阅读说明：本技术 一种医学检测数据的数据处理方法及系统 (Data processing method and system for medical detection data ) 是由 许元 王俊 宋琛 于 2021-09-09 设计创作，主要内容包括：本发明提供的一种医学检测数据的数据处理方法及系统,涉及医学检测技术领域。在本发明中,首先,获取医学影像检测设备发送的多帧医学检测图像,其中,医学影像检测设备基于对目标对象进行图像采集得到多帧医学检测图像；其次,对多帧医学检测图像进行筛选,得到多帧医学检测图像对应的至少一帧目标医学检测图像；然后,基于预先训练得到的目标健康状态检测模型对至少一帧目标医学检测图像进行识别处理,得到目标对象的健康状态检测结果数据,其中,目标健康状态检测模型为神经网络模型,健康状态检测结果数据用于表征目标对象的健康状态。基于上述方法,可以改善现有技术中存在的健康检测结果的可靠性不佳的问题。(The invention provides a data processing method and a data processing system for medical detection data, and relates to the technical field of medical detection. Firstly, acquiring a plurality of frames of medical detection images sent by medical image detection equipment, wherein the medical image detection equipment acquires the plurality of frames of medical detection images based on image acquisition of a target object; secondly, screening multiple frames of medical detection images to obtain at least one frame of target medical detection image corresponding to the multiple frames of medical detection images; and then, identifying at least one frame of target medical detection image based on a target health state detection model obtained through pre-training to obtain health state detection result data of the target object, wherein the target health state detection model is a neural network model, and the health state detection result data is used for representing the health state of the target object. Based on the method, the problem that the reliability of the health detection result is poor in the prior art can be solved.)

1. A data processing method of medical detection data is applied to a medical data processing server, the medical data processing server is in communication connection with medical image detection equipment, and the data processing method of the medical detection data comprises the following steps:

acquiring a plurality of frames of medical detection images sent by the medical image detection equipment, wherein the medical image detection equipment acquires the plurality of frames of medical detection images based on image acquisition of a target object;

screening the multiple frames of medical detection images to obtain at least one frame of target medical detection image corresponding to the multiple frames of medical detection images;

and identifying the at least one frame of target medical detection image based on a target health state detection model obtained by pre-training to obtain health state detection result data of the target object, wherein the target health state detection model is a neural network model, and the health state detection result data is used for representing the health state of the target object.

2. The data processing method of medical examination data according to claim 1, wherein the step of acquiring a plurality of frames of medical examination images transmitted by the medical image examination apparatus comprises:

judging whether the health state detection processing is needed or not, and generating corresponding health state detection notification information when the health state detection processing is judged to be needed;

sending the health state detection notification information to the medical image detection device, wherein the medical image detection device is used for acquiring an image of a target object based on the health state detection notification information to obtain a plurality of frames of medical detection images and sending the medical detection images to the medical data processing server after receiving the health state detection notification information;

and acquiring the multi-frame medical detection image sent by the medical image detection equipment based on the health state detection notification information.

3. The data processing method of medical examination data according to claim 2, wherein the step of determining whether or not the health status detection process is required, and generating the corresponding health status detection notification information when it is determined that the health status detection process is required, includes:

judging whether health state detection request information sent by target user terminal equipment in communication connection is received, wherein the target user terminal equipment is used for generating the health state detection request information based on health state detection request operation performed by corresponding medical personnel;

if the health state detection request information sent by the target user terminal equipment is received, judging that health state detection processing is needed, and generating corresponding health state detection notification information;

and if the health state detection request information sent by the target user terminal equipment is not received, judging that the health state detection processing is required.

4. The method for processing medical examination data according to claim 1, wherein the step of screening the plurality of frames of medical examination images to obtain at least one target medical examination image corresponding to the plurality of frames of medical examination images comprises:

classifying the plurality of frames of medical detection images based on the shooting angle corresponding to each frame of medical detection image to obtain a plurality of detection image classification sets, wherein the shooting angles corresponding to any two frames of medical detection images in any one detection image classification set are the same, and the shooting angles corresponding to the medical detection images between any two detection image classification sets are different;

and screening the multiple frames of medical detection images included in the detection image classification set aiming at each detection image classification set to obtain at least one frame of target medical detection image corresponding to the detection image classification set.

5. The data processing method of medical test data according to claim 4, wherein the step of performing a screening process on a plurality of frames of medical test images included in each test image classification set to obtain at least one frame of target medical test image corresponding to the test image classification set includes:

for each detection image classification set, sequencing the multi-frame medical detection images included in the detection image classification set according to the corresponding image acquisition sequence to obtain a detection image sequence corresponding to the detection image classification set;

and for each detection image sequence, screening multiple frames of medical detection images included in the detection image sequence based on the sequencing relation among the medical detection images in the detection image sequence to obtain at least one frame of target medical detection image corresponding to the detection image sequence.

6. The method as claimed in any one of claims 1 to 5, wherein the step of performing recognition processing on the at least one frame of target medical examination image based on a pre-trained target health status detection model to obtain the health status detection result data of the target object comprises:

aiming at each frame of target medical detection image in the at least one frame of target medical detection image, carrying out identification processing on the target medical detection image based on a target health state detection model obtained by pre-training to obtain health state detection result data corresponding to the target medical detection image;

and determining health state detection result data of the target object based on the health state detection result data corresponding to each frame of the target medical detection image.

7. The method as claimed in claim 6, wherein the step of identifying, for each target medical detection image of the at least one target medical detection image, the target medical detection image based on a pre-trained target health status detection model to obtain the health status detection result data corresponding to the target medical detection image comprises:

acquiring a plurality of frames of medical detection sample images, wherein the plurality of frames of medical detection sample images comprise a plurality of frames of medical detection sample images corresponding to each sample object in a plurality of sample objects, the plurality of sample objects comprise a plurality of first sample objects and a plurality of second sample objects, a target part or a target organ corresponding to each first sample object belongs to a healthy state, the target part or the target organ corresponding to each second sample object belongs to an unhealthy state, the plurality of second sample objects at least have different unhealthy states corresponding to two second sample objects, a plurality of frames of medical detection sample images corresponding to each first sample object are acquired by image acquisition based on a plurality of shooting angles between the first sample image and the target part or the target organ, and a plurality of frames of medical detection sample images corresponding to each second sample object are acquired based on a plurality of frames of medical detection sample images between the second sample object and the target part or the target organ Acquiring images at a plurality of shooting angles;

classifying the plurality of frames of medical detection sample images to obtain a first sample image set and a second sample image set, wherein a sample object corresponding to each frame of medical detection sample image included in the first sample image set belongs to the first sample object, and a sample object corresponding to each frame of medical detection sample image included in the second sample image set belongs to the second sample object;

selecting a plurality of frames of medical detection sample images from the first sample image set and selecting a plurality of frames of medical detection sample images from the second sample image set based on preset proportion information, constructing to obtain a first sample image training set, and constructing to obtain a first sample image testing set based on other medical detection sample images except the first sample image training set;

training a target neural network model based on the first sample image training set to obtain a corresponding health state detection model, and testing the health state detection model based on the first sample image testing set to obtain a corresponding test result;

if the test result meets the preset test condition, determining the health state detection model as a target health state detection model, if the test result does not meet the preset test condition, reselecting a plurality of frames of medical detection sample images in the first sample image set based on the preset proportion information, reselecting a plurality of frames of medical detection sample images in the second sample image set, constructing to obtain a second sample image training set, constructing to obtain a second sample image test set based on other medical detection sample images except the first sample image training set, and until the test result obtained by testing the health state detection model trained based on the second sample image training set based on the second sample image test set meets the test condition;

and aiming at each frame of target medical detection image in the at least one frame of target medical detection image, identifying the target medical detection image based on the target health state detection model to obtain health state detection result data corresponding to the target medical detection image.

8. A data processing system of medical detection data is applied to a medical data processing server, the medical data processing server is connected with medical image detection equipment in a communication mode, and the data processing system of the medical detection data comprises:

the medical detection image acquisition module is used for acquiring a plurality of frames of medical detection images sent by the medical image detection equipment, wherein the medical image detection equipment acquires the plurality of frames of medical detection images based on image acquisition of a target object;

the medical detection image screening module is used for screening the plurality of frames of medical detection images to obtain at least one frame of target medical detection image corresponding to the plurality of frames of medical detection images;

and the medical detection image identification module is used for identifying the at least one frame of target medical detection image based on a target health state detection model obtained through pre-training to obtain health state detection result data of the target object, wherein the target health state detection model is a neural network model, and the health state detection result data is used for representing the health state of the target object.

9. The data processing system of medical examination data of claim 8, wherein the medical examination image screening module is specifically configured to:

classifying the plurality of frames of medical detection images based on the shooting angle corresponding to each frame of medical detection image to obtain a plurality of detection image classification sets, wherein the shooting angles corresponding to any two frames of medical detection images in any one detection image classification set are the same, and the shooting angles corresponding to the medical detection images between any two detection image classification sets are different;

and screening the multiple frames of medical detection images included in the detection image classification set aiming at each detection image classification set to obtain at least one frame of target medical detection image corresponding to the detection image classification set.

10. The data processing system of medical examination data of claim 8, wherein the medical examination image recognition module is specifically configured to:

aiming at each frame of target medical detection image in the at least one frame of target medical detection image, carrying out identification processing on the target medical detection image based on a target health state detection model obtained by pre-training to obtain health state detection result data corresponding to the target medical detection image;

and determining health state detection result data of the target object based on the health state detection result data corresponding to each frame of the target medical detection image.

Technical Field

The invention relates to the technical field of medical detection, in particular to a data processing method and system of medical detection data.

Background

After a normal physical examination or a disease or injury, the patient needs to go to the medical field to perform a corresponding examination, and then, based on the examination result, a treatment is performed, and the like. In the prior art, an important examination means is to acquire an image of a target part or a target organ of a patient or a user based on a medical image detection device to obtain a corresponding medical examination image, and then a doctor determines based on the acquired medical examination image, a learned medical instruction and corresponding experience to obtain whether the target part or the target organ of the patient or the user is healthy or not, and a specific pathological change condition when the target part or the target organ is unhealthy or not.

Based on this, since the health examination result obtained finally completely depends on the knowledge level and the treatment experience of the doctor, and different examination results may be obtained according to different subjective states of the doctor at that time, the problem of poor reliability of the health examination result is likely to occur.

Disclosure of Invention

In view of the above, an object of the present invention is to provide a data processing method and system for medical testing data, so as to solve the problem of poor reliability of the health testing result in the prior art.

In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:

a data processing method of medical detection data is applied to a medical data processing server, the medical data processing server is in communication connection with medical image detection equipment, and the data processing method of the medical detection data comprises the following steps:

acquiring a plurality of frames of medical detection images sent by the medical image detection equipment, wherein the medical image detection equipment acquires the plurality of frames of medical detection images based on image acquisition of a target object;

screening the multiple frames of medical detection images to obtain at least one frame of target medical detection image corresponding to the multiple frames of medical detection images;

and identifying the at least one frame of target medical detection image based on a target health state detection model obtained by pre-training to obtain health state detection result data of the target object, wherein the target health state detection model is a neural network model, and the health state detection result data is used for representing the health state of the target object.

In some preferred embodiments, in the above method for processing medical examination data, the step of acquiring multiple frames of medical examination images sent by the medical image examination device includes:

judging whether the health state detection processing is needed or not, and generating corresponding health state detection notification information when the health state detection processing is judged to be needed;

sending the health state detection notification information to the medical image detection device, wherein the medical image detection device is used for acquiring an image of a target object based on the health state detection notification information to obtain a plurality of frames of medical detection images and sending the medical detection images to the medical data processing server after receiving the health state detection notification information;

and acquiring the multi-frame medical detection image sent by the medical image detection equipment based on the health state detection notification information.

In some preferred embodiments, in the above data processing method of medical examination data, the step of determining whether or not health status detection processing is required, and generating corresponding health status detection notification information when it is determined that the health status detection processing is required, includes:

judging whether health state detection request information sent by target user terminal equipment in communication connection is received, wherein the target user terminal equipment is used for generating the health state detection request information based on health state detection request operation performed by corresponding medical personnel;

if the health state detection request information sent by the target user terminal equipment is received, judging that health state detection processing is needed, and generating corresponding health state detection notification information;

and if the health state detection request information sent by the target user terminal equipment is not received, judging that the health state detection processing is required.

In some preferred embodiments, in the above method for processing medical test data, the step of screening the plurality of frames of medical test images to obtain at least one frame of target medical test image corresponding to the plurality of frames of medical test images includes:

classifying the plurality of frames of medical detection images based on the shooting angle corresponding to each frame of medical detection image to obtain a plurality of detection image classification sets, wherein the shooting angles corresponding to any two frames of medical detection images in any one detection image classification set are the same, and the shooting angles corresponding to the medical detection images between any two detection image classification sets are different;

and screening the multiple frames of medical detection images included in the detection image classification set aiming at each detection image classification set to obtain at least one frame of target medical detection image corresponding to the detection image classification set.

In some preferred embodiments, in the data processing method of medical inspection data, the step of, for each inspection image classification set, performing a screening process on multiple frames of medical inspection images included in the inspection image classification set to obtain at least one frame of target medical inspection image corresponding to the inspection image classification set includes:

for each detection image classification set, sequencing the multi-frame medical detection images included in the detection image classification set according to the corresponding image acquisition sequence to obtain a detection image sequence corresponding to the detection image classification set;

and for each detection image sequence, screening multiple frames of medical detection images included in the detection image sequence based on the sequencing relation among the medical detection images in the detection image sequence to obtain at least one frame of target medical detection image corresponding to the detection image sequence.

In some preferred embodiments, in the above method for processing medical detection data, the step of performing recognition processing on the at least one frame of target medical detection image based on a target health status detection model obtained by pre-training to obtain health status detection result data of the target object includes:

aiming at each frame of target medical detection image in the at least one frame of target medical detection image, carrying out identification processing on the target medical detection image based on a target health state detection model obtained by pre-training to obtain health state detection result data corresponding to the target medical detection image;

and determining health state detection result data of the target object based on the health state detection result data corresponding to each frame of the target medical detection image.

In some preferred embodiments, in the above method for processing medical detection data, the step of, for each target medical detection image of the at least one frame of target medical detection image, performing recognition processing on the target medical detection image based on a target health status detection model obtained through pre-training to obtain health status detection result data corresponding to the target medical detection image includes:

acquiring a plurality of frames of medical detection sample images, wherein the plurality of frames of medical detection sample images comprise a plurality of frames of medical detection sample images corresponding to each sample object in a plurality of sample objects, the plurality of sample objects comprise a plurality of first sample objects and a plurality of second sample objects, a target part or a target organ corresponding to each first sample object belongs to a healthy state, the target part or the target organ corresponding to each second sample object belongs to an unhealthy state, the plurality of second sample objects at least have different unhealthy states corresponding to two second sample objects, a plurality of frames of medical detection sample images corresponding to each first sample object are acquired by image acquisition based on a plurality of shooting angles between the first sample image and the target part or the target organ, and a plurality of frames of medical detection sample images corresponding to each second sample object are acquired based on a plurality of frames of medical detection sample images between the second sample object and the target part or the target organ Acquiring images at a plurality of shooting angles;

classifying the plurality of frames of medical detection sample images to obtain a first sample image set and a second sample image set, wherein a sample object corresponding to each frame of medical detection sample image included in the first sample image set belongs to the first sample object, and a sample object corresponding to each frame of medical detection sample image included in the second sample image set belongs to the second sample object;

selecting a plurality of frames of medical detection sample images from the first sample image set and selecting a plurality of frames of medical detection sample images from the second sample image set based on preset proportion information, constructing to obtain a first sample image training set, and constructing to obtain a first sample image testing set based on other medical detection sample images except the first sample image training set;

training a target neural network model based on the first sample image training set to obtain a corresponding health state detection model, and testing the health state detection model based on the first sample image testing set to obtain a corresponding test result;

if the test result meets the preset test condition, determining the health state detection model as a target health state detection model, if the test result does not meet the preset test condition, reselecting a plurality of frames of medical detection sample images in the first sample image set based on the preset proportion information, reselecting a plurality of frames of medical detection sample images in the second sample image set, constructing to obtain a second sample image training set, constructing to obtain a second sample image test set based on other medical detection sample images except the first sample image training set, and until the test result obtained by testing the health state detection model trained based on the second sample image training set based on the second sample image test set meets the test condition;

and aiming at each frame of target medical detection image in the at least one frame of target medical detection image, identifying the target medical detection image based on the target health state detection model to obtain health state detection result data corresponding to the target medical detection image.

The embodiment of the present invention further provides a data processing system of medical detection data, which is applied to a medical data processing server, wherein the medical data processing server is connected with a medical image detection device in a communication manner, and the data processing system of medical detection data comprises:

the medical detection image acquisition module is used for acquiring a plurality of frames of medical detection images sent by the medical image detection equipment, wherein the medical image detection equipment acquires the plurality of frames of medical detection images based on image acquisition of a target object;

the medical detection image screening module is used for screening the plurality of frames of medical detection images to obtain at least one frame of target medical detection image corresponding to the plurality of frames of medical detection images;

and the medical detection image identification module is used for identifying the at least one frame of target medical detection image based on a target health state detection model obtained through pre-training to obtain health state detection result data of the target object, wherein the target health state detection model is a neural network model, and the health state detection result data is used for representing the health state of the target object.

In some preferred embodiments, in the data processing system for medical examination data, the medical examination image screening module is specifically configured to:

classifying the plurality of frames of medical detection images based on the shooting angle corresponding to each frame of medical detection image to obtain a plurality of detection image classification sets, wherein the shooting angles corresponding to any two frames of medical detection images in any one detection image classification set are the same, and the shooting angles corresponding to the medical detection images between any two detection image classification sets are different;

and screening the multiple frames of medical detection images included in the detection image classification set aiming at each detection image classification set to obtain at least one frame of target medical detection image corresponding to the detection image classification set.

In some preferred embodiments, in the data processing system for medical examination data, the medical examination image recognition module is specifically configured to:

aiming at each frame of target medical detection image in the at least one frame of target medical detection image, carrying out identification processing on the target medical detection image based on a target health state detection model obtained by pre-training to obtain health state detection result data corresponding to the target medical detection image;

and determining health state detection result data of the target object based on the health state detection result data corresponding to each frame of the target medical detection image.

The data processing method and system for medical detection data provided by the embodiment of the invention can firstly acquire multi-frame medical detection images acquired by image acquisition of a target object by medical image detection equipment, then screen the acquired multi-frame medical detection images to obtain at least one frame of target medical detection image, so that the target medical detection images can be identified and processed based on a target health state detection model obtained by pre-training to obtain health state detection result data of the target object And (5) performing comparison to obtain a more reliable detection result.

In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is a schematic block diagram of an application of a medical data processing server according to an embodiment of the present invention.

Fig. 2 is a schematic flow chart illustrating steps included in a data processing method for medical examination data according to an embodiment of the present invention.

Fig. 3 is a block diagram illustrating functional modules included in a data processing system for medical examination data according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1, an embodiment of the present invention provides a medical data processing server. Wherein the medical data processing server may comprise a memory and a processor.

In detail, the memory and the processor are electrically connected directly or indirectly to realize data transmission or interaction. For example, they may be electrically connected to each other via one or more communication buses or signal lines. The memory may have stored therein at least one software functional module, which may be in the form of software or firmware. The processor may be configured to execute the executable computer program stored in the memory, so as to implement the data processing method for medical examination data provided by the embodiment of the present invention, that is, implement the following steps to obtain corresponding health status examination result data:

acquiring a plurality of frames of medical detection images sent by medical image detection equipment, wherein the medical image detection equipment acquires the plurality of frames of medical detection images based on image acquisition of a target object; screening multiple frames of medical detection images to obtain at least one frame of target medical detection image corresponding to the multiple frames of medical detection images; and identifying at least one frame of target medical detection image based on a target health state detection model obtained by pre-training to obtain health state detection result data of the target object, wherein the target health state detection model is a neural network model, and the health state detection result data is used for representing the health state of the target object.

Alternatively, in one possible example, the Memory may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), a System on Chip (SoC), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.

Alternatively, in a possible example, the structure shown in fig. 1 is only an illustration, and the medical data processing server may further include more or less components than those shown in fig. 1, for example, may include a communication unit for information interaction with other devices.

With reference to fig. 2, an embodiment of the present invention further provides a data processing method for medical test data, which can be applied to the medical data processing server. The method steps defined by the flow related to the data processing method of the medical detection data can be realized by the medical data processing server, and the medical data processing server is in communication connection with a medical image detection device.

The specific process shown in FIG. 2 will be described in detail below.

And step S110, acquiring a plurality of frames of medical detection images sent by the medical image detection equipment.

In the embodiment of the present invention, the medical data processing server may first acquire a plurality of frames of medical detection images sent by the medical image detection device. The medical image detection equipment acquires a plurality of frames of medical detection images based on image acquisition of a target object.

And step S120, screening the plurality of frames of medical detection images to obtain at least one frame of target medical detection image corresponding to the plurality of frames of medical detection images.

In the embodiment of the present invention, the medical data processing server may screen the plurality of frames of medical detection images after acquiring the plurality of frames of medical detection images, so as to obtain at least one frame of target medical detection image corresponding to the plurality of frames of medical detection images.

Step S130, the at least one frame of target medical detection image is identified and processed based on the target health state detection model, and health state detection result data of the target object is obtained.

In the embodiment of the present invention, after obtaining the at least one frame of target medical detection image, the medical data processing server may perform recognition processing on the at least one frame of target medical detection image based on a target health status detection model obtained through pre-training, so as to obtain health status detection result data of the target object. The target health state detection model is a neural network model, and the health state detection result data is used for representing the health state of the target object.

Based on the method, a plurality of frames of medical detection images acquired by medical image detection equipment for acquiring images of a target object can be acquired, then at least one frame of target medical detection image is acquired by screening the acquired plurality of frames of medical detection images, so that the target medical detection images can be identified and processed based on a target health state detection model obtained by pre-training to acquire health state detection result data of the target object, and based on the health state detection result data, the acquired health detection result has higher reliability on the basis of ensuring that corresponding sample data is more sufficient by utilizing a mature neural network technology, thereby solving the problem of poor reliability of the health detection result in the prior art, and comparing the health state detection result data acquired based on the target health state detection model with a diagnosis and treatment result made by a doctor according to experience, thereby obtaining more reliable detection results.

Optionally, in a possible example, step S110 may include the following steps to acquire the plurality of frames of medical detection images:

step a01, determining whether a health status detection process is required, and generating corresponding health status detection notification information when it is determined that the health status detection process is required;

step a02, sending the health status detection notification information to the medical image detection device, where the medical image detection device is configured to send, after receiving the health status detection notification information, a multi-frame medical detection image obtained by image acquisition of a target object based on the health status detection notification information to the medical data processing server;

step a03, acquiring the multiple frames of medical detection images sent by the medical image detection device based on the health status detection notification information.

Optionally, in a possible example, a01 in the above example may include the following steps to determine whether the health status detection process needs to be performed:

step A011, judging whether health state detection request information sent by a target user terminal device in communication connection is received, wherein the target user terminal device is used for generating the health state detection request information based on health state detection request operation performed by corresponding medical personnel;

a step a012 of determining that a health state detection process is required and generating corresponding health state detection notification information, if the health state detection request information transmitted by the target user terminal device is received;

step A013, if the health status detection request information sent by the target user terminal device is not received, determining that the step needs to be performed with health status detection processing.

Optionally, in a possible example, the step S120 may include the following steps to obtain the at least one frame of target medical detection image:

step A11, classifying the multiple frames of medical detection images based on the shooting angle corresponding to each frame of medical detection image to obtain multiple detection image classification sets, wherein any two frames of medical detection images in any one detection image classification set have the same shooting angle, and the shooting angles corresponding to the medical detection images between any two detection image classification sets are different;

step A12, for each detection image classification set, performing screening processing on multiple frames of medical detection images included in the detection image classification set to obtain at least one frame of target medical detection image corresponding to the detection image classification set.

Optionally, in a possible example, a12 in the above example may include the following steps to obtain at least one frame of target medical detection image corresponding to each detection image classification set:

step A121, for each detection image classification set, sequencing multiple frames of medical detection images included in the detection image classification set according to the corresponding image acquisition sequence to obtain a detection image sequence corresponding to the detection image classification set;

step A122, for each detection image sequence, based on the ordering relationship between the medical detection images in the detection image sequence, performing screening processing on multiple frames of medical detection images included in the detection image sequence to obtain at least one frame of target medical detection image corresponding to the detection image sequence.

Optionally, in a possible example, a122 in the above example may include the following steps to obtain at least one frame of target medical detection image corresponding to each detection image sequence:

step a1221, for each of the detection image sequences, determining a frame of medical detection image with target identification information as a first medical detection image in the detection image sequence, and using other medical detection images as second medical detection images, where the target identification information is generated in response to a selection operation performed by a corresponding medical worker after the first medical detection image is acquired by the medical image detection device (for example, a frame that the medical worker considers to be optimal);

step a1222, for each of the detection image sequences, performing a target screening operation based on a relationship between the second medical detection image and the first medical detection image of each frame in the detection image sequence, so as to obtain at least one frame of target medical detection image corresponding to the detection image sequence.

Wherein the target screening operation may include:

step 1, determining each frame of the second medical detection image as a first medical detection image and a second medical detection image based on the image similarity between the first medical detection image and the second medical detection image (the image similarity between the first medical detection image and the first medical detection image of each frame is greater than or equal to a predetermined image similarity threshold value, and the image similarity between the second medical detection image and the first medical detection image of each frame is less than the image similarity threshold value);

step 2, determining a first type of target medical detection image in a plurality of frames of the first type of medical detection images based on the corresponding image acquisition sequence, wherein the image acquisition sequence corresponding to the first type of target medical detection image is closest to the image acquisition sequence corresponding to the first type of medical detection image in the image acquisition sequence corresponding to the first type of medical detection images in the plurality of frames of the first type of medical detection images;

step 3, determining a target second-class medical detection image from the plurality of frames of second-class medical detection images based on the corresponding image acquisition sequence, wherein the image acquisition sequence corresponding to the target second-class medical detection image is closest to the image acquisition sequence corresponding to the first medical detection image in the plurality of frames of second-class medical detection images;

step 4, determining an image acquisition sequence range to obtain an image acquisition sequence range based on the image acquisition sequence corresponding to the target first-class medical detection image and the image acquisition sequence corresponding to the target second-class medical detection image, counting the image frame numbers of the medical detection images included in the detection image sequence in the image acquisition sequence range, and determining the size relation between the image frame numbers and a predetermined frame number threshold;

step 5, if the image frame number is larger than or equal to the frame number threshold, determining the image acquisition sequence range as a target image acquisition sequence range;

step 6, if the number of the image frames is smaller than the frame number threshold, determining a first acquisition sequence range based on the image acquisition sequence corresponding to the plurality of frames of the first type of medical detection images, determining a second acquisition sequence range based on the image acquisition sequence corresponding to the plurality of frames of the second type of medical detection images, and determining an intersection range between the first acquisition sequence range and the second acquisition sequence range as a target image acquisition sequence range;

and 7, selecting a first medical detection image of which the corresponding image acquisition sequence belongs to the target image acquisition sequence range from the plurality of frames of first medical detection images as a target medical detection image, selecting a second medical detection image of which the corresponding image acquisition sequence belongs to the target image acquisition sequence range from the plurality of frames of second medical detection images as a target medical detection image, and taking the first medical detection image as the target medical detection image.

Optionally, in a possible example, the step S130 may include the following steps to obtain the health status detection result data of the target object:

step A31, aiming at each frame of target medical detection image in the at least one frame of target medical detection image, identifying the target medical detection image based on a pre-trained target health state detection model to obtain health state detection result data corresponding to the target medical detection image;

step a32, determining health status detection result data of the target object based on the health status detection result data corresponding to each frame of the target medical detection image (for example, as long as the health status detection result data corresponding to one frame of the target medical detection image represents that the health status of the target object is abnormal, the health status detection result data of the target object may be considered as abnormal, or the health status detection result data may be comprehensively judged by combining the diagnosis result of the medical staff).

Optionally, in a possible example, the step a31 in the above example may include the following steps to obtain the health status detection result data corresponding to each frame of the target medical detection image:

step a311, obtaining a plurality of frames of medical test sample images, where the plurality of frames of medical test sample images include a plurality of frames of medical test sample images corresponding to each of a plurality of sample objects, the plurality of sample objects include a plurality of first sample objects and a plurality of second sample objects, a target portion or a target organ corresponding to each of the first sample objects belongs to a healthy state, the target portion or the target organ corresponding to each of the second sample objects belongs to an unhealthy state, the plurality of second sample objects have at least two unhealthy states corresponding to the second sample objects to different degrees, the plurality of frames of medical test sample images corresponding to each of the first sample objects are obtained by image acquisition based on a plurality of shooting angles between the first sample image and the target portion or the target organ, and the plurality of frames of medical test sample images corresponding to each of the second sample objects are obtained by image acquisition based on a plurality of shooting angles between the second sample image and the target portion or the target organ Acquiring images at a plurality of shooting angles between the target organs;

step a312, classifying the multiple frames of medical detection sample images (based on corresponding sample objects) to obtain a first sample image set and a second sample image set, where a sample object corresponding to each frame of medical detection sample image included in the first sample image set belongs to the first sample object, and a sample object corresponding to each frame of medical detection sample image included in the second sample image set belongs to the second sample object;

a313, selecting a plurality of frames of medical detection sample images from the first sample image set and selecting a plurality of frames of medical detection sample images from the second sample image set based on preset proportion information, constructing to obtain a first sample image training set, and constructing to obtain a first sample image testing set based on other medical detection sample images except the first sample image training set;

step A314, training a target neural network model based on the first sample image training set to obtain a corresponding health state detection model, and testing the health state detection model based on the first sample image testing set to obtain a corresponding test result;

step A315, if the test result meets a preset test condition, determining the health state detection model as a target health state detection model, if the test result does not meet the preset test condition, reselecting multiple frames of medical detection sample images from the first sample image set based on the preset proportion information, reselecting multiple frames of medical detection sample images from the second sample image set, constructing a second sample image training set (which is not completely the same as the first sample image training set), and constructing a second sample image test set (which is not completely the same as the first sample image test set) based on other medical detection sample images except the first sample image training set until a test result obtained by testing the health state detection model trained based on the second sample image training set based on the second sample image test set is obtained The results satisfied the test conditions (cycle at least once);

step A316, for each frame of target medical detection image in the at least one frame of target medical detection image, performing identification processing on the target medical detection image based on the target health status detection model to obtain health status detection result data corresponding to the target medical detection image.

With reference to fig. 3, an embodiment of the present invention further provides a data processing system for medical examination data, which is applicable to the medical data processing server, where the medical data processing server is communicatively connected to a medical image examination device. The data processing system of the medical detection data can comprise a medical detection image acquisition module, a medical detection image screening module and a medical detection image identification module.

In detail, the medical detection image obtaining module is configured to obtain multiple frames of medical detection images sent by the medical image detection device, where the medical image detection device obtains the multiple frames of medical detection images based on image acquisition of a target object. The medical detection image screening module is used for screening the plurality of frames of medical detection images to obtain at least one frame of target medical detection image corresponding to the plurality of frames of medical detection images. The medical detection image recognition module is configured to perform recognition processing on the at least one frame of target medical detection image based on a target health state detection model obtained through pre-training to obtain health state detection result data of the target object, where the target health state detection model is a neural network model, and the health state detection result data is used to represent a health state of the target object.

Optionally, in a possible example, the medical examination image screening module is specifically configured to:

classifying the plurality of frames of medical detection images based on the shooting angle corresponding to each frame of medical detection image to obtain a plurality of detection image classification sets, wherein the shooting angles corresponding to any two frames of medical detection images in any one detection image classification set are the same, and the shooting angles corresponding to the medical detection images between any two detection image classification sets are different; and screening the multiple frames of medical detection images included in the detection image classification set aiming at each detection image classification set to obtain at least one frame of target medical detection image corresponding to the detection image classification set.

Optionally, in a possible example, the medical examination image recognition module is specifically configured to:

aiming at each frame of target medical detection image in the at least one frame of target medical detection image, carrying out identification processing on the target medical detection image based on a target health state detection model obtained by pre-training to obtain health state detection result data corresponding to the target medical detection image; and determining health state detection result data of the target object based on the health state detection result data corresponding to each frame of the target medical detection image.

In summary, the data processing method and system for medical detection data provided by the present invention can acquire a plurality of frames of medical detection images acquired by a medical image detection device for image acquisition of a target object, and then screen the acquired plurality of frames of medical detection images to obtain at least one frame of target medical detection image, so that the target medical detection image can be identified based on a target health status detection model obtained by pre-training, and health status detection result data of the target object can be obtained And comparing the results to obtain a more reliable detection result.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.