An ai-assisted thyroid scanning system and method

Abstract

The application provides an AI-assisted thyroid scanning system and method, and relates to the technical field of ultrasonic examination.The application comprises the following steps: acquiring a video stream, dividing the video stream into multiple video clips in chronological order and associating time intervals, extracting optimal thyroid ultrasonic images from the video clips, uploading the optimal thyroid ultrasonic images to a cloud analysis system to obtain corresponding cloud analysis results, and distributing each video clip to multiple AI analysis lightweight models to obtain corresponding local analysis results; establishing a cache queue, adding video clips that have obtained local analysis results and cloud analysis results to the cache queue, adjusting the order, and then sequentially outputting each video clip and analysis result in the cache queue.The beneficial effects are as follows: the cost is reduced, the popularity of the system is improved, the real-time performance is improved, the data transmission pressure is reduced, the delay of diagnosis results caused by poor network conditions is reduced, and the reliability of diagnosis results is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ultrasonic examination, in particular to a scanning thyroid system and method based on AI assistance. BACKGROUND

[0002] The existing ultrasonic equipment mainly includes two categories: high-resolution equipment used by professional medical institutions and portable equipment. The professional equipment is known for its high-resolution image acquisition capability, which can capture high-quality ultrasonic images in real time and quickly transmit image data to intelligent devices or servers for AI analysis through high-bandwidth and low-latency transmission technology. The AI analysis system uses a deep learning model to process a large amount of data in a short time and provides high-accuracy diagnostic results. However, these devices are expensive and ordinary users cannot afford them. In contrast, portable ultrasonic equipment is more affordable and easy to carry, but it usually only provides image acquisition functions, and users need to have certain professional knowledge to correctly use and interpret the images, which limits its popularity in non-professional fields.

[0003] In addition, whether it is a professional device or a portable device, there is a problem of data transmission and real-time performance. The transmission of high-resolution ultrasonic image data requires high-bandwidth and low-latency technical support, but existing network conditions may not meet this demand, resulting in data transmission delays and affecting the real-time performance of diagnosis. At the same time, the data transmission and analysis process is highly dependent on internet connection, and once the network condition is poor, data uploading and analysis will be affected, which may cause delays in diagnostic results.

[0004] In addition, for non-professionals, the operation of existing ultrasonic equipment is complex and it is difficult to accurately judge the image results, which also limits the widespread application of portable ultrasonic equipment. Therefore, the development of ultrasonic equipment needs to be improved and innovated to address these problems and improve the practicality and popularity of the equipment. SUMMARY

[0005] In view of the problems in the prior art, the present application provides a scanning thyroid system based on AI assistance, which comprises:

[0006] An image acquisition module connected to the probe of a portable ultrasonic scanning device, used to acquire a video stream containing multiple frames of thyroid ultrasonic images in the scanning process of a patient, divide the video stream into multiple video segments in chronological order and associate the time intervals;

[0007] an image distribution module connected to the image acquisition module, configured to extract an optimal thyroid ultrasound image from each of the video segments, upload the optimal thyroid ultrasound image to a cloud analysis system to obtain a corresponding cloud analysis result, and distribute each of the video segments to a plurality of AI analysis lightweight models configured in a host of a portable ultrasound scanning device to obtain a corresponding local analysis result;

[0008] a result output module connected to the image distribution module, configured to establish a cache queue, add the video segments having obtained the local analysis result and the cloud analysis result to the cache queue, adjust the order in the cache queue according to the associated time intervals, and then output each of the video segments in the cache queue and the associated local analysis result and cloud analysis result in sequence.

[0009] Preferably, the image acquisition module comprises:

[0010] a video segmentation unit configured to divide the video stream in sequence to obtain a plurality of video segments according to a preset initial interval length, and then associate the time interval of each video segment with the video segment by taking the corresponding system time of the thyroid ultrasound images at the beginning and end of the video segment as the time interval of the video segment.

[0011] Preferably, the image acquisition module further comprises an interval adjustment unit connected to the video segmentation unit, configured to obtain the image processing capability of the AI analysis lightweight model, and adjust the initial interval length according to the image processing capability.

[0012] The video segmentation unit divides the video stream in sequence to obtain a plurality of video segments according to the adjusted initial interval length.

[0013] Preferably, the interval adjustment unit comprises:

[0014] a storage subunit configured to store a standard video segment having an interval length equal to the initial interval length, the standard video segment being associated with a corresponding standard processing time length;

[0015] a capability testing subunit connected to the storage subunit, configured to input the standard video segment into the AI analysis lightweight model multiple times to obtain a plurality of test processing time lengths, and take the average of all the test processing time lengths as the image processing capability of the AI analysis lightweight model.

[0016] a length adjustment subunit connected to the capability testing unit, configured to calculate the ratio of the image processing capability to the standard processing time length as an adjustment ratio, multiply the initial interval length by the adjustment ratio to obtain the adjusted initial interval length.

[0017] Preferably, the image distribution module comprises:

[0018] a quality evaluation unit, configured to, for each video segment, take each frame of the thyroid ultrasound image contained in the video segment as a candidate image, and then perform a comprehensive score quality evaluation on each candidate image according to the clarity, contrast, noise, integrity and artifact to obtain a corresponding image quality score;

[0019] an image extraction unit connected to the quality evaluation unit, configured to take the candidate image with the highest image quality score as the optimal thyroid ultrasound image.

[0020] The application further provides a scanning thyroid method based on AI assistance, applied to the scanning thyroid system and comprising:

[0021] Step S1, the scanning thyroid system acquires a video stream containing multiple frames of thyroid ultrasound images in the scanning process of a patient, divides the video stream into multiple video segments in chronological order and associates a time interval with each video segment;

[0022] Step S2, the scanning thyroid system acquires, for each video segment, an optimal thyroid ultrasound image extracted from the video segment, uploads the optimal thyroid ultrasound image to a cloud analysis system to obtain a corresponding cloud analysis result, and respectively distributes each video segment to multiple AI analysis lightweight models configured in a host of a portable ultrasound scanning device to obtain a corresponding local analysis result;

[0023] Step S3, the scanning thyroid system acquires a cache queue, adds the video segments that have obtained the local analysis result and the cloud analysis result to the cache queue, adjusts the order of the video segments in the cache queue according to the associated time interval, and then sequentially outputs each video segment in the cache queue and its associated local analysis result and cloud analysis result.

[0024] Preferably, the step S1 comprises:

[0025] Step S11, the scanning thyroid system acquires a video stream containing multiple frames of thyroid ultrasound images in the scanning process of a patient by a probe of the portable ultrasound scanning device;

[0026] Step S12, the scanning thyroid system sequentially divides the video stream into multiple video segments according to a pre-set initial interval length, and then associates a time interval with each video segment by taking the corresponding system time of the thyroid ultrasound image at the beginning and end of the video segment as the time interval of the video segment.

[0027] Preferably, the step S1 further comprises, before the step S11 is performed:

[0028] The step S10 comprises:

[0029] The step S12 comprises, according to the adjusted initial interval length, dividing the video stream to obtain a plurality of video clips.

[0030] Preferably, the scanning thyroid system stores a standard video clip with an interval length equal to the initial interval length, and the standard video clip is associated with a corresponding standard processing time.

[0031] The step S10 comprises:

[0032] The step S101 comprises:

[0033] The step S102 comprises:

[0034] Preferably, the step S2 comprises:

[0035] The step S21 comprises:

[0036] The step S22 comprises:

[0037] The step S23 comprises:

[0038] The above technical solution has the following advantages or beneficial effects:

[0039] 1. The use of a portable ultrasound scanning device as the front end reduces the cost burden of professional high-resolution equipment, making it affordable for ordinary users to have thyroid ultrasound detection. At the same time, the combination of cloud and local AI analysis reduces the dependence on a single high-performance device, further indirectly reducing costs and improving the popularity of the system.

[0040] 2. The optimal thyroid ultrasound image is uploaded to the cloud analysis system through the image distribution module, which greatly reduces the network bandwidth requirement compared to transmitting the entire video stream, and reduces the dependence on network conditions. At the same time, multiple AI analysis lightweight models are used for parallel processing locally, and the rate is guaranteed through wired data transmission, improving real-time performance and reducing data transmission pressure.

[0041] 3. The result output module establishes a cache queue, adjusts the order of video segments according to time intervals, ensures the order of output results, and avoids output confusion caused by network delay. This design reduces the diagnostic result delay caused by poor network conditions, improves the reliability of the diagnostic result and the user experience. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 For the preferred embodiment of the present application, a result diagram of an AI-assisted thyroid scanning system is provided.

[0043] Figure 2 For the preferred embodiment of the present application, a flow diagram of an AI-assisted thyroid scanning method is provided.

[0044] Figure 3 For the preferred embodiment of the present application, a sub-flow diagram of step S1 is provided.

[0045] Figure 4 For the preferred embodiment of the present application, a sub-flow diagram of step S10 is provided.

[0046] Figure 5 For the preferred embodiment of the present application, a sub-flow diagram of step S2 is provided. DETAILED DESCRIPTION

[0047] The present application will be described in detail below in conjunction with the drawings and specific embodiments. The present application is not limited to this embodiment, and other embodiments that meet the spirit of the present application can also fall within the scope of the present application.

[0048] In the preferred embodiment of the present application, based on the above-mentioned problems existing in the prior art, an AI-assisted thyroid scanning system is provided, as shown in Figure 1 , comprising:

[0049] An image acquisition module 1 connected to a probe of a portable ultrasound scanning device, configured to acquire a video stream containing multiple frames of thyroid ultrasound images in a scanning process of a patient, divide the video stream into multiple video clips in chronological order and associate time intervals with the video clips;

[0050] An image distribution module 2 connected to the image acquisition module 1, configured to, for each video clip, extract an optimal thyroid ultrasound image from the video clip, upload the optimal thyroid ultrasound image to a cloud analysis system to obtain a corresponding cloud analysis result, and distribute each video clip to multiple AI analysis lightweight models configured in a host of the portable ultrasound scanning device to obtain a corresponding local analysis result;

[0051] A result output module 3 connected to the image distribution module 3, configured to establish a cache queue, add the video clips having obtained the local analysis result and the cloud analysis result to the cache queue, adjust the order of the video clips in the cache queue according to the associated time intervals, and then output the video clips in the cache queue and the associated local analysis results and cloud analysis results in sequence.

[0052] Specifically, high-resolution devices used by professional medical institutions are relatively expensive, which is difficult for ordinary users to afford. The system in the embodiment uses a portable ultrasound scanning device as the front end of image acquisition. Such devices are generally more affordable and easy to carry. At the same time, the system combines cloud and local AI analysis to improve diagnostic efficiency and reduce dependence on a single high-performance device, thereby indirectly reducing costs.

[0053] Specifically, existing ultrasound devices face data transmission and real-time issues, especially the transmission of high-resolution image data requires high-bandwidth and low-latency technical support. The cloud analysis system has strong computing power and rich data resources, and can perform in-depth analysis on the uploaded optimal thyroid ultrasound images to provide high-accuracy diagnostic results. In the embodiment, the image distribution module uploads the optimal thyroid ultrasound image in each video clip to the cloud analysis system. Since multiple single-frame images are uploaded to the cloud analysis system, compared with the existing ultrasound device that transmits video streams, only a small amount of network bandwidth is required, reducing the dependence on and stringent requirements for network conditions, and the influence of the network is smaller.

[0054] At the same time, multiple AI analysis lightweight models configured in the host of the portable device are used for local analysis. Through wired data transmission in the local area, the data transmission rate is guaranteed. This parallel processing method reduces the pressure of single data transmission and improves real-time performance.

[0055] The result output module establishes a cache queue, adjusts the order of the video segments according to the time intervals, ensures the order of the output results, avoids output disorder caused by network delay, and reduces diagnostic result delay caused by poor network conditions.

[0056] In a preferred embodiment of the present application, the image acquisition module 1 comprises:

[0057] The video segmentation unit 11 is configured to divide the video stream into multiple video segments according to the preset initial interval length, and then associate the time interval of each video segment with the video segment by taking the corresponding system time of the thyroid ultrasound images at the beginning and end of the video segment as the time interval of the video segment.

[0058] In a preferred embodiment of the present application, the image acquisition module further comprises an interval adjustment unit 12 connected to the video segmentation unit 11, configured to obtain the image processing capability of the AI analysis lightweight model, and adjust the initial interval length according to the image processing capability.

[0059] The video segmentation unit 11 divides the video stream into multiple video segments according to the adjusted initial interval length.

[0060] Specifically, in this embodiment, the image processing capability of the AI analysis lightweight model is obtained by the interval adjustment unit, which can dynamically adjust the initial interval length used in the video segmentation unit. This means that the system can optimize the division of video segments according to the performance of the actual analysis model, ensuring that the length of each video segment is neither too long (causing processing delay) nor too short (affecting image continuity).

[0061] The optimized video segment length can ensure that the AI analysis lightweight model can fully exert its performance when processing each segment, neither increasing the computational burden due to processing too many image frames, nor missing key information due to too few image frames. This helps to improve the efficiency and accuracy of analysis.

[0062] Different AI analysis lightweight models may have different image processing capabilities and processing speeds. Through the adjustment of the interval adjustment unit, the system can adapt to different analysis models, ensuring optimal analysis results in various scenarios.

[0063] After introducing the interval adjustment unit 12, the system can more flexibly cope with possible performance improvements or changes in the future. For example, when the AI analysis lightweight model is upgraded or a new more efficient model is introduced, the system can fully utilize these performance improvements by adjusting the initial interval length.

[0064] By dynamically adjusting the length of video segments, the system can more effectively utilize computing resources and bandwidth. For example, in the case of limited bandwidth, the system can reduce the number of image frames per video segment to reduce data transmission, thereby ensuring the real-time and stability of data transmission.

[0065] In summary, by introducing the video segmentation unit 11 and the interval adjustment unit 12 in the image acquisition module 1, the system of the present embodiment can dynamically adjust the length of video segments to adapt to the image processing capabilities of different AI analysis lightweight models. This not only improves the analysis efficiency and accuracy, but also enhances the flexibility and scalability of the system, providing better support for future performance improvement and changes. At the same time, this design also helps to optimize resource utilization, ensuring stable operation of the system in various scenarios.

[0066] In a preferred embodiment of the present application, the interval adjustment unit 12 includes:

[0067] The storage sub-unit 121 is used to store a standard video segment with an interval length equal to the initial interval length, and the standard video segment is associated with a corresponding standard processing time;

[0068] The capability testing sub-unit 122 is connected to the storage sub-unit 121 and is used to input the standard video segment multiple times into the AI analysis lightweight model to obtain multiple test processing times, and the average of all test processing times is taken as the image processing capability of the AI analysis lightweight model;

[0069] The length adjustment sub-unit 123 is connected to the capability testing sub-unit 122 and is used to calculate the ratio of the image processing capability to the standard processing time as an adjustment ratio, and multiply the initial interval length by the adjustment ratio to obtain the adjusted initial interval length.

[0070] Specifically, in the present embodiment, the standard video segment is input multiple times into the AI analysis lightweight model by the capability testing sub-unit, and the average of the multiple test processing times is calculated, which can more accurately evaluate the image processing capability of the AI analysis lightweight model. This evaluation method reduces the randomness of single test and improves the reliability of evaluation results.

[0071] The length adjustment sub-unit dynamically adjusts the initial interval length according to the image processing capability of the AI analysis lightweight model. This ensures that the length of each video segment can match the processing capability of the AI analysis lightweight model, neither causing processing delay due to too long segments nor affecting the integrity of image information due to too short segments.

[0072] By dynamically adjusting the length of video segments, the system can more effectively utilize computing resources and bandwidth. When the AI analysis lightweight model has strong processing capacity, the system can increase the length of the video segment, thereby analyzing more image frames in one processing and improving processing efficiency. Conversely, when the processing capacity is weak, the system can reduce the segment length to reduce the processing burden.

[0073] This design enables the system to adapt to AI analysis lightweight models with different performance. When the model is upgraded or performance is improved, the system can automatically adjust the length of the video segment to ensure that the analysis efficiency and accuracy are always at the optimal state.

[0074] The standard video segments and corresponding standard processing durations stored by the storage subunit 121 can be part of the system configuration, facilitating implementation and maintenance. When the system performance needs to be adjusted, only the standard video segments or the standard processing durations need to be updated.

[0075] In summary, by introducing the storage subunit 121, the capability testing subunit 122, and the length adjustment subunit 123 in the interval adjustment unit, the system of the present embodiment can accurately evaluate the image processing capacity of the AI analysis lightweight model and dynamically adjust the length of the video segment according to the evaluation results. This not only improves the analysis efficiency and accuracy, but also optimizes resource utilization, enhances the flexibility and scalability of the system. At the same time, this design also facilitates the implementation and maintenance of the system, providing better support for future performance improvement and changes.

[0076] In a preferred embodiment of the present application, the image distribution module 2 comprises:

[0077] The quality evaluation unit 21 is configured to, for each video segment, take each frame of thyroid ultrasound image contained therein as a candidate image, and then perform comprehensive scoring quality evaluation on each candidate image according to clarity, contrast, noise, integrity, and artifacts to obtain a corresponding image quality score.

[0078] The image extraction unit 22 is connected to the quality evaluation unit 21 and is configured to take the candidate image with the highest image quality score as the optimal thyroid ultrasound image.

[0079] Specifically, in the present embodiment, the quality evaluation unit performs comprehensive scoring quality evaluation on each frame of thyroid ultrasound image in each video segment, which ensures that the selected image is optimal in terms of clarity, contrast, noise control, integrity, and artifacts. Such an image is more easily analyzed and diagnosed accurately by the AI analysis lightweight model.

[0080] By directly selecting the image with the highest quality score as the optimal thyroid ultrasound image through the image extraction unit 22, the processing of a large number of redundant or low-quality images is avoided, thereby improving the analysis efficiency. This helps to reduce the waste of computing resources and speed up the diagnosis.

[0081] High-quality images are the basis for accurate diagnosis. The introduction of the quality evaluation unit and the image extraction unit ensures that the AI analysis lightweight model receives the best quality images, thereby improving the accuracy of diagnosis. This is crucial for the early detection and effective treatment of thyroid diseases.

[0082] The comprehensive scoring criteria in the quality evaluation unit can be adjusted according to actual needs to adapt to different application scenarios. For example, in some cases, clarity may be more important than other factors; while in other cases, more attention may be needed to the integrity of the image. This flexibility enables the system to be optimized according to different needs. Specifically, the user can select the factors to focus on before using the device, and the system can select the corresponding ultrasound image quality evaluation method based on the focus factors. A variety of ultrasound image quality evaluation methods available in the prior art can be used.

[0083] In summary, by introducing the quality evaluation unit and the image extraction unit in the image distribution module, the system of the present embodiment can automatically select the optimal thyroid ultrasound image for analysis and diagnosis. This not only improves the image quality and analysis efficiency, but also enhances the accuracy of diagnosis, providing strong support for the early detection and effective treatment of thyroid diseases. At the same time, this design also improves the degree of automation of the system, reduces the cost and risk of manual intervention, and provides more possibilities for future system upgrades and optimization.

[0084] The present application also provides an AI-assisted scanning method for the thyroid, which is applied to the scanning system for the thyroid as described above, as shown in the accompanying drawings, comprising: Figure 2 as shown, including:

[0085] Step S1, the scanning system for the thyroid obtains a video stream containing multiple thyroid ultrasound images in the scanning process of the patient, divides the video stream into multiple video segments in chronological order and associates the time interval;

[0086] Step S2, the scanning system for the thyroid obtains for each video segment, extracts the optimal thyroid ultrasound image from the video segment, uploads the optimal thyroid ultrasound image to the cloud analysis system to obtain the corresponding cloud analysis result, and distributes each video segment to the multiple AI analysis lightweight models configured in the host of the portable ultrasound scanning device to obtain the corresponding local analysis result;

[0087] Step S3, the scanning thyroid system obtains a cache queue, adds the video clip with the local analysis result and the cloud analysis result obtained to the cache queue, adjusts the order of the video clip in the cache queue according to the associated time interval, and then outputs the video clip in the cache queue and the associated local analysis result and cloud analysis result in sequence.

[0088] As shown in the preferred embodiment of the present application, Figure 3 Step S1 includes:

[0089] Step S11, the scanning thyroid system obtains a video stream containing multiple thyroid ultrasound images in the scanning process of the probe of the portable ultrasonic scanning device on the patient;

[0090] Step S12, the scanning thyroid system divides the video stream into multiple video clips according to the preset initial interval length in sequence, and then associates the time interval of the video clip with the video clip by taking the corresponding system time of the thyroid ultrasound images at the beginning and end of the video clip as the time interval of the video clip.

[0091] As shown in the preferred embodiment of the present application, Figure 3 Step S1 further includes:

[0092] Step S10, the scanning thyroid system obtains the image processing capability of the AI analysis lightweight model, and adjusts the initial interval length according to the image processing capability.

[0093] In step S12, the video stream is divided into multiple video clips according to the adjusted initial interval length.

[0094] In the preferred embodiment of the present application, the scanning thyroid system stores a standard video clip with an interval length equal to the initial interval length, and the standard video clip is associated with a corresponding standard processing time;

[0095] As shown in the preferred embodiment of the present application, Figure 4 Step S10 includes:

[0096] Step S101, the scanning thyroid system inputs the standard video clip into the AI analysis lightweight model multiple times to obtain multiple test processing times, and takes the average value of all test processing times as the image processing capability of the AI analysis lightweight model.

[0097] Step S102, the scanning thyroid system calculates the ratio of the image processing capability to the standard processing time as an adjustment ratio, and multiplies the initial interval length by the adjustment ratio to obtain the adjusted initial interval length.

[0098] As shown in the preferred embodiment of the present application, Figure 5 Step S2 includes:

[0099] Step S21, the thyroid scanning system takes each frame of thyroid ultrasound image contained in each video clip as a candidate image, and then performs comprehensive scoring quality evaluation on each candidate image according to the definition, contrast, noise, integrity and artifact to obtain a corresponding image quality score;

[0100] Step S22, the thyroid scanning system takes the candidate image with the highest image quality score as the optimal thyroid ultrasound image, and uploads the optimal thyroid ultrasound image to the cloud analysis system to obtain a corresponding cloud analysis result;

[0101] Step S23, the thyroid scanning system respectively assigns each video clip to a plurality of AI analysis lightweight models configured in the host of the portable ultrasound scanning device to obtain a corresponding local analysis result.

[0102] The above is only the preferred embodiment of the present application, and does not limit the implementation and protection scope of the present application. For those skilled in the art, it should be realized that any equivalent replacement and obvious changes made by applying the content of the present application and drawings should be included in the protection scope of the present application.

Claims

1. An AI-assisted based scanning thyroid system, characterized in that, The application relates to a system for thyroid ultrasound image analysis, comprising: an image acquisition module connected to a probe of a portable ultrasound scanning device, configured to acquire a video stream containing multiple thyroid ultrasound images during a scanning process of a patient, divide the video stream into multiple video clips in chronological order and associate time intervals with the video clips; an image distribution module connected to the image acquisition module, configured to extract an optimal thyroid ultrasound image from each video clip, upload the optimal thyroid ultrasound image to a cloud analysis system to obtain a corresponding cloud analysis result, and distribute each video clip to multiple AI analysis lightweight models configured in a host of the portable ultrasound scanning device to obtain a corresponding local analysis result; a result output module connected to the image distribution module, configured to establish a cache queue, add the video clips having obtained the local analysis result and the cloud analysis result to the cache queue, adjust the order of the video clips in the cache queue according to the associated time intervals, and then output the video clips in the cache queue and the associated local analysis result and cloud analysis result in sequence.

2. The thyroid scanning system of claim 1, wherein, The image acquisition module comprises: a video segmentation unit configured to divide the video stream into multiple video clips in sequence according to a preset initial interval length, and then associate the time interval of the first and last thyroid ultrasound images of each video clip with the video clip.

3. The thyroid scanning system of claim 2, wherein, The image acquisition module further comprises an interval adjustment unit connected to the video segmentation unit, configured to acquire the image processing capability of the AI analysis lightweight model, and adjust the initial interval length according to the image processing capability. The video segmentation unit divides the video stream into multiple video clips in sequence according to the adjusted initial interval length.

4. The thyroid scanning system of claim 3, wherein, The interval adjustment unit comprises: a storage subunit configured to store a standard video clip with an interval length equal to the initial interval length, and the standard video clip is associated with a corresponding standard processing time; a capability test subunit connected to the storage subunit, configured to input the standard video clip into the AI analysis lightweight model multiple times to obtain multiple test processing times, and take the average of all the test processing times as the image processing capability of the AI analysis lightweight model; a length adjustment subunit connected to the capability test subunit, configured to calculate the ratio of the image processing capability to the standard processing time as an adjustment ratio, and multiply the initial interval length by the adjustment ratio to obtain the adjusted initial interval length.

5. The thyroid scanning system of claim 1, wherein, The image distribution module comprises: a quality evaluation unit configured to take each frame of the thyroid ultrasound image contained in each video clip as a candidate image, and then perform comprehensive scoring quality evaluation on each candidate image according to the clarity, contrast, noise, integrity and artifacts to obtain a corresponding image quality score; an image extraction unit connected to the quality evaluation unit, configured to take the candidate image with the highest image quality score as the optimal thyroid ultrasound image.

6. An AI-assisted based method of scanning thyroid, characterized in that, Applied to the thyroid scanning system as claimed in any one of claims 1-5, comprising: Step S1, the thyroid scanning system acquires a video stream containing multiple thyroid ultrasound images in the scanning process of a patient, divides the video stream into multiple video clips in chronological order and associates a time interval; Step S2, the thyroid scanning system acquires each video clip, extracts the optimal thyroid ultrasound image from the video clip, uploads the optimal thyroid ultrasound image to the cloud analysis system to obtain the corresponding cloud analysis result, and assigns each video clip to the multiple AI analysis lightweight models configured in the host of the portable ultrasound scanning device to obtain the corresponding local analysis result; Step S3, the thyroid scanning system acquires a cache queue, adds the video clips that have obtained the local analysis result and the cloud analysis result to the cache queue, adjusts the order in the cache queue according to the associated time interval, and then outputs each video clip in the cache queue and its associated local analysis result and cloud analysis result in turn.

7. The method of scanning a thyroid according to claim 6, wherein, The step S1 comprises: Step S11, the thyroid scanning system acquires a video stream containing multiple thyroid ultrasound images in the scanning process of a patient by the probe of the portable ultrasound scanning device; Step S12, the thyroid scanning system divides the video stream into multiple video clips in turn according to the preset initial interval length, and then associates the time interval of each video clip with the video clip by taking the corresponding system time of the thyroid ultrasound images at the beginning and end of the video clip as the time interval of the video clip.

8. The method of claim 7, wherein the step of scanning the thyroid comprises, The step S1 further comprises, before performing the step S11: Step S10, the thyroid scanning system acquires the image processing capability of the AI analysis lightweight model, and adjusts the initial interval length according to the image processing capability; The step S12 divides the video stream into multiple video clips in turn according to the adjusted initial interval length.

9. The method of claim 8, wherein the step of scanning the thyroid comprises, The thyroid scanning system stores a standard video clip with an interval length equal to the initial interval length, and the standard video clip is associated with a corresponding standard processing time; The step S10 comprises: Step S101, the thyroid scanning system inputs the standard video clip multiple times into the AI analysis lightweight model to obtain multiple test processing times, and takes the average of all the test processing times as the image processing capability of the AI analysis lightweight model; Step S102, the thyroid scanning system calculates the ratio of the image processing capability to the standard processing time as an adjustment ratio, and multiplies the initial interval length by the adjustment ratio to obtain the adjusted initial interval length.

10. The method of claim 6, wherein the step of scanning the thyroid comprises, The step S2 comprises: Step S21, the scanning thyroid system takes each frame of the thyroid ultrasound image contained in each video segment as a candidate image, then scores the quality of each frame of the candidate image according to the clarity, contrast, noise, integrity and artifact to obtain the corresponding image quality score; Step S22, the scanning thyroid system takes the candidate image with the highest image quality score as the optimal thyroid ultrasound image, uploads the optimal thyroid ultrasound image to the cloud analysis system to obtain the corresponding cloud analysis result; Step S23, the scanning thyroid system respectively assigns each video segment to a plurality of AI analysis lightweight models configured in the host of the portable ultrasound scanning device to obtain the corresponding local analysis result.

Images