on the abnormal positioning cause further includes:

[0027] In a case where the abnormal positioning cause includes an artifact occlusion, an occlusion ratio in the image to be detected is calculated;

[0028] Positioning correction information is output according to the occlusion ratio.

[0029] In a second aspect, a positioning correction apparatus is provided in the embodiments, including: an acquisition module, a determination module, and an output module; wherein:

[0030] The acquisition module is configured to acquire an image to be detected; wherein the image to be detected is an image generated after a magnetic resonance object fails to be positioned in a magnetic resonance scan;

[0031] The determination module is configured to perform abnormal positioning analysis on the image to be detected, and determine an abnormal positioning cause for the magnetic resonance object.

[0032] The output module is configured to determine and output positioning correction information based on the abnormal positioning cause.

[0033] In a third aspect, a storage medium having a computer program stored thereon is provided in the embodiments, and the program is executed by a processor to implement the positioning correction method of the first aspect.

[0034] Compared with the related art, the positioning correction method, the positioning correction apparatus, and the storage medium provided in the embodiments can acquire an image to be detected; wherein the image to be detected is an image generated after a magnetic resonance object fails to be positioned in a magnetic resonance scan; perform abnormal positioning analysis on the image to be detected, and determine an abnormal positioning cause for the magnetic resonance object; and finally determine and output positioning correction information based on the abnormal positioning cause. The positioning correction method, the positioning correction apparatus, and the storage medium can analyze the cause of positioning failure, and provide a positioning correction scheme, so as to facilitate a user to successfully complete automatic positioning based on the positioning correction information, and thus reduce the operation complexity of the magnetic resonance scan.

[0035] Details of one or more embodiments of the present application are presented in the following drawings and description to make other features, objects, and advantages of the present application more apparent. BRIEF DESCRIPTION OF DRAWINGS

[0036] The drawings described herein are intended to provide further understanding of the present application, form a part of the present application, and are used to explain the present application and its illustrative embodiments, and do not constitute improper limitations to the present application. In the drawings:

[0037] Figure 1 is a hardware structure block diagram of a terminal of the positioning correction method of the present embodiment;

[0038] Figure 2 is a flow chart of the positioning correction method of the present embodiment;

[0039] Figure 3 is a flow chart of the abnormal positioning analysis method of the present preferred embodiment;

[0040] Figure 4 is a flow chart of the correction prompting method of the laser lamp positioning deviation of the present preferred embodiment;

[0041] Figure 5 is a flow chart of the correction prompting method of the implant artifact shielding of the present preferred embodiment;

[0042] Figure 6 is a flow chart of the correction prompting method of the scan posture abnormality of the present preferred embodiment;

[0043] Figure 7 is a structure block diagram of the positioning correction device of the present embodiment. DETAILED DESCRIPTION

[0044] In order to more clearly understand the purpose, technical scheme and advantages of the present application, the present application is described and explained below in combination with the drawings and embodiments.

[0045] Unless otherwise defined, technical terms or scientific terms used in the present application shall have the same meaning as those commonly understood by a person of ordinary skill in the art to which the present application belongs. The terms "one", "a", "an", "the", "these", and similar terms in the present application do not indicate quantity of limitation, and they can be singular or plural. The terms "include", "contain", "have", and any variant thereof in the present application are intended to cover non-exclusive inclusion; for example, a process, method, and system, product or device containing a series of steps or modules (units) are not limited to the listed steps or modules (units), but can include steps or modules (units) not listed, or can include other steps or modules (units) inherent to the process, method, product or device. The terms "connect", "connected", "couple" and similar terms in the present application are not limited to physical or mechanical connection, but can include electrical connection, whether direct or indirect. The term "multiple" in the present application refers to two or more. The term "and / or" describes the association between the associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that A exists alone, A and B exist together, and B exists alone. Generally, the character " / " represents an "or" relationship between the associated objects. The terms "first", "second", "third" and the like in the present application are only used to distinguish similar objects, and do not represent a specific order of the objects.

[0046] The method embodiments provided in the present embodiment can be executed in a terminal, a computer or a similar computing device. For example, the method embodiments are executed on a terminal, Figure 1 is a hardware structure diagram of a terminal of the positioning correction method of the present embodiment. As shown in Figure 1 , the terminal can include one or more (only one is shown in Figure 1 ) processor 102 and memory 104 for storing data, wherein the processor 102 can include but not limited to processing devices such as microprocessor MCU or programmable logic device FPGA. The above terminal can also include a transmission device 106 for communication function and an input / output device 108. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above terminal. For example, the terminal can include more or less components than those shown in Figure 1 , or have a different configuration from that shown in Figure 1 .

[0047] The memory 104 can be used to store computer programs, such as software programs of application software and modules, such as the computer program corresponding to the positioning correction method in the embodiment. The processor 102 can execute various functional applications and data processing, i.e., implement the method described above, by running the computer program stored in the memory 104. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include memories remotely arranged with respect to the processor 102, which can be connected to the terminal through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0048] The transmission device 106 is configured to receive or send data via a network. The network includes a wireless network provided by a communication provider of the terminal. In an example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In an example, the transmission device 106 can be a radio frequency (RF) module, which is configured to communicate with the Internet in a wireless manner.

[0049] In the embodiment, a positioning correction method is provided, Figure 2 The flowchart of the positioning correction method in the embodiment is shown in FIG. 2, which includes the following steps: Figure 2

[0050] In step S210, an image to be detected is acquired. The image to be detected is an image generated after a positioning failure of a magnetic resonance object in a magnetic resonance scan.

[0051] In step S220, an abnormal positioning analysis is performed on the image to be detected, and a cause of abnormal positioning of the magnetic resonance object is determined.

[0052] ​In a magnetic resonance scan of a magnetic resonance object, in order to realize a targeted scan of the magnetic resonance object and improve the scan efficiency, a part of the magnetic resonance object that needs to be scanned can be located. Thus, a positioning algorithm can be pre-set in a magnetic resonance system, which identifies a structure to be located from a positioning image to realize automatic positioning of the magnetic resonance object. In the case of automatic positioning failure, the image obtained after the positioning failure is extracted for analyzing the cause of the positioning failure. In the magnetic resonance scan scene, there are various factors affecting the positioning result, including but not limited to laser light positioning deviation, implanted objects in the magnetic resonance object causing artifacts shielding, and the pre-selected standard scan position not matching the actual scan position of the magnetic resonance object. Based on image detection technology, the cause of the current magnetic resonance scan positioning failure is determined by image analysis of the to-be-detected image generated after the positioning failure. The cause of the positioning failure is an abnormal positioning cause.

[0053] For different abnormal positioning causes, the abnormal positioning analysis method applicable thereto can be different. To improve the accuracy of abnormal positioning cause analysis, different image detection technologies can be used to analyze the to-be-detected image to determine the final abnormal positioning cause. The image detection technology can be determined according to the specific application scenario. For example, the image detection technology can include a general target positioning algorithm, a target detection algorithm, a target segmentation algorithm, or a target classification algorithm. By analyzing the to-be-detected image to determine and analyze the specific abnormal positioning situation, the operator of the magnetic resonance scan can be provided with a clear positioning failure cause, so that the operator can correct the positioning operation in the magnetic resonance scan process based on the positioning failure cause, thereby reducing the operation complexity of the magnetic resonance scan.

[0054] Preferably, the to-be-detected image can be subjected to target detection to determine whether the abnormal positioning cause is laser light positioning deviation or implanted object artifact shielding. By predicting the scan posture of the to-be-detected image, it can be determined whether the abnormal positioning cause is that the pre-selected standard scan position does not match the actual scan position of the magnetic resonance object. For example, a target recognition algorithm can also be used to determine whether the target structure to be located is detected, and whether the positioning recognition point is deformed, to determine whether the positioning failure is caused by a lesion of the target structure. The target structure can be a part of the magnetic resonance object that needs to be scanned. In addition, the stability of the algorithm, the system version or the software version can be analyzed to determine other causes of positioning failure.

[0055] In step S230, positioning correction information is determined and output based on the abnormal positioning cause.

[0056] The positioning correction information refers to information determined based on the abnormal positioning cause and indicating that the current positioning operation needs to be adjusted. Specifically, it can include offset amount information determined based on the offset distance of the laser lamp from the magnetic resonance object, artifact occlusion information output based on the artifact occlusion ratio of the implant to the magnetic resonance object, or prompt information for selecting a correct scan posture, etc. Based on the analysis of the abnormal positioning cause, the embodiment further determines the positioning correction information, which can provide clear positioning correction guidance for the operation of the magnetic resonance scan, thereby further reducing the complexity of the operation of the magnetic resonance scan.

[0057] The steps S210 to S230 above acquire the to-be-detected image; the to-be-detected image is an image generated after the positioning of the magnetic resonance object fails in the magnetic resonance scan; the to-be-detected image is subjected to abnormal positioning analysis to determine the abnormal positioning cause for the magnetic resonance object; and the positioning correction information is determined and output based on the abnormal positioning cause. It can analyze the cause of the positioning failure and provide a positioning correction scheme, thereby facilitating the user to successfully complete automatic positioning based on the positioning correction information, and further reducing the operation complexity of the magnetic resonance scan.

[0058] Further, in an embodiment, the abnormal positioning analysis includes at least one of artifact occlusion analysis, positioning offset analysis, and scan posture abnormality analysis. The artifact occlusion analysis refers to the analysis of the artifact occlusion of the metal or non-metal implant, the positioning offset analysis can include the analysis of the positioning offset of the laser lamp to the to-be-scanned part, and the scan posture abnormality analysis can specifically be the analysis of whether the actual scan posture of the magnetic resonance object is consistent with the standard scan posture selected before the scan.

[0059] Additionally, in an embodiment, based on the step S220 above, the to-be-detected image is subjected to abnormal positioning analysis to determine the abnormal positioning cause for the magnetic resonance object, which can specifically include:

[0060] The target detection model is trained to detect the target structure and the artifact in the to-be-detected image, to obtain the target structure detection result and the artifact detection result; the to-be-detected image is subjected to artifact occlusion analysis based on the target structure detection result and the artifact detection result, to determine whether the abnormal positioning cause includes artifact occlusion; and the to-be-detected image is subjected to positioning offset analysis based on the target structure detection result, to determine whether the abnormal positioning cause includes positioning offset.

[0061] The target detection model can be any network model that realizes extraction of a bounding box of a target structure in the to-be-detected image and target classification. For example, the Yolo network can be used as the target detection model in the embodiment. The target structure refers to a part of a magnetic resonance object that needs to be scanned by magnetic resonance, such as a bony structure or soft tissue. Thus, the target structure detection refers to detection of a part that needs to be scanned by magnetic resonance, such as a bony structure or soft tissue, in the to-be-detected image by using the target detection model. The target structure detection result refers to image data of the part that needs to be scanned by magnetic resonance, which can be represented by a bounding box in the to-be-detected image. The bounding box of the target structure can indicate the position information of the target structure in the to-be-detected image. The artifact detection refers to detection of a metal implant or a non-metal implant in the to-be-detected image by using the target detection model. Specifically, the artifact detection can be realized by identifying an abnormally highlighted signal in the to-be-detected image. The artifact detection result refers to image data of the detected implant in the to-be-detected image, which can also be represented by a bounding box in the to-be-detected image. It can be understood that the bounding box of the implant and the bounding box of the target structure can be associated with different categories and corresponding identification information.

[0062] After obtaining the target structure detection result and the artifact detection result, the abnormal positioning analysis of the positioning deviation and the artifact occlusion can be performed. Specifically, whether the detected target structure deviates and is incomplete in the to-be-detected image can be determined based on the position of the target structure in the to-be-detected image, so as to determine whether the abnormal positioning is caused by the positioning deviation. For example, whether the positioning deviation exists can be determined based on the distance between the center of the bounding box of the target structure and the center of the to-be-detected image, the coordinate information of the center of the bounding box of the target structure, or the positional relationship between the bounding box of the target structure and the boundary of the to-be-detected image, so as to complete the positioning deviation analysis. In addition, in the case where the target structure detection result and the artifact detection result indicate that the target structure and the artifact are detected in the to-be-detected image at the same time, the artifact occlusion analysis can be realized by comparing the overlap ratio of the two. For example, whether the artifact occlusion exists can be determined by calculating the overlapping area between the bounding box of the target structure and the bounding box of the artifact.

[0063] Furthermore, during the training of this target detection model, the network can be pre-trained for images from different scanning sites. For example, experienced clinicians can manually annotate key bony structures or tissues in different scanning sites, as well as artifacts formed by implants in the images, for network training, thereby obtaining a fully trained target detection model. During anomaly localization analysis, the image to be detected is input into the target detection model, which processes it and outputs predictions of the target structure and implant artifacts of the MRI object. Specifically, the target detection model outputs the bounding box and category of the target structure as the target structure detection result, and outputs the bounding box and category of the implant artifact as the artifact detection result. The target structure of the MRI object can be the anatomical structure, treatment structure, or examination structure of the MRI object that requires MRI scanning.

[0064] The following explanation uses an MRI scan of the ankle joint as an example. After the automatic localization of the ankle joint fails and an image to be detected is generated, this image is input into the aforementioned target detection model for ankle joint structure detection and artifact detection. This yields target detection results including the ankle joint bounding box and predicted probability, as well as artifact detection results including the bounding box and predicted probability of the implant artifact. Then, based on the information from the ankle joint bounding box and the implant artifact bounding box, localization deviation analysis and artifact occlusion analysis are performed. For example, if the distance between the ankle joint bounding box and the center of the image to be detected reaches a preset distance threshold, the cause of abnormal localization is determined to be localization deviation; and if the overlap ratio between the ankle joint bounding box and the implant artifact bounding box exceeds a preset ratio threshold, the cause of abnormal localization is determined to be artifact occlusion.

[0065] To address the issue that related technologies cannot analyze the specific reasons for positioning failures, leading to high operational complexity of magnetic resonance imaging (MRI) scans, this embodiment uses a well-trained target detection model to perform target structure detection and artifact detection on the image to be detected. Based on the detection results, it determines whether there is artifact occlusion or positioning deviation, enabling accurate analysis of the causes of abnormal positioning. It can also promptly alert and correct positioning deviations of the laser lamp or artifact occlusion of the implant, thereby reducing the operational complexity of MRI scans.

[0066] In another embodiment, based on the above step S220, performing anomaly localization analysis on the image to be detected to determine the cause of the anomaly localization for the magnetic resonance object may further include:

[0067] Using a well-trained regression model, the rotation amount of the image to be detected is predicted to obtain the rotation amount of the image to be detected in a preset direction. Based on the rotation amount of the image to be detected in the preset direction, the scanning posture anomaly analysis of the image to be detected is performed to determine whether the cause of the anomaly localization includes scanning posture anomaly.

[0068] The regression model can be any network model suitable for regression analysis of the image to be detected. For example, the regression model can be a ResNet residual network model. The aforementioned scanning posture detection can specifically be the prediction of the MRI scan position. In this embodiment, the image to be detected is input into a fully trained regression model to predict the rotation amount and determine the rotation amount of the target structure in the image to be detected in a preset direction. Here, the rotation amount refers to the rotation angle of the target structure in the image to be detected in one or more preset directions. The corresponding direction can be set by a pre-selected standard scanning position. Predicting the rotation amount of the image to be detected means predicting the rotation angle of the target structure in the image to be detected in a preset direction. For example, if the operator pre-selects a supine, head-first position as the standard scanning position of the MRI object coordinate system, the image to be detected is first transposed based on this standard scanning position. If the patient's position in the image to be detected is supine with feet first, the head-to-feet direction in the image is rotated 180 degrees. Then, the rotation amount is predicted in the transposed image. Specifically, the rotation angle of the target structure in the image relative to the standard scanning position is predicted in three preset directions: head-to-feet, front-back, and left-right. Finally, based on these rotation angles and the pre-selected standard scanning position, it is determined whether the actual scanning position in the image to be detected matches the standard scanning position, thus achieving scanning posture anomaly analysis. If the actual scanning position does not match the standard scanning position, the cause of the anomaly is considered to be abnormal scanning posture.

[0069] Continuing with the example of an ankle MRI scan, the image to be examined is first transposed based on the standard ankle scanning position. This standard scanning position ensures the sole of the foot is perpendicular to the lower leg, the ankle joint is in a neutral position with no internal or external rotation, the toes point in the same direction as the left-right axis of the image, and the normal vector of the sole aligns with the head-to-foot direction. After transposing the image based on this standard scanning position, the regression model predicts the rotation angles of the image relative to the standard ankle position image in the head-to-foot, front-to-back, and left-to-right directions, thus obtaining the predicted rotation amount. If the predicted rotation amount indicates that the ankle joint structure in the image has a rotation angle greater than a preset angle in any of these directions (e.g., greater than 90 degrees), it is identified as an abnormal scanning posture.

[0070] In this embodiment, by predicting the rotation amount of the image to be detected, it is possible to determine whether there is an abnormal scanning posture. This can accurately detect positioning failures caused by discrepancies between the actual scanning posture and the pre-scanning posture. In such cases, it provides the operator with a clear indication of the cause of the abnormal positioning, making it easier for the operator to adjust the scanning posture in a timely manner, thereby reducing the operational complexity of magnetic resonance scanning.

[0071] In one embodiment, based on the target structure detection results and artifact detection results, an artifact occlusion analysis is performed on the image to be detected to determine whether the cause of abnormal localization includes artifact occlusion. This may include: if the target structure detection results and artifact detection results indicate that the overlap ratio between the target structure and the artifact in the image to be detected reaches a preset overlap threshold, determining that the cause of abnormal localization of the magnetic resonance object includes artifact occlusion; based on the target structure detection results, a localization deviation analysis is performed on the image to be detected to determine whether the cause of abnormal localization includes localization deviation. This may include: if the target structure detection results indicate that the deviation distance between the target structure and the center of the image to be detected reaches a preset distance threshold, determining that the cause of abnormal localization is localization deviation.

[0072] In the image to be detected, if the overlap ratio of the artifacts of the target structure and the implant exceeds a certain level, it can be determined that there is a localization failure due to artifact occlusion, thus identifying artifact occlusion as a cause of abnormal localization. Specifically, based on the bounding boxes of the detected target structure and the implant artifact, the size of the overlapping region of the two bounding boxes can be calculated, and the ratio of the size of the overlapping region to the size of the target structure's bounding box is used as the overlap ratio. If this overlap ratio exceeds a preset overlap threshold, it indicates that the overlapping region of the target structure and the artifact in the image to be detected has reached the preset overlap threshold, and is therefore considered to have artifact occlusion.

[0073] Similarly, in the image to be detected, if the deviation of the target structure from the center of the image reaches a certain preset distance threshold, it is considered that there is a localization failure due to localization deviation, thus determining that the cause of abnormal localization includes localization deviation. For example, the center of the bounding box of the target structure and the center of the image to be detected can be calculated, and the distance between the center of the bounding box of the target structure and the center of the image to be detected can be compared. If the distance is greater than the preset distance threshold, it is considered that there is a localization deviation.

[0074] In this embodiment, based on preset artifact occlusion conditions, the existence of artifact occlusion is determined by combining the target structure detection structure and artifact detection results, which improves the accuracy of artifact occlusion detection. Simultaneously, the determination of positioning deviation based on the target structure detection results and positioning deviation conditions improves the accuracy of positioning deviation detection. Therefore, this embodiment can improve the accuracy of anomaly positioning analysis.

[0075] In one embodiment, based on the rotation amount of the image to be detected in a preset direction, a scanning posture anomaly analysis is performed on the image to be detected to determine whether the cause of the anomaly localization includes scanning posture anomalies, which may include:

[0076] If the rotation of the image to be detected in a preset direction reaches a preset rotation threshold, the scanning posture of the magnetic resonance imaging (MRI) object is determined to be abnormal. For example, when there are multiple preset directions, if the rotation of the image to be detected in any preset direction exceeds the preset rotation threshold, it is considered that the actual scanning position does not conform to the standard position, and the scanning posture of the MRI object is determined to be abnormal. This embodiment compares the rotation amount with a preset threshold to determine abnormal scanning posture, which can improve the accuracy of abnormal scanning posture determination.

[0077] Additionally, in one embodiment, based on the above step S230, determining and outputting positioning correction information based on the cause of the abnormal positioning may include:

[0078] When the cause of abnormal positioning includes positioning deviation, the amount of positioning deviation that occurred after locating the magnetic resonance object in the image to be detected is calculated; positioning correction information is output based on the positioning deviation amount. The positioning deviation amount can be integrated with the information on positioning deviation of the laser point into positioning correction information, and the operator can be prompted through an associated output device so that the operator can take appropriate action based on the above positioning correction information. The above output device can be a display device, audio device, etc.

[0079] Furthermore, different positioning correction information can be output based on the degree to which the laser pointer deviates from the region of interest (ROI) of the image being inspected—that is, the degree to which the laser pointer deviates from the position of the target structure in the image. If the laser pointer deviates significantly from the ROI, the target structure cannot be identified in the image, or only a small portion of the target structure can be identified. Therefore, the corresponding deviation distance cannot be calculated, and a prompt message indicating that the laser pointer positioning has deviated and needs to be repositioned can be output. If the laser pointer deviates slightly from the ROI and most of the target structure can be identified, then after calculating the deviation distance, a prompt message indicating that the laser pointer positioning needs to move forward or backward by the corresponding distance can be output.

[0080] In one embodiment, based on the above step S230, determining and outputting positioning correction information based on the cause of abnormal positioning may further include:

[0081] When artifact occlusion is a cause of abnormal localization, the occlusion ratio in the image to be detected is calculated; localization correction information is output based on the occlusion ratio. Specifically, when artifact occlusion is a cause of abnormal localization, the indication of implant artifact occlusion and the specific occlusion ratio can be integrated into localization correction information, which is then displayed through an associated output device so that the operator can process the localization correction information. Furthermore, the implant can be either a metallic or non-metallic implant. If it is a metallic implant, a prompt indicating whether scanning needs to be stopped can be output; if it is a non-metallic implant, a prompt indicating that manual localization is required can be output.

[0082] Additionally, if the cause of abnormal localization does not fall under any of the categories of artifact occlusion, localization deviation, or abnormal scanning posture, then the cause can be classified as another reason. For example, a lesion might prevent the algorithm from recognizing the required target structure during the localization process, or the lesion might cause deformation of the target structure's localization identification points, resulting in localization failure. In such cases, manual localization by the operator is required. Other reasons for localization failure include unstable localization algorithms, outdated system versions, or incompatible software versions.

[0083] The present embodiment will now be described and illustrated through preferred embodiments.

[0084] In a preferred embodiment, an anomaly localization analysis method is provided to determine the cause of anomalies in the image to be detected, based on the process described in the above embodiments. Figure 3 This is a flowchart of the anomaly localization analysis method according to a preferred embodiment. For example... Figure 3 As shown, this anomaly localization analysis method includes the following steps:

[0085] Step S301: Obtain the image to be detected; wherein the image to be detected is an image generated after the failure to locate the magnetic resonance object, such as an ankle joint image generated after the failure to locate the ankle joint of the magnetic resonance object.

[0086] Step S302: Using a fully trained target detection model, predict the bounding boxes of the target structures and implant artifacts in the image to be detected. Specifically, the target detection model can be any model that detects specific targets in an image, such as the YOLO model. Inputting the image to be detected into the fully trained target detection model enables the detection of specified target structures and implant artifacts. The target structure can be a bony structure or soft tissue structure to be scanned using magnetic resonance imaging (MRI). The implant can include metallic implants and non-metallic implants.

[0087] Step S303: A well-trained regression model is used to predict the rotation of the target structure in the image to be detected relative to the standard scanning position in the head-to-toe direction, the front-to-back direction, and the left-to-right direction. Specifically, the head-to-toe direction, the front-to-back direction, and the left-to-right direction in the image to be detected are pre-defined as the directions for predicting the rotation. The rotation can be a rotation angle. The well-trained regression model is used to predict the rotation angle of the target structure in the image to be detected in one or more of the above-mentioned preset directions.

[0088] Step S304: Based on the result of step S302, if the overlap ratio between the target structure and the artifact exceeds a preset threshold, then artifact occlusion is confirmed. Specifically, the overlap ratio between the bounding box of the target structure and the bounding box of the artifact can be calculated. For example, the ratio of the size of the overlapping area between the two bounding boxes to the size of the bounding box of the target structure can be used as the overlap ratio. If this overlap ratio exceeds the preset overlap threshold, then it can be confirmed that there is a localization failure caused by artifact occlusion of the implant in the magnetic resonance scan.

[0089] Step S305: Based on the result of step S302, if the target structure deviates from the center of the image to be detected by more than a preset distance threshold, it is confirmed that there is a positioning deviation; that is, when it is detected that the distance between the center of the bounding box of the target structure and the center of the image to be detected exceeds the preset distance threshold, it is confirmed that there is a positioning failure caused by the positioning deviation of the laser lamp.

[0090] Step S306: Based on the result of step S303, if the rotation amount in any direction exceeds a preset rotation amount threshold, then the scanning posture is confirmed to be abnormal. This abnormal scanning posture can be due to the actual scanning posture not matching the standard scanning posture. When the rotation angle of the image to be detected in any preset direction exceeds the preset rotation amount threshold, it is confirmed that the positioning failure is caused by the actual scanning position not matching the selected standard scanning position.

[0091] It should also be noted that when predicting the rotation amount of the image to be detected in step S303, the pre-set direction can also be other directions, and the specific direction can be determined based on the pre-selected standard scanning position. By performing target structure detection and artifact detection on the image to be detected based on a well-trained target detection model, and determining whether there is artifact occlusion or positioning deviation based on the detection results, and predicting the rotation amount of the image to be detected based on a well-trained regression model, the scanning posture anomaly analysis can be achieved. This enables accurate analysis of the causes of anomaly positioning, and timely reminders and corrections for laser positioning deviation or implant artifact occlusion, thereby reducing the operational complexity of magnetic resonance scanning.

[0092] In addition, regarding the process of performing positioning deviation analysis on the image to be detected in the above embodiments, a method for correcting and prompting positioning deviation of laser lights is provided in a preferred embodiment. Figure 4 This is a flowchart of the laser light positioning deviation correction and prompting method according to this preferred embodiment. Figure 4 As shown, the correction prompt method includes the following steps:

[0093] Step S401: Obtain the deviation distance of the target structure relative to the center of the image to be detected; specifically, the deviation distance of the center of the bounding box of the target structure in the image to be detected relative to the center of the image to be detected can be calculated. The detection of this target structure can be achieved by processing the image to be detected using a well-trained target detection model.

[0094] Step S402: Determine whether the laser light deviation is too large based on preset conditions; if so, proceed to step S403; otherwise, proceed to step S404. Specifically, when the deviation distance in step S401 exceeds the preset distance threshold, it can be confirmed that the laser light deviation is too large, that is, in the image to be detected, the target structure is far from the center of the image to be detected. At this time, the target structure may not be fully displayed in the image to be detected, thus causing the positioning failure.

[0095] Step S403: Output a prompt message indicating that the laser positioning is significantly off and needs to be repositioned. If the laser positioning is significantly off, it is impossible to accurately position the target structure. Therefore, it is necessary to notify the relevant personnel to reposition the laser point.

[0096] Step S404: Confirm that the laser light positioning deviation is minimal, calculate and output the required entry or exit distance from the bed based on the deviation distance. When the laser light positioning deviation is minimal, the required entry or exit distance from the bed can be directly calculated based on this deviation distance to improve the efficiency of successful positioning.

[0097] Furthermore, regarding the process of analyzing artifact occlusion in the image to be detected in the above embodiments, a preferred embodiment also provides a method for correcting and prompting implant artifact occlusion. Figure 5 This is a flowchart of the implant artifact occlusion correction and prompting method according to a preferred embodiment of this invention. Figure 5 As shown, the correction prompt method includes the following steps:

[0098] Step S501: Determine whether it is a metal artifact; if yes, proceed to step S502; otherwise, proceed to step S503. When an implant artifact is detected in the image to be detected, the implant artifact can be determined as a metal artifact based on a preset metal implant detection method. For example, the implant artifact can be determined as a metal artifact based on a metal implant detection method commonly used in the field of magnetic resonance scanning.

[0099] Step S502: Output a prompt message indicating the presence of metal artifacts and asking whether to stop the scan. The presence of metal artifacts indicates the presence of metal implants in the patient, which will interfere with the magnetic resonance scan. Therefore, a prompt message needs to be output to the staff to determine whether to stop the scan.

[0100] Step S503: Output a prompt message indicating the presence of a non-metallic implant, requiring manual positioning.

[0101] In view of the above embodiments, a method for correcting and prompting abnormal scanning posture of the image to be detected is provided in a preferred embodiment. Figure 6 This is a flowchart of the scanning posture abnormality correction prompt method according to this preferred embodiment. Figure 6 As shown, the correction prompt method includes the following steps:

[0102] Step S601: Confirm that the actual scanning position does not match the standard scanning position; for example, if the rotation amount of the image to be detected in the preset direction exceeds the preset rotation amount threshold, confirm that the actual scanning position of the magnetic resonance object does not match the pre-selected standard scanning position, thereby determining that the abnormal positioning is caused by abnormal scanning posture.

[0103] Step S602: Output a prompt message indicating that the correct standard scanning position has been selected.

[0104] This embodiment also provides a positioning correction device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. The terms "module," "unit," "subunit," etc., used below refer to combinations of software and / or hardware that perform a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0105] Figure 7 This is a structural block diagram of the positioning correction device 70 in this embodiment, as shown below. Figure 7 As shown, the positioning correction device 70 includes: an acquisition module 72, a determination module 74, and an output module 76; wherein: the acquisition module 72 is used to acquire an image to be detected; wherein the image to be detected is an image generated after the positioning of the magnetic resonance object fails during magnetic resonance scanning; the determination module 74 is used to perform abnormal positioning analysis on the image to be detected to determine the cause of the abnormal positioning of the magnetic resonance object; the output module 76 is used to determine and output positioning correction information based on the cause of the abnormal positioning.

[0106] The aforementioned positioning correction device 70 can analyze the reasons for positioning failure and provide a positioning correction scheme, thereby facilitating users to successfully complete automatic positioning based on positioning correction information, and thus reducing the operational complexity of magnetic resonance scanning.

[0107] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0108] Furthermore, in conjunction with the positioning correction methods provided in the above embodiments, this embodiment can also provide a storage medium for implementation. The storage medium stores a computer program; when executed by a processor, the computer program implements any of the positioning correction methods described in the above embodiments.

[0109] It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. All other embodiments derived by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0110] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0111] Obviously, the accompanying drawings are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar situations based on these drawings without any creative effort. Furthermore, it is understood that although the work done in this development process may be complex and lengthy, for those skilled in the art, certain design, manufacturing, or production modifications made based on the technical content disclosed in this application are merely conventional technical means and should not be considered as insufficient disclosure of this application.

[0112] The term "embodiment" in this application refers to a specific feature, structure, or characteristic described in connection with an embodiment that may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily imply the same embodiment, nor does it imply that it is mutually exclusive with or independent of other embodiments. It will be clearly or implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0113] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of patent protection. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims

1. A positioning correction method, characterized in that, include: Acquire the image to be detected; wherein, the image to be detected is an image generated after the magnetic resonance object fails to be located during magnetic resonance scanning; Anomaly localization analysis is performed on the image to be detected to determine the cause of the anomaly localization for the magnetic resonance object; Based on the aforementioned abnormal location causes, location correction information is determined and output; wherein: The step of performing anomaly localization analysis on the image to be detected to determine the cause of the anomaly localization for the magnetic resonance object includes: Using a well-trained target detection model, target structure detection and artifact detection are performed on the image to be detected, obtaining target structure detection results and artifact detection results; based on the target structure detection results and the artifact detection results, artifact occlusion analysis is performed on the image to be detected to determine whether the cause of the abnormal localization includes artifact occlusion; based on the target structure detection results, localization deviation analysis is performed on the image to be detected to determine whether the cause of the abnormal localization includes localization deviation; and / or, Using a well-trained regression model, the rotation amount of the image to be detected is predicted to obtain the rotation amount of the image to be detected in a preset direction; based on the rotation amount of the image to be detected in the preset direction, the scanning posture anomaly analysis is performed on the image to be detected to determine whether the cause of the anomaly localization includes scanning posture anomaly.

2. The positioning correction method according to claim 1, characterized in that: The step of performing artifact occlusion analysis on the image to be detected based on the target structure detection results and the artifact detection results to determine whether the cause of the abnormal localization includes artifact occlusion includes: If the target structure detection result and the artifact detection result indicate that the overlap ratio between the target structure and the artifact in the image to be detected reaches a preset overlap threshold, the abnormal localization cause of the magnetic resonance object is determined to include artifact occlusion. The step of performing a positioning deviation analysis on the image to be detected based on the target structure detection results to determine whether the cause of the abnormal positioning includes positioning deviation includes: If the target structure detection result indicates that the deviation distance of the target structure relative to the center of the image to be detected reaches a preset distance threshold, the cause of the abnormal positioning is determined to include positioning deviation.

3. The positioning correction method according to claim 1, characterized in that, The step of performing a scanning posture anomaly analysis on the image to be detected based on the rotation amount of the image to be detected in a preset direction, in order to determine whether the cause of the anomaly localization includes scanning posture anomalies, includes: If the rotation of the image to be detected in a preset direction reaches a preset rotation threshold, the scanning posture of the magnetic resonance object is determined to be abnormal.

4. The positioning correction method according to any one of claims 1 to 3, characterized in that, The process of determining and outputting positioning correction information based on the cause of the abnormal positioning includes: If the cause of the abnormal positioning includes positioning deviation, calculate the amount of positioning deviation that occurred after locating the magnetic resonance object in the image to be detected; Positioning correction information is output based on the positioning deviation.

5. The positioning correction method according to any one of claims 1 to 3, characterized in that, The step of determining and outputting positioning correction information based on the cause of the abnormal positioning also includes: If the cause of the abnormal localization includes artifact occlusion, calculate the occlusion ratio in the image to be detected; Positioning correction information is output based on the occlusion ratio.

6. A positioning and correction device, characterized in that, include: The module consists of an acquisition module, a judgment module, and an output module; where: The acquisition module is used to acquire the image to be detected; wherein, the image to be detected is an image generated after the magnetic resonance object fails to be located during magnetic resonance scanning; The determination module is used to perform anomaly localization analysis on the image to be detected and determine the cause of the anomaly localization for the magnetic resonance object; The output module is used to determine and output positioning correction information based on the cause of the abnormal positioning; wherein: The step of performing anomaly localization analysis on the image to be detected to determine the cause of the anomaly localization for the magnetic resonance object includes: Using a well-trained target detection model, target structure detection and artifact detection are performed on the image to be detected, obtaining target structure detection results and artifact detection results; based on the target structure detection results and the artifact detection results, artifact occlusion analysis is performed on the image to be detected to determine whether the cause of the abnormal localization includes artifact occlusion; based on the target structure detection results, localization deviation analysis is performed on the image to be detected to determine whether the cause of the abnormal localization includes localization deviation; and / or, Using a well-trained regression model, the rotation amount of the image to be detected is predicted to obtain the rotation amount of the image to be detected in a preset direction; based on the rotation amount of the image to be detected in the preset direction, the scanning posture anomaly analysis is performed on the image to be detected to determine whether the cause of the anomaly localization includes scanning posture anomaly.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the positioning correction method according to any one of claims 1 to 5.

Images