A method and system for screening quality of thyroid ultrasound scanning video

Abstract

The present invention provides a thyroid ultrasound scan video quality screening method and system, which obtains a scan video of the thyroid; performs decoding processing to obtain a plurality of frames of ultrasound images, and counts the total number of frames of the ultrasound images; identifies the thyroid in each frame of the ultrasound image, and counts the first frame number of the ultrasound image in which the thyroid exists; determines whether the ratio of the first frame number to the total frame number exceeds a preset ratio, and if not, outputs a judgment result that the scan video quality is unqualified, and if so, continues to determine whether the quality of the scan video is qualified according to the thyroid feature information, and uploads the scan video to a server if it is qualified. The waste of network and computing resources is reduced, and the scan video that meets the video quality requirements is greatly improved, and the efficiency of the diagnostic doctor in diagnosing the patient's disease is greatly improved, and the effectiveness and accuracy of the diagnosis report are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to a method and system for screening thyroid ultrasound scanning video quality. Background Art

[0002] When the front-end doctor uses the ultrasound scanning equipment to obtain the thyroid ultrasound scanning video, the video is directly uploaded to the cloud server, all the videos are directly processed on the cloud server, and then the doctor reads the ultrasound image and fills out the diagnosis report. When the front-end doctor scans the patient's thyroid gland, some thyroid parts may be missed or scanned incorrectly for various reasons. A good quality thyroid ultrasound scanning video usually contains clear and effective thyroid images, involving the left transverse section, right transverse section, isthmus, left longitudinal section, right longitudinal section and other thyroid parts. Check whether the number of frames of the effective image in the scanning video meets the standard, and whether the length of the scanning video meets the requirements.

[0003] In the prior art, the thyroid scan video is directly uploaded to the cloud server, resulting in the following problems:

[0004] (1) It is impossible to screen out scan videos with low quality and promptly remind the scanning doctor to re-scan the patient;

[0005] (2) Failure to promptly alert doctors when they miss scanning certain areas of the thyroid gland;

[0006] (3) It is impossible to remind doctors to correct incorrect scanning methods in time when the scanning time is too short or too long;

[0007] (4) When low-quality scan videos are transmitted to the cloud server, network resources and computing resources of the cloud server are occupied;

[0008] (5) Scanning videos of low quality or those that miss certain areas of the thyroid gland will seriously affect the doctor's diagnosis of the patient's condition, leading to failure of the diagnosis. Summary of the invention

[0009] Based on the problems existing in the prior art, the present invention provides a thyroid ultrasound scanning video quality screening method and system, aiming to solve the technical problems in the prior art such as diagnosis failure caused by low thyroid ultrasound scanning video quality and waste of network resources.

[0010] A method for screening thyroid ultrasound scanning video quality comprises the following steps:

[0011] Step A1, performing an ultrasonic scan on the patient's thyroid gland to obtain a scan video of the thyroid gland;

[0012] Step A2, performing decoding processing on the scanned video to obtain a plurality of frames of ultrasound images, and counting the total number of frames of the ultrasound images;

[0013] Step A3, using the first recognition model to recognize the thyroid gland in each frame of the ultrasound image, and outputting the recognition result;

[0014] Step A4, counting the number of first frames of the ultrasound image in which the thyroid gland is present according to the recognition result of the first recognition model;

[0015] Step A5, determining whether the ratio of the first frame number to the total frame number exceeds a preset ratio;

[0016] If yes, proceed to step A6;

[0017] If not, the judgment result that the scanned video quality is unqualified is output;

[0018] Step A6, using the second recognition model to identify thyroid feature information in the ultrasound image belonging to the first frame number, classifying and counting feature information of the ultrasound image belonging to the first frame number according to the thyroid feature information, and judging whether the video quality of the scanned video is qualified:

[0019] If yes, proceed to step A7;

[0020] Step A7, uploading the scanned video to the server.

[0021] Furthermore, before step A2, the following steps are performed:

[0022] Step B1, reading the total duration of the scanned video;

[0023] Step B2, determining whether the total duration of the scanned video exceeds a preset time range:

[0024] If yes, then output the judgment result that the scanned video quality is unqualified;

[0025] If not, continue to step A2.

[0026] Further, step A6 includes the following steps:

[0027] Step B3, for the ultrasound image belonging to the first frame number, using the second recognition model to recognize the contour features of the thyroid gland in the ultrasound image;

[0028] Step B4, acquiring thyroid site information in the ultrasound image according to the contour features, the thyroid site information including the transverse section, the longitudinal section and the isthmus;

[0029] Step B5, respectively counting and identifying the second frame number of the ultrasonic image of the transverse section, the third frame number of the ultrasonic image of the longitudinal section, and the fourth frame number of the ultrasonic image of the isthmus;

[0030] Step B6, determining whether the second frame number is greater than the first threshold, whether the third frame number is greater than the second threshold, and whether the fourth frame number is greater than the third threshold:

[0031] If the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time, continue to execute step A7;

[0032] If at least one of the second frame number, the third frame number and the fourth frame number is not greater than the corresponding threshold value, a judgment result that the video quality of the scanned video is unqualified is output.

[0033] Further, in step B6, when the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time, the following steps are further performed before step A7:

[0034] Step B7, judging whether the transverse section is a left-side transverse section or a right-side transverse section for the ultrasound image of the transverse section, if it is a left-side transverse section, recording the ultrasound image as a left-side thyroid image, and if it is not a left-side transverse section, recording the ultrasound image as a right-side thyroid image;

[0035] Step B8, counting the number of frames of the left thyroid image and the number of frames of the right thyroid image respectively;

[0036] Step B9, judging whether there is at least one frame of the left thyroid image and at least one frame of the right thyroid image according to the frame number of the left thyroid image and the frame number of the right thyroid image:

[0037] If yes, proceed to step A7;

[0038] If not, continue to step B10;

[0039] Step B10, when there is no left thyroid image, outputting a judgment result that there is no left thyroid image; when there is no right thyroid image, outputting a judgment result that there is no right thyroid image.

[0040] Furthermore, it also includes:

[0041] Step A8, when the judgment result that the video quality of the scanned video is unqualified is output, a first prompt message requesting a re-scan is generated.

[0042] Furthermore, after step B10, the following steps are further included:

[0043] Step B11, when the judgment result that the left thyroid image is not available is output, a second prompt information indicating that the left thyroid image is not available is generated;

[0044] Step B12: When the judgment result that the right thyroid image is not available is output, a third prompt information indicating that the right thyroid image is not obtained is generated.

[0045] Furthermore, the preset time range is 8 to 12 seconds.

[0046] A thyroid ultrasound scanning video quality screening system, comprising the aforementioned thyroid ultrasound scanning video quality screening method, including:

[0047] A video acquisition module is used to perform an ultrasonic scan of the patient's thyroid gland and obtain a scan video of the thyroid gland;

[0048] A decoding module, connected to the video acquisition module, is used to perform decoding processing on the scanned video to obtain a plurality of frames of ultrasound images;

[0049] A first statistical module, connected to the decoding module, for counting the total number of frames of the ultrasound image;

[0050] A first recognition module, connected to the decoding module, for identifying the thyroid gland in each frame of the ultrasound image using a first recognition model and outputting a recognition result;

[0051] a second statistical module, connected to the first recognition module, for counting the number of first frames of the ultrasound image in which the thyroid gland exists according to the recognition result of the first recognition model;

[0052] A first judgment module, connected to the second statistical module and the first statistical module, is used to judge whether the ratio of the first frame number to the total frame number exceeds a preset ratio, and output a judgment result;

[0053] a second recognition module connected to the first judgment module, configured to, when a judgment result shows that the ratio of the first frame number to the total frame number exceeds a preset ratio, use a second recognition model to recognize thyroid feature information in the ultrasound image belonging to the first frame number;

[0054] a second judgment module, connected to the second recognition module, for classifying and counting characteristic information of the ultrasound images in the first frame number according to the thyroid characteristic information, and judging whether the video quality of the scanned video is qualified;

[0055] The uploading module is connected to the second judging module and the video acquisition module, and is used to upload the scanning video to the server when the video quality of the scanning video is qualified.

[0056] Furthermore, it also includes:

[0057] The time acquisition module is connected to the video acquisition module and is used to read the total duration of the scanned video;

[0058] The third judgment module is connected to the time acquisition module and is used to judge whether the total duration of the scanned video exceeds the preset time range;

[0059] The decoding module is also connected to the third judgment module and is used to perform decoding processing on the scanning video to obtain several frames of ultrasound images when the total duration of the scanning video does not exceed a preset time range.

[0060] Furthermore, it also includes:

[0061] The thyroid characteristic information includes the contour characteristics of the thyroid gland;

[0062] A part acquisition module, connected to the second recognition module and the second judgment module, for acquiring thyroid part information in an ultrasonic image according to contour features of the thyroid gland, the thyroid part information including a transverse section, a longitudinal section and an isthmus;

[0063] The second judgment module also includes:

[0064] A first statistical unit is used to respectively count the second frame number of the ultrasonic image identifying the transverse section, the third frame number of the ultrasonic image identifying the longitudinal section, and the fourth frame number of the ultrasonic image identifying the isthmus;

[0065] A first judging unit, connected to the first statistical unit, for judging whether the second frame number is greater than the first threshold, whether the third frame number is greater than the second threshold, and whether the fourth frame number is greater than the third threshold;

[0066] The uploading module is used to upload the scanned video to the server when the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time.

[0067] The beneficial technical effect of the present invention is: by providing a thyroid ultrasound scanning video quality screening method and system, when the number of frames of effective images in the scanning video does not meet the standard, the doctor is notified in time so that the doctor can re-scan, and the video with video quality that meets the requirements is uploaded to the server, thereby reducing the waste of network and computing resources. The scanning video that meets the video quality requirements greatly improves the efficiency of the diagnostic doctor in diagnosing the patient's disease and improves the effectiveness and accuracy of the diagnosis report. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 A flowchart of the steps of a method for screening thyroid ultrasound scan video quality according to the present invention;

[0069] Figure 2 A flowchart of a preferred embodiment of a method for screening thyroid ultrasound scan video quality according to the present invention;

[0070] Figure 3 A flowchart of a preferred embodiment of a method for screening thyroid ultrasound scan video quality according to the present invention;

[0071] Figure 4A flowchart of a preferred embodiment of a method for screening thyroid ultrasound scan video quality according to the present invention;

[0072] Figure 5 A flowchart of the steps of a thyroid ultrasound scanning video quality screening system of the present invention;

[0073] Figure 6 A schematic diagram of a module of a preferred embodiment of a thyroid ultrasound scanning video quality screening system of the present invention;

[0074] Figure 7 A schematic diagram of a module of a preferred embodiment of a thyroid ultrasound scanning video quality screening system of the present invention;

[0075] Figure 8 A schematic diagram of a module of a preferred embodiment of a thyroid ultrasound scanning video quality screening system of the present invention;

[0076] Fig. 9 A schematic diagram of a module of a preferred embodiment of a thyroid ultrasound scanning video quality screening system of the present invention; DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0079] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0080] See also Figure 1 The present invention provides a method for screening thyroid ultrasound scan video quality, comprising the following steps:

[0081] Step A1, performing an ultrasonic scan on the patient's thyroid gland to obtain a scan video of the thyroid gland;

[0082] Step A2, performing decoding processing on the scanned video to obtain a plurality of frames of ultrasound images, and counting the total number of frames of the ultrasound images;

[0083] Step A3, using the first recognition model to recognize the thyroid gland in each frame of the ultrasound image, and outputting the recognition result;

[0084] Step A4, according to the recognition result of the first recognition model, counting the number of first frames of the ultrasound image in which the thyroid gland exists,

[0085] Step A5, determining whether the ratio of the first frame number to the total frame number exceeds a preset ratio;

[0086] If yes, proceed to step A6;

[0087] If not, the judgment result that the scanned video quality is unqualified is output;

[0088] Step A6, using the second recognition model to identify thyroid feature information in the ultrasound image belonging to the first frame number, classifying and counting feature information of the ultrasound image belonging to the first frame number according to the thyroid feature information, and judging whether the video quality of the scanned video is qualified:

[0089] If yes, proceed to step A7;

[0090] Step A7, uploading the scanned video to the server.

[0091] See also Figure 2 , further, before step A2, perform the following steps:

[0092] Step B1, reading the total duration of the scanned video;

[0093] Step B2, determining whether the total duration of the scanned video exceeds a preset time range:

[0094] If yes, then output the judgment result that the scanned video quality is unqualified;

[0095] If not, continue to step A2.

[0096] See also Figure 3 , further, step A6 includes the following steps:

[0097] Step B3, for the ultrasound image belonging to the first frame number, using the second recognition model to recognize the contour features of the thyroid gland in the ultrasound image;

[0098] Step B4, acquiring thyroid site information in the ultrasound image according to the contour features, the thyroid site information including the transverse section, the longitudinal section and the isthmus;

[0099] Step B5, respectively counting and identifying the second frame number of the ultrasonic image of the transverse section, the third frame number of the ultrasonic image of the longitudinal section, and the fourth frame number of the ultrasonic image of the isthmus;

[0100] Step B6, determining whether the second frame number is greater than the first threshold, whether the third frame number is greater than the second threshold, and whether the fourth frame number is greater than the third threshold:

[0101] If the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time, continue to execute step A7;

[0102] If at least one of the second frame number, the third frame number and the fourth frame number is not greater than the corresponding threshold value, a judgment result that the video quality of the scanned video is unqualified is output.

[0103] See also Figure 4 Further, in step B6, when the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time, the following steps are further performed before step A7:

[0104] Step B7, judging whether the transverse section is a left-side transverse section or a right-side transverse section for the ultrasound image of the transverse section, if it is a left-side transverse section, recording the ultrasound image as a left-side thyroid image, and if it is not a left-side transverse section, recording the ultrasound image as a right-side thyroid image;

[0105] Step B8, counting the number of frames of the left thyroid image and the number of frames of the right thyroid image respectively;

[0106] Step B9, judging whether there is at least one frame of the left thyroid image and at least one frame of the right thyroid image according to the frame number of the left thyroid image and the frame number of the right thyroid image:

[0107] If yes, proceed to step A7;

[0108] If not, continue to step B10;

[0109] Step B10, when there is no left thyroid image, outputting a judgment result that there is no left thyroid image; when there is no right thyroid image, outputting a judgment result that there is no right thyroid image.

[0110] During a thyroid ultrasound scan, the patient's thyroid gland needs to be scanned in five areas, namely the left transverse section, left longitudinal section, isthmus, right transverse section, and right longitudinal section. In addition, the doctor needs to scan two videos at the same time, the left thyroid image and the right thyroid image.

[0111] Furthermore, it also includes:

[0112] Step A8, when the judgment result that the video quality of the scanned video is unqualified is output, a first prompt message requesting a re-scan is generated.

[0113] Furthermore, after step B10, the following steps are further included:

[0114] Step B11, when the judgment result that the left thyroid image is not available is output, a second prompt information indicating that the left thyroid image is not available is generated;

[0115] Step B12: When the judgment result that the right thyroid image is not available is output, a third prompt information indicating that the right thyroid image is not obtained is generated.

[0116] Specifically, determine whether the identified thyroid cross-section image is a left cross-section. If it is a left cross-section, record the identification of a normal left thyroid image; if it is not a left cross-section, record the identification of a normal right thyroid image. Determine whether the doctor has scanned a left and a right thyroid image. If one side of the thyroid image is missing, remind the doctor to re-scan the thyroid video on that side; if it is not missing, upload the scanned thyroid video to the server.

[0117] Furthermore, the preset time range is 8 to 12 seconds.

[0118] If the total time T<8s, a first prompt message is generated to remind the doctor to re-scan. If the total time T>12s, a first prompt message is generated to remind the doctor to pause the current scan and re-scan.

[0119] Specifically, the first recognition model is an AI model.

[0120] Specifically, the second recognition model is an AI model.

[0121] Specifically, the server is a cloud server.

[0122] Specifically, the preset ratio is 0.5. That is to say, if the first frame number of the ultrasound image with the thyroid gland accounts for more than 50% of the total frame number of the ultrasound image, it means that the ultrasound scan video quality meets the standard, and if the first frame number of the ultrasound image with the thyroid gland accounts for less than 50% of the total frame number of the ultrasound image, it means that the video quality is low, and the first prompt information is generated to remind the scanning doctor to re-scan.

[0123] Specifically, the file formats of ultrasound images are png, jpg, and bmp.

[0124] Specifically, the first threshold, the second threshold and the third threshold are all the same value, preferably, all are 5.

[0125] See also Figure 5 The present invention also provides a thyroid ultrasound scan video quality screening system, comprising a thyroid ultrasound scan video quality screening method as described above, including:

[0126] A video acquisition module (1) is used to perform an ultrasonic scan on the patient's thyroid gland and obtain a scan video of the thyroid gland;

[0127] A decoding module (2), connected to the video acquisition module (1), is used to perform decoding processing on the scanned video to obtain a plurality of frames of ultrasound images;

[0128] A first statistical module (3), connected to the decoding module (2), is used to count the total number of frames of the ultrasound image;

[0129] A first recognition module (4), connected to the decoding module (2), is used to recognize the thyroid gland in each frame of the ultrasound image using a first recognition model and output a recognition result;

[0130] A second statistical module (5), connected to the first recognition module (4), is used to count the number of first frames of the ultrasound image containing the thyroid gland according to the recognition result of the first recognition model;

[0131] A first judgment module (6) is connected to the second statistical module (5) and the first statistical module (3), and is used to judge whether the ratio of the first frame number to the total frame number exceeds a preset ratio, and output a judgment result;

[0132] A second recognition module (7) is connected to the first judgment module (6) and is used to use a second recognition model to recognize thyroid feature information in the ultrasound image belonging to the first frame number when the judgment result is that the ratio of the first frame number to the total frame number exceeds a preset ratio;

[0133] A second judgment module (8), connected to the second identification module (7), is used to classify and count the characteristic information of the ultrasound images in the first frame number according to the thyroid characteristic information, and to judge whether the video quality of the scanned video is qualified;

[0134] The uploading module (9) is connected to the second judging module (8) and the video collecting module (1) and is used to upload the scanning video to the server when the video quality of the scanning video is qualified.

[0135] See also Figure 6 , further, as another preferred embodiment of the present invention, it also includes:

[0136] A time acquisition module (10), connected to the video acquisition module (1), is used to read the total duration of the scanned video;

[0137] A third judgment module (11), connected to the time acquisition module (10), is used to judge whether the total duration of the scanned video exceeds a preset time range;

[0138] The decoding module (2) is also connected to the third judgment module (11) and is used to perform decoding processing on the scanning video to obtain a plurality of frames of ultrasound images when the total duration of the scanning video does not exceed a preset time range.

[0139] See also Figure 7 , further, as another preferred embodiment of the present invention, it also includes:

[0140] The thyroid characteristic information includes the contour characteristics of the thyroid gland;

[0141] A part acquisition module (12) connected to the second recognition module (7) and the second judgment module (8) is used to acquire thyroid part information in the ultrasound image according to the contour features of the thyroid gland, wherein the thyroid part information includes a transverse section, a longitudinal section and an isthmus;

[0142] The second judgment module (8) also includes:

[0143] A first statistical unit (81) is used to respectively count the second frame number of the ultrasonic image of the transverse section, the third frame number of the ultrasonic image of the longitudinal section, and the fourth frame number of the ultrasonic image of the isthmus;

[0144] A first judgment unit (82), connected to the first statistical unit (81), is used to judge whether the second frame number is greater than the first threshold, whether the third frame number is greater than the second threshold, and whether the fourth frame number is greater than the third threshold;

[0145] An uploading module (9) is used to upload the scanned video to a server when the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time.

[0146] See also Fig. 9 , further, as another preferred embodiment of the present invention, it also includes:

[0147] The first reminder module (13) is connected to the third judgment module (11) and is used to generate a first reminder message requesting a re-scan when the total duration of the scanned video exceeds a preset time range. The first reminder message includes a reason for failure, in which the reason for failure is that the total duration of the scanned video is unqualified, and reminds the scanning doctor to pay attention to the scanning time.

[0148] The first reminder module (13) is also connected to the first judgment module (6) and is used to generate a first prompt message requesting a re-scan when it is judged that the ratio of the first frame number to the total frame number does not exceed a preset ratio. The first prompt message includes the reason for failure, and here, the reason for failure is that the number of ultrasound images containing the thyroid gland is insufficient, so as to remind the doctor to pay attention to the scanning method, find the reason for the failure of the scanning, and correct the wrong scanning method.

[0149] The first reminder module (13) is also connected to the second judgment module (8), and is used to generate a first prompt message requesting a re-scan when at least one of the second frame number, the third frame number, and the fourth frame number is not greater than the corresponding threshold value. The first prompt message includes the reason for failure, and here, the reason for failure is that the number of ultrasound images containing information of one or more parts of the thyroid gland is insufficient, for example, the number of ultrasound images of at least one part of the transverse part, the longitudinal part, and the isthmus is insufficient, so as to remind the doctor to pay attention to the scanning method, find the reason for the failure of the scanning, correct the wrong scanning method, and focus on scanning the parts with fewer ultrasound images.

[0150] See also Figure 8 Further, as another preferred embodiment of the present invention, the second judgment module (8) also includes:

[0151] A second judgment unit (83) is connected to the first judgment unit (82) and judges whether the transected part is a left transected part or a right transected part based on the ultrasound image of the transected part;

[0152] A classification marking unit (84), connected to the second judgment unit (83), is used to record the ultrasound image as a left thyroid image if the ultrasound image is a left cross section, and to record the ultrasound image as a right thyroid image if the ultrasound image is not a left cross section;

[0153] A second statistical unit (85), connected to the classification labeling unit (84), is used to respectively count the number of frames of the left thyroid image and the number of frames of the right thyroid image;

[0154] A third judgment unit (86), connected to the second statistical unit (85), is used to judge whether there is at least one frame of the left thyroid image and at least one frame of the right thyroid image according to the frame number of the left thyroid image and the frame number of the right thyroid image;

[0155] The uploading module (9) is used to upload the scanned video to the server when there is at least one frame of the left thyroid image and at least one frame of the right thyroid image.

[0156] See also Fig. 9 , further, as another preferred embodiment of the present invention, it also includes:

[0157] A second reminder module (14) connected to the second judgment module (8) is used to generate second prompt information that the left thyroid image has not been obtained when the second judgment module outputs a judgment result that the left thyroid image is not present;

[0158] The third reminder module (15) is connected to the second judgment module (8) and is used to generate a third reminder message indicating that the right thyroid image has not been obtained when the second judgment module outputs a judgment result indicating that the right thyroid image is not available, thereby reminding the doctor to perform an ultrasonic scan on the missing left thyroid image or right thyroid image.

[0159] Furthermore, the first threshold, the second threshold and the third threshold are all the same value.

[0160] Specifically, the server is a cloud server.

[0161] Specifically, the preset ratio is 0.5. That is to say, if the first frame number of the ultrasound image with the thyroid gland accounts for more than 50% of the total frame number of the ultrasound image, it means that the ultrasound scan video quality meets the standard, and if the first frame number of the ultrasound image with the thyroid gland accounts for less than 50% of the total frame number of the ultrasound image, it means that the video quality is low, and the first prompt information is generated to remind the scanning doctor to re-scan.

[0162] Specifically, the file formats of ultrasound images are png, jpg, and bmp.

[0163] Specifically, the preferred preset time range is 8 to 12 seconds. If the total time T<8 seconds, a first prompt message is generated to remind the doctor to re-scan. If the total time T>12 seconds, a first prompt message is generated to remind the doctor to pause the current scan and re-scan.

[0164] Specifically, the first recognition model is an AI model.

[0165] Specifically, the second recognition model is an AI model.

[0166] The present invention proposes a thyroid ultrasound scanning video quality screening system and method, which realizes real-time screening of thyroid ultrasound scanning video quality, and real-time evaluation of the quality of the front-end doctor's thyroid scanning video. When the evaluated video quality is too low, the scanning doctor is promptly reminded to re-scan the patient. When the scanning doctor misses certain parts when scanning the thyroid, the scanning doctor is reminded in real time to re-scan. When the scanning doctor scans for too short or too long, the doctor is promptly reminded to correct the wrong scanning method. Videos that meet the video quality requirements are uploaded to the cloud server, reducing the waste of network and cloud computing resources. Scanning videos that meet the video quality requirements greatly improve the efficiency of the diagnostic doctor in diagnosing the patient's symptoms and improve the effectiveness and accuracy of the diagnosis report.

[0167] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for screening thyroid ultrasound scan video quality, characterized in that: The steps include: Step A1, performing an ultrasonic scan on the patient's thyroid gland to obtain a scan video of the thyroid gland; Step A2, performing decoding processing on the scanned video to obtain a plurality of frames of ultrasound images, and counting the total number of frames of the ultrasound images; Step A3, using the first recognition model to recognize the thyroid gland in each frame of the ultrasound image, and outputting the recognition result; Step A4, counting the number of first frames of the ultrasound image in which the thyroid gland exists according to the recognition result of the first recognition model; Step A5, determining whether the ratio of the first frame number to the total frame number exceeds a preset ratio; If yes, proceed to step A6; If not, outputting a judgment result that the video quality of the scanned video is unqualified; Step A6, using a second recognition model to identify thyroid feature information in the ultrasound image belonging to the first number of frames, classifying and counting feature information of the ultrasound image belonging to the first number of frames according to the thyroid feature information, and determining whether the video quality of the scanned video is qualified: If yes, proceed to step A7; Step A7, uploading the scan video to the server.

2. The method for screening thyroid ultrasound scan video quality according to claim 1, characterized in that, before step A2, the following steps are performed: Step B1, reading the total duration of the scanning video; Step B2, determining whether the total duration of the scanned video exceeds a preset time range: If yes, outputting a judgment result that the video quality of the scanned video is unqualified; If not, continue with step A2.

3. The method for screening thyroid ultrasound scan video quality according to claim 1, wherein step A6 comprises the following steps: Step B3, for the ultrasound image in the first frame number, using a second recognition model to recognize contour features of the thyroid gland in the ultrasound image; Step B4, acquiring thyroid part information in the ultrasonic image according to the contour feature, the thyroid part information including a transverse section, a longitudinal section and an isthmus; Step B5, respectively counting and identifying the second frame number of the ultrasonic image of the transverse section, the third frame number of the ultrasonic image of the longitudinal section, and the fourth frame number of the ultrasonic image of the isthmus; Step B6, determining whether the second frame number is greater than the first threshold, whether the third frame number is greater than the second threshold, and whether the fourth frame number is greater than the third threshold: If the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time, continue to execute step A7; If at least one of the second frame number, the third frame number and the fourth frame number is not greater than the corresponding threshold value, a judgment result that the video quality of the scanned video is unqualified is output.

4. The method for screening thyroid ultrasound scan video quality according to claim 3, characterized in that, in step B6, when the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time, the following steps are further performed before step A7: Step B7, judging whether the transverse section is a left-side transverse section or a right-side transverse section for the ultrasound image of the transverse section, if it is a left-side transverse section, recording the ultrasound image as a left-side thyroid image, and if it is not a left-side transverse section, recording the ultrasound image as a right-side thyroid image; Step B8, respectively counting the number of frames of the left thyroid image and the number of frames of the right thyroid image; Step B9, judging whether there is at least one frame of the left thyroid image and at least one frame of the right thyroid image according to the frame number of the left thyroid image and the frame number of the right thyroid image: If yes, continue with step A7; If not, continue to step B10; Step B10, when the left thyroid image is not available, outputting a judgment result that the left thyroid image is not available; when the right thyroid image is not available, outputting a judgment result that the right thyroid image is not available.

5. The method for screening thyroid ultrasound scan video quality according to claim 3, further comprising: Step A8, when the judgment result that the video quality of the scanned video is unqualified is output, a first prompt message requesting a re-scan is generated.

6. A method for screening thyroid ultrasound scan video quality as claimed in claim 4, characterized in that: The step B10 further includes: Step B11, when the judgment result that the left thyroid image is not available is output, a second prompt information indicating that the left thyroid image is not available is generated; Step B12: When the judgment result that the right thyroid image is not present is output, a third prompt message indicating that the right thyroid image is not obtained is generated.

7. A thyroid ultrasound scanning video quality screening method as described in claim 2, characterized in that the preset time range is 8~12s.

8. A thyroid ultrasound scan video quality screening system, characterized in that: include: A video acquisition module is used to perform an ultrasonic scan of the patient's thyroid gland and obtain a scan video of the thyroid gland; A decoding module, connected to the video acquisition module, for performing decoding processing on the scan video to obtain a plurality of frames of ultrasound images; A first statistical module, connected to the decoding module, for counting the total number of frames of the ultrasound image; A first recognition module, connected to the decoding module, configured to recognize the thyroid gland in each frame of the ultrasound image using a first recognition model and output a recognition result; a second statistical module, connected to the first recognition module, and configured to count the number of first frames of the ultrasound image in which the thyroid gland exists according to the recognition result of the first recognition model; a first judging module, connected to the second statistical module and the first statistical module, configured to judge whether a ratio of the first frame number to the total frame number exceeds a preset ratio, and output a judgment result; a second recognition module, connected to the first judgment module, configured to, when the judgment result is that the ratio of the first frame number to the total frame number exceeds the preset ratio, use a second recognition model to recognize thyroid feature information in the ultrasound image belonging to the first frame number; a second judgment module, connected to the second recognition module, for classifying and counting characteristic information of the ultrasound images in the first number of frames according to the thyroid characteristic information, and judging whether the video quality of the scanned video is qualified; The uploading module is connected to the second judgment module and the video acquisition module, and is used to upload the scanning video to the server when the video quality of the scanning video is qualified.

9. A thyroid ultrasound scan video quality screening system as claimed in claim 8, characterized in that: Also includes: A time acquisition module, connected to the video acquisition module, for reading the total duration of the scanning video; A third judgment module, connected to the time acquisition module, is used to judge whether the total duration of the scanning video exceeds a preset time range; The decoding module is also connected to the third judgment module, and is used to perform decoding processing on the scanning video to obtain several frames of ultrasound images when the total duration of the scanning video does not exceed the preset time range.

10. The thyroid ultrasound scan video quality screening system according to claim 8, characterized in that: Also includes: The thyroid characteristic information includes contour characteristics of the thyroid gland; a part acquisition module, connected to the second recognition module and the second judgment module, for acquiring thyroid part information in the ultrasound image according to the contour features of the thyroid, the thyroid part information including a transverse part, a longitudinal part and an isthmus; The second judgment module also includes: A first statistical unit is used to respectively count the second frame number of the ultrasonic image identifying the transverse section, the third frame number of the ultrasonic image identifying the longitudinal section, and the fourth frame number of the ultrasonic image identifying the isthmus; a first judging unit, connected to the first statistical unit, for judging whether the second frame number is greater than a first threshold, whether the third frame number is greater than the second threshold, and whether the fourth frame number is greater than a third threshold; The uploading module is used to upload the scanning video to the server when the second frame number, the third frame number and the fourth frame number are all greater than their corresponding thresholds at the same time.

Images