Data processing method, medical device, storage medium and computer program product

Abstract

The invention provides a data processing method, medical equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a target gestational week stage of a target fetus, and determining a plurality of disease categories corresponding to the target gestational week stage in a disease category database based on the target gestational week stage; acquiring disease characterization information of the target fetus, wherein the disease characterization information comprises one or more of a target section image and a disease description text of the target fetus; and for each disease category corresponding to the target gestational week stage, based on the disease characterization information of the target fetus, calculating the disease probability that the target fetus suffers from the disease of the disease category. And based on each disease probability, determining a final disease category in a plurality of disease categories corresponding to the target gestational week stage, and outputting a standard disease name corresponding to the final disease category. According to the invention, the realization difficulty of subsequent medical data statistics can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more specifically to a data processing method, a medical device, a storage medium and a computer program product. Background Art

[0002] In the medical field, accurate assessment of the health of the fetus is crucial to ensure the safety of mother and baby. With the advancement of medical technology, ultrasound examinations have become part of routine pregnancy examinations. Doctors can judge the type of fetal disease based on the collected ultrasound images and manually fill them into the fetal test report.

[0003] However, the naming of the same disease category may be reflected as different texts in the test report due to different naming habits of doctors. For example, if the left heart of the fetus is hypoplastic, it may be reflected in the test report as "hypoplastic left heart syndrome", "hypoplastic left heart", "aortic stenosis, small left heart", "aortic stenosis, small left ventricle, mitral stenosis", "aortic stenosis, small left ventricle, mitral atresia" and other texts. Therefore, when statistical medical data (such as the diagnosis rate of a specific disease category in a specific year, the total number of confirmed fetuses, etc.), it is difficult to automatically count the test reports. In addition, the use of different names for the same disease category in the test report is not conducive to improving the readability of the test report, which is not conducive to the hospital's quality inspection and management of the test report, and reduces the hospital's work efficiency. Summary of the invention

[0004] The present invention is proposed in view of the above problems. The present invention provides a data processing method, a medical device, a storage medium and a computer program product.

[0005] According to one aspect of the present invention, a data processing method is provided, the data processing method comprising: obtaining a target gestational age stage of a target fetus, and based on the target gestational age stage, determining multiple disease categories corresponding to the target gestational age stage in a disease database, wherein the disease database comprises multiple preset gestational age stages and multiple disease categories corresponding to each preset gestational age stage, and each disease category corresponds to a standard disease name; Acquiring disease representation information of the target fetus, wherein the disease representation information includes one or more of a target section image of the target fetus and a disease description text; For each disease category corresponding to the target gestational age stage, based on the disease representation information of the target fetus, the probability of the target fetus suffering from the disease of the disease category is calculated; Based on each disease probability, a final disease category is determined from multiple disease categories corresponding to the target gestational age stage, and its corresponding standard disease name is output.

[0006] Exemplarily, the disease representation information includes a text describing the disease of the target fetus; Correspondingly, for each disease category corresponding to the target gestational age stage, based on the disease characterization information of the target fetus, the probability of the target fetus suffering from the disease of the disease category is calculated, including: Obtaining first text information from the disease description text; Compare the first text information with the first preset information corresponding to each disease category corresponding to the target gestational age stage, so as to obtain a first similarity between the first text information and each first preset information; Each disease probability is determined based on each first similarity.

[0007] Exemplarily, the disease representation information of the target fetus also includes a target section image of the target fetus; Correspondingly, each disease probability is determined based on each first similarity, including for each disease category corresponding to the target gestational age stage, if its corresponding first similarity is less than the preset similarity threshold, its corresponding target section image is input into the preset first algorithm to calculate its corresponding disease probability; otherwise, its corresponding disease probability is determined based on its corresponding first similarity.

[0008] Exemplarily, after obtaining the first similarity between the first text information and each first preset information, the data processing method further includes: If each first similarity is less than a preset similarity threshold, obtaining second text information from the disease description text; For each second preset information of each disease category corresponding to the target gestational age stage, the second text information is compared with the second preset information to obtain a second similarity between the second text information and the second preset information; The determining of each disease probability based on each first similarity is replaced by determining each disease probability based on each second similarity.

[0009] Exemplarily, the disease representation information of the target fetus includes a target section image of the target fetus; Correspondingly, for each disease category corresponding to the target gestational age stage, the probability of the target fetus suffering from a disease of that disease category is calculated based on the disease characterization information of the target fetus, including: for each disease category corresponding to the target gestational age stage, the corresponding target cross-sectional image is input into a preset first algorithm to calculate the corresponding disease probability.

[0010] Exemplarily, the target section image includes a plurality of target section images belonging to a plurality of preset section types, and the disease database further includes: for each disease category, at least one preset section type corresponding to the disease category and a section weight of each preset section type corresponding to the disease category, wherein each section weight is used to indicate the importance of the target section image of the preset section type as a basis for calculating the probability of illness of the disease category; Correspondingly, for each disease category corresponding to the target gestational age stage, the corresponding target section image is input into a preset first algorithm to calculate the corresponding disease probability, including: For each preset section type corresponding to each disease category corresponding to the target gestational age stage, inputting the target section image of the preset section type into a preset first algorithm, and calculating a first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type; Based on the section weights of each preset section type corresponding to the disease category and each first probability of the disease category, the probability of the target fetus suffering from the disease of the disease category is determined.

[0011] Exemplarily, the preset first algorithm includes a first detection model; Correspondingly, the target section image of the preset section type is input into a preset first algorithm to calculate the first probability that the target fetus suffers from the disease of the disease category in the target section image of the preset section type, including: inputting the target section image of the preset section type into a first detection model corresponding to the preset section type to obtain the first probability that the target fetus suffers from the disease of the disease category in the target section image of the preset section type, wherein the first detection model corresponding to the preset section type is trained based on the training image corresponding to the disease category, and the first detection model corresponding to the preset section type is different in different disease categories.

[0012] Exemplarily, the first detection model is trained by the following steps: For each disease category in each preset gestational age stage, a training image of a baseline fetus of a preset section type corresponding to the disease category at the preset gestational age stage in the training data set is input into the first initial model for training, and the model parameters of the first initial model are adjusted based on the first label of the training image to obtain a first detection model, wherein the first label is used to indicate whether the actual disease category of the training image is the disease category corresponding to the training image.

[0013] Exemplarily, the preset first algorithm includes a second detection model; Correspondingly, the target section image of the preset section type is input into a preset first algorithm, and a first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type is calculated, including: The target section image of the preset section type is input into a second detection model corresponding to the preset section type to obtain a first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type, wherein the second detection model corresponding to the preset section type is trained based on multiple training images corresponding to each disease category including at least the preset section type.

[0014] Exemplarily, the second detection model is trained by the following steps: For each disease category in each preset gestational age stage, a training image of a baseline fetus of a preset section type corresponding to the disease category at the preset gestational age stage in the training data set is input into the second initial model for training, and the model parameters of the second initial model are adjusted based on the second label of the training image to obtain a second detection model, wherein the second label is used to represent the actual disease category of the training image.

[0015] Exemplarily, the above data processing method also includes: calculating the target gestational age stage based on the target section image according to a preset second algorithm.

[0016] Exemplarily, based on each disease probability, a final disease category is determined from a plurality of disease categories corresponding to the target gestational age stage, including: The disease category corresponding to the maximum disease probability among all disease probabilities is determined as the final disease category.

[0017] Exemplarily, the corresponding standard disease names are output, including: Displaying the standard disease name corresponding to the final disease category in the first interface, wherein the first interface also includes a test report of the target fetus; In response to the user selecting a standard disease name in the first interface, the standard disease name is filled into the area corresponding to the disease category displayed in the test report.

[0018] According to another aspect of the present invention, there is also provided a medical device, the medical device comprising a signal generating device and a processor for executing the above data processing method; The signal generating device is used to transmit a preset signal and receive an echo signal generated by the preset signal acting on the target fetus; The processor is further configured to generate a target section image of the target fetus based on the echo signal.

[0019] According to yet another aspect of the present invention, a storage medium is provided, storing computer program instructions, wherein the computer program instructions are used to execute the data processing method when running.

[0020] According to yet another aspect of the present invention, a computer program product is provided, comprising computer program instructions, wherein the computer program instructions are used to execute the data processing method when executed by a processor.

[0021] According to the above scheme of the embodiment of the present invention, the target gestational age stage of the target fetus can be obtained, and based on the target gestational age stage, multiple disease categories corresponding to the target gestational age stage can be determined in the disease database. Then, the disease characterization information of the target fetus is obtained. For each disease category corresponding to the target gestational age stage, the probability of the target fetus suffering from the disease of the disease category is calculated based on the disease characterization information of the target fetus. Finally, based on each probability of illness, a final disease category is determined from the multiple disease categories corresponding to the target gestational age stage, and the corresponding standard disease name is output. By obtaining the target gestational age stage, the above scheme can determine the final disease category of the target fetus from the multiple disease categories corresponding to the target gestational age stage, and then generate a more accurate standard disease name. Specifically, if the actual diseases suffered by different fetuses are the same, the standard disease names corresponding to the above different fetuses are also the same, which is conducive to reducing the difficulty of implementing subsequent medical data statistics. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other purposes, features and advantages of the present invention will become more apparent by describing the embodiments of the present invention in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0023] Figure 1 A schematic flow chart of a data processing method according to an embodiment of the present invention is shown; Figure 2 A schematic block diagram of a medical device according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical scheme and advantages of the present invention more obvious, the exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described in the present invention, all other embodiments obtained by those skilled in the art without creative work should fall within the protection scope of the present invention.

[0025] In order to at least partially solve the above problems, an embodiment of the present invention provides a data processing method. Figure 1 FIG. 1 is a schematic flow chart of a data processing method according to an embodiment of the present invention. Figure 1 As shown, the above data processing method may include steps S110 to S140.

[0026] In step S110, the target gestational age of the target fetus is obtained, and based on the target gestational age, a plurality of disease categories corresponding to the target gestational age are determined in a disease database.

[0027] The target fetus may be a fetus that needs medical examination, for example, a fetus that needs ultrasound examination or magnetic resonance examination. The target gestational age stage may be the gestational age stage that the target fetus is currently in. In one example, it can be extracted based on the relevant text of the target gestational age stage filled in the test report. In another example, the target gestational age stage can also be obtained by the user through input devices (such as mouse, keyboard, etc.).

[0028] The disease database may include multiple preset gestational stages and multiple disease categories corresponding to each preset gestational stage, and each disease category corresponds to a standard disease name. The above-mentioned preset gestational stages, disease categories, and standard disease names can be maintained by developers or users, and the embodiments of the present invention are not described in detail here. In combination with actual scenarios, the multiple preset gestational stages included in the disease database can be any gestational stage that the fetus may be in. For example, the disease database may include: preset gestational stages from week 1 to week 12, preset gestational stages from week 13 to week 27, preset gestational stages from week 28 to week 40, etc. The specific duration and total number of gestational stages can be set by developers or users in combination with actual conditions, and the embodiments of the present invention are not limited here. The disease categories corresponding to the preset gestational stages can be used to indicate the categories of diseases that the fetus may suffer from when it is in the preset gestational stage, or the categories of diseases that are more likely to suffer from. The standard disease names of disease categories can be used to indicate the unified and standardized disease names of the disease categories. The standard disease name can be filled into the test report of the target fetus during the test phase, or updated into the test report during the report maintenance phase, so that the names used in the test reports for diseases that are actually in the same disease category are also consistent.

[0029] In step S120, disease representation information of the target fetus is obtained.

[0030] The disease representation information may be data related to the disease suffered by the target fetus. For example, the disease representation information includes one or more of a target section image of the target fetus and a disease description text.

[0031] The target section image may be an image obtained by detecting the target fetus using ultrasound technology or magnetic resonance technology in the relevant technology. The target section image may be an image of the target fetus on the section that the user needs to observe. For example, the section that the user needs to observe may include an abdominal cross section, a spinal sagittal plane, a spinal cross section, a head cross section, etc., which may be determined according to the actual needs of the user.

[0032] The above-mentioned disease description text may include text related to the disease that the target fetus may suffer from. For example, it may include text on the category of the disease that the target fetus may suffer from, text related to the symptoms of the pregnant woman, and description text of the visual features of the target section image of the target fetus, etc. The above information can be obtained based on the content filled in the test report of the target fetus. Specifically, the text extraction algorithm in the relevant technology can be used to extract text from the area in the test report that has been filled with the disease description text to obtain the above-mentioned disease description text.

[0033] In step S130, for each disease category corresponding to the target gestational age stage, based on the disease characterization information of the target fetus, the probability of the target fetus suffering from a disease of the disease category is calculated.

[0034] In one example, for each disease category corresponding to the target gestational age stage, the probability of the target fetus suffering from the disease of the disease category can be determined based on the disease characterization information of the target fetus through a preset detection algorithm for detecting the disease category. In another example, the disease characterization information can also be input into a detection model for detecting the disease category in the related art to obtain the probability of the target fetus suffering from the disease of the disease category. Specifically, if the above-mentioned disease characterization information is a disease description text, the disease description text can be input into a detection model that can classify the text to obtain the above-mentioned disease probability; if the above-mentioned disease characterization information is a target section image, the target section image can be input into a detection model that can classify the image to obtain the above-mentioned disease probability. If the above-mentioned disease characterization information includes both the disease description text and the target section image, the disease description text can be input into a detection model that can classify the text to obtain the disease probability based on the text, and the target section image can be input into a detection model that can classify the image to obtain the disease probability based on the image. The probability of the target fetus suffering from a disease of the disease category can be determined by combining the disease probability obtained based on the text and the disease probability obtained based on the image. For example, the highest disease probability between the disease probability obtained based on the text and the disease probability obtained based on the image can be used as the disease probability of the target fetus suffering from the disease category. For another specific example, the average value or the weighted average value between the disease probability obtained based on the text and the disease probability obtained based on the image can be used as the disease probability of the target fetus suffering from the disease category. The specific combination rules can be determined by the user based on actual conditions, and the embodiments of the present invention are not limited here.

[0035] In step S140, based on each disease probability, a final disease category is determined from a plurality of disease categories corresponding to the target gestational age stage, and the corresponding standard disease name is output.

[0036] In one example, the disease category with the highest probability of disease can be directly used as the above-mentioned final disease category. In another example, the disease category corresponding to the target gestational age stage can be combined with preset rules to screen out the disease categories that the user does not need to check this time (for example, the pregnant woman has undergone relevant tests for the disease category) to determine a final disease category from the multiple disease categories after screening. For example, the disease category with the highest probability of disease among the multiple disease categories after screening can be used as the above-mentioned final disease category. In another example, the actual probability of occurrence of the disease category (which can be obtained by the user based on actual experience) can also be combined to determine the weight value corresponding to each disease category, and then the weight value corresponding to the disease category is multiplied by the probability of disease corresponding to the disease category. The disease category with the highest probability of disease after multiplication can be used as the above-mentioned final disease category. In this example, the final disease category can also be determined in combination with the preset rules in the previous example, and the embodiments of the present invention will not be described in detail here.

[0037] In one example, the standard disease name can be output to the display screen to inform the user of the standard disease name of the disease suffered by the target fetus. The user can fill in the test report of the target fetus based on this. In another example, the standard disease name can also be automatically filled in the corresponding area of ​​the test report of the target fetus to reduce labor costs.

[0038] According to the above scheme of the embodiment of the present invention, the target gestational age stage of the target fetus can be obtained, and based on the target gestational age stage, multiple disease categories corresponding to the target gestational age stage are determined in the disease database. Then, the disease characterization information of the target fetus is obtained. For each disease category corresponding to the target gestational age stage, the probability of the target fetus suffering from the disease of the disease category is calculated based on the disease characterization information of the target fetus. Finally, based on each probability of illness, a final disease category is determined from the multiple disease categories corresponding to the target gestational age stage, and the corresponding standard disease name is output. By obtaining the target gestational age stage, the above scheme can determine the final disease category of the target fetus from the multiple disease categories corresponding to the target gestational age stage, and then generate a more accurate standard disease name. Specifically, if the actual diseases suffered by different fetuses are the same, the standard disease names corresponding to the above different fetuses are also the same, which is conducive to reducing the difficulty of implementing subsequent medical data statistics and facilitating quality inspection or other management of test reports.

[0039] For example, the disease representation information may include a text describing the disease of the target fetus.

[0040] Correspondingly, step S130, for each disease category corresponding to the target gestational age stage, based on the disease characterization information of the target fetus, calculates the probability of the target fetus suffering from a disease of the disease category, which may include steps S131a to S133a.

[0041] In step S131a, first text information is obtained from the disease description text.

[0042] The first text information may be a part of the disease description text. Specifically, for example, the disease description text may be segmented into sentences or words, and the sentences or words obtained by segmentation are used as the first text information.

[0043] In step S132a, the first text information is compared with the first preset information corresponding to each disease category corresponding to the target gestational age stage to obtain a first similarity between the first text information and each first preset information.

[0044] The above comparison processing may be the calculation of cosine similarity, Jaccard similarity, word embedding similarity, etc. in the related art, which may be determined by the developer or user according to the actual situation, and the embodiment of the present invention does not limit this.

[0045] The first preset information corresponding to the above disease category may be text information related to the disease category. For example, text of the disease category that the reference fetus (which may be the same as or different from the target fetus and is used in the stage of establishing the disease database) may suffer from, text related to the symptoms of pregnant women with the reference fetus, and text describing the visual features of each cross-sectional image of the reference fetus. The above first preset information can be pre-saved by the developer or user in the above disease database.

[0046] In step S133a, each disease probability is determined based on each first similarity.

[0047] For the first similarity between the first text information and each first preset information, in one example, the first similarity can be directly used as the probability that the target fetus suffers from a disease of the disease category corresponding to the first preset information. In another example, if the disease category corresponds to multiple first preset information, the first preset information with the highest first similarity to the first text information among the multiple first preset information can be used as the probability that the target fetus suffers from the disease of the disease category. In another example, if the disease description text includes multiple first text information, the first similarity between the first text information and each first preset information can be determined for each first text information. Then, the highest first similarity can be used as the target similarity. Finally, the highest target similarity among the target similarities of all the first text information can be used as the probability that the target fetus suffers from a disease of the target disease category. The target disease category is the disease category corresponding to the first preset information with the highest target similarity.

[0048] According to the above scheme of an embodiment of the present invention, when the disease characterization information includes the disease description text of the target fetus, the first text information can be obtained from the disease description text. Then, the first text information is compared with the first preset information corresponding to each disease category corresponding to the target gestational age stage, so as to obtain the first similarity between the first text information and each first preset information. Finally, each disease probability is determined based on each first similarity. The above scheme can use the first text information to determine the probability of the target fetus suffering from the disease category corresponding to the target gestational age stage. Since the processed data is at the text level, the data processing amount is less, so the generation efficiency of the disease probability can be improved, which is conducive to improving the output efficiency of the standard disease name.

[0049] Exemplarily, the disease characterization information of the target fetus also includes a target cross-sectional image of the target fetus.

[0050] Correspondingly, step S133a, determining each disease probability based on each first similarity, includes step S133a1.

[0051] In step S133a1, for each disease category corresponding to the target gestational age stage, if its corresponding first similarity is less than the preset similarity threshold, its corresponding target section image is input into the preset first algorithm to calculate its corresponding disease probability; otherwise, its corresponding disease probability is determined based on its corresponding first similarity.

[0052] Combined with the actual scene, since the target section image is actually collected for the target fetus, its representativeness is higher than the first text information. If the first similarity between each disease category and the first text information is less than the similarity threshold (the specific value can be set by the developer or user according to the actual situation), it can be regarded as that the comparison at the text level cannot determine the accurate probability of illness. Therefore, the probability of illness can be determined at the image level. Specifically, the target section image corresponding to the disease category can be input into the preset first algorithm to obtain the probability of the target fetus suffering from the disease of the disease category. The target section image corresponding to the disease category can be the section image required for detecting the disease category, which can be set by the developer or user according to the actual situation. For example, if the spinal disease category pays more attention to the sagittal plane of the spine, the target section image of the sagittal plane of the spine can be used as the target section image corresponding to the disease category.

[0053] In one example, the above-mentioned preset first algorithm may be an image comparison algorithm in the related art. For each disease category, the disease category may correspond to a standard section image of a reference fetus. The standard section image is compared with the target section image to obtain a similarity. The similarity may be used as the probability that the target fetus suffers from a disease of the disease category. In another example, the above-mentioned preset first algorithm may be a detection model for detecting the disease category, and the target section image is input into the detection model to obtain the confidence of the detection model that the category of the target section image is the disease category. The confidence may be used as the probability that the target fetus suffers from a disease of the disease category.

[0054] For specific content of determining the corresponding disease probability based on the corresponding first similarity, reference may be made to the relevant content in step S133a, which will not be elaborated in detail in this embodiment of the present invention.

[0055] According to the above scheme of the embodiment of the present invention, the disease characterization information of the target fetus may include the disease description text of the target fetus and the target section image of the target fetus. For each disease category corresponding to the target gestational age stage, if the corresponding first similarity is less than the preset similarity threshold, the corresponding target section image may be input into the preset first algorithm to calculate the corresponding disease probability; otherwise, the corresponding disease probability may be determined based on the corresponding first similarity. The above scheme may use the target section image to determine the above disease probability when the first similarity is less than the similarity threshold, thereby improving the accuracy of the disease probability, which is conducive to improving the accuracy of the output standard disease name.

[0056] Exemplarily, in step S132a, after the first text information is compared with the first preset information corresponding to each disease category corresponding to the target gestational age stage to obtain the first similarity between the first text information and each first preset information, the above data processing method also includes: steps S133b to S135b.

[0057] In step S133b, if each first similarity is less than a preset similarity threshold, second text information is obtained from the disease description text.

[0058] The second text information is different from the first text information, and the representativeness of the second text information may be lower than that of the first text information. For example, the first text information may be the name of the disease, and the second text information may be what is seen in the image. For another example, the first text information may be what is seen in the image, and the second text information may be a description of the symptoms of the pregnant woman. What is seen in the above-mentioned image may be what is seen in the ultrasound or MRI in the relevant technology, and can be used to summarize the visual characteristics of the target section image. For example, the cross-section is oval and the liver is visible, and no pressure marks are seen on the fetal neck. Combined with the actual scene, here we take the first text information as the name of the disease and the second text information as what is seen in the image as an example. Matching the name of the disease for the target fetus is more representative than matching what is seen in the image, so the first text information can be used preferentially to match the first preset information.

[0059] In step S134b, for each second preset information of each disease category corresponding to the target gestational age stage, the second text information is compared with the second preset information to obtain a second similarity between the second text information and the second preset information.

[0060] The second preset information can be used to represent text information related to the disease category, which may be less representative than the first preset information, such as text related to the image or symptoms of pregnant women mentioned above. The second preset information can be pre-saved by the developer or user in the disease database mentioned above.

[0061] The above comparison processing may be the calculation of cosine similarity, Jaccard similarity, word embedding similarity, etc. in the related art, which may be determined by the developer or user according to the actual situation, and the embodiment of the present invention does not limit this.

[0062] In step S135b, step S133a, determining each disease probability based on each first similarity, is replaced by determining each disease probability based on each second similarity.

[0063] For the second similarity between the second text information and each second preset information, in one example, the second similarity can be directly used as the probability that the target fetus suffers from a disease of the disease category corresponding to the second preset information. In another example, if the disease category corresponds to multiple second preset information, the second preset information with the highest second similarity to the second text information among the multiple second preset information can be used as the probability that the target fetus suffers from the disease of the disease category. In another example, if the disease description text includes multiple second text information, the second similarity between the second text information and each second preset information can be determined for each second text information. Then, the highest second similarity can be used as the target similarity. Finally, the highest target similarity among the target similarities of all the second text information can be used as the probability that the target fetus suffers from a disease of the target disease category. The target disease category is the disease category corresponding to the second preset information with the highest target similarity.

[0064] According to the above scheme of the embodiment of the present invention, if each first similarity is less than the preset similarity threshold, the processor can obtain the second text information from the disease description text. Then, for each second preset information of each disease category corresponding to the target gestational age stage, the second text information is compared with the second preset information to obtain the second similarity between the second text information and the second preset information. Finally, each probability of illness is determined based on each second similarity. In the above scheme, if each first similarity is less than the preset similarity threshold, it can be regarded as that the representativeness of the first similarity is low, so the second text information different from the first text information can be used to match the second preset information, which can improve the representativeness of the obtained probability of illness, and is conducive to improving the accuracy of the output standard disease name.

[0065] Exemplarily, the disease representation information of the target fetus may include a target cross-sectional image of the target fetus.

[0066] Correspondingly, step S130, for each disease category corresponding to the target gestational age stage, based on the disease characterization information of the target fetus, calculates the probability that the target fetus suffers from a disease of the disease category, which may include step S131c.

[0067] In step S131c, for each disease category corresponding to the target gestational age stage, the corresponding target section image is input into a preset first algorithm to calculate the corresponding disease probability.

[0068] The target section image corresponding to the disease category may be the section image required for detecting the disease category, which may be set by the developer or user according to the actual situation. For example, if the spinal disease category focuses more on the sagittal plane of the spine, the target section image of the sagittal plane of the spine may be used as the target section image corresponding to the disease category.

[0069] In one example, the above-mentioned preset first algorithm may be an image comparison algorithm in the related art. For each disease category, the disease category may correspond to a standard section image of a reference fetus. The standard section image is compared with the target section image to obtain a similarity. The similarity may be used as the probability that the target fetus suffers from a disease of the disease category. In another example, the above-mentioned preset first algorithm may be a detection model for detecting the disease category, and the target section image is input into the detection model to obtain the confidence of the detection model that the category of the target section image is the disease category. The confidence may be used as the probability that the target fetus suffers from a disease of the disease category.

[0070] According to the above method of an embodiment of the present invention, when the disease characterization information of the target fetus includes the target section image of the target fetus, for each disease category corresponding to the target gestational age stage, the corresponding target section image can be input into a preset first algorithm to calculate the corresponding disease probability. The above scheme can determine the disease probability of each disease category suffered by the target fetus based on the target section image, thereby improving the accuracy of the disease probability, which is conducive to improving the accuracy of the output standard disease name.

[0071] Exemplarily, if the disease characterization information of the target fetus includes at least a target cross-sectional image of the target fetus, the above-mentioned data processing method further includes: calculating the target gestational age stage based on the target cross-sectional image according to a preset second algorithm.

[0072] In one example, the above-mentioned preset second algorithm may be an image comparison algorithm in the related art. For each gestational stage, the gestational stage may correspond to a standard section image of a reference fetus. The standard section image is compared with the target section image to obtain a similarity. The gestational stage with the highest similarity in each gestational stage may be used as the target gestational stage of the target fetus. In another example, the above-mentioned preset second algorithm may be a detection model for detecting the gestational stage, and the target section image is input into the detection model to obtain the confidence of the detection model for each gestational stage in which the fetus in the target section image is located. The gestational stage with the highest confidence in each gestational stage may be used as the target gestational stage of the target fetus.

[0073] According to the above solution of the embodiment of the present invention, the target gestational age stage can be calculated based on the target section image according to the preset second algorithm. The above solution can automatically determine the target gestational age stage of the target fetus, which can reduce labor costs.

[0074] For example, the target section image may include multiple target section images belonging to multiple preset section types. The preset section types may be the types of sections that the user needs to observe, such as the abdominal cross section, spinal sagittal plane, spinal cross section, head cross section, etc. mentioned above.

[0075] The disease database also includes: for each disease category, at least one preset section type corresponding to the disease category and a section weight of each preset section type corresponding to the disease category. Each section weight is used to indicate the importance of the target section image of the preset section type as a basis for calculating the probability of the disease category.

[0076] In combination with actual scenarios, the preset section type corresponding to the disease category may be the type of section that needs to be observed when detecting the disease category. For example, if the spinal disease category pays more attention to the sagittal plane of the spine and the cross-section of the spine, the preset section type corresponding to the disease category may include the sagittal plane of the spine and the cross-section of the spine. If the sagittal plane of the spine needs to be paid more attention to when determining the disease category, the section weight corresponding to the sagittal plane of the spine may be higher than the section weight of the cross-section of the spine. The preset section types corresponding to the above-mentioned disease categories and the weights corresponding to the preset section types can be set in advance by developers or users, and the embodiments of the present invention are not limited here.

[0077] Correspondingly, for each disease category corresponding to the target gestational age stage in step S133a1 or step S131c, the corresponding target cross-sectional image is input into a preset first algorithm to calculate the corresponding disease probability, which may include steps S131c1 to S131c2.

[0078] In step S131c1, for each preset section type corresponding to each disease category corresponding to the target gestational age stage, the target section image of the preset section type is input into a preset first algorithm to calculate the first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type.

[0079] For the relevant contents of the preset first algorithm, reference may be made to the relevant contents of step S133a1 or step S131c above, and the embodiments of the present invention will not be elaborated herein.

[0080] In step S131c2, based on the section weights of each preset section type corresponding to the disease category and each first probability of the disease category, the probability of the target fetus suffering from a disease of the disease category is determined.

[0081] For each preset section type of each disease category, the section weight of the preset section type corresponding to the disease category can be multiplied by the first probability of the target section image of the preset section type to obtain the target probability of the target fetus suffering from the disease of the disease category for the preset section type. In one example, the highest target probability of all preset section types corresponding to the disease category can be used as the probability of the target fetus suffering from the disease of the disease category. In another example, the target probabilities of all preset section types corresponding to the disease category can also be summed to obtain the probability of the target fetus suffering from the disease of the disease category.

[0082] According to the above scheme of the embodiment of the present invention, for each preset section type corresponding to each disease category corresponding to the target gestational age stage, the target section image of the preset section type can be input into the preset first algorithm to calculate the first probability that the target fetus suffers from the disease of the disease category in the target section image of the preset section type. Then, based on the section weight of each preset section type corresponding to the disease category and each first probability of the disease category, the probability of the target fetus suffering from the disease of the disease category is determined. In the above scheme, each disease category can correspond to a preset section type and a section weight corresponding to the preset section type, so based on the target fetal images of multiple preset section types, a more representative probability of disease can be determined to improve the accuracy of the output standard disease name.

[0083] Exemplarily, the preset first algorithm includes a first detection model.

[0084] The specific model architecture of the first detection model mentioned above is not limited in the embodiment of the present invention. For example, it can be a convolutional neural network architecture, a recurrent neural network architecture, a support vector machine architecture, etc.

[0085] Correspondingly, in step S131c1, the target section image of the preset section type is input into a preset first algorithm to calculate the first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type, including: inputting the target section image of the preset section type into a first detection model corresponding to the preset section type to obtain the first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type.

[0086] For multiple preset section types corresponding to disease categories, the first detection model corresponding to the preset section type can be obtained by training based on the training image corresponding to the disease category, and the first detection model corresponding to the preset section type in different disease categories may be different. The generation method of the above-mentioned training image may be the same as the generation method of the target section image, and the embodiments of the present invention are not described in detail here. For each disease category, the first detection model corresponding to the disease category may be different from the first detection models corresponding to other disease categories, so as to achieve targeted detection of the disease category. The multiple preset section types corresponding to the disease category may correspond to the same first detection model. Furthermore, the multiple preset sections corresponding to the disease category may also correspond to different first detection models to further improve the accuracy of the first probability. In this case, the first detection model corresponding to the preset section type is obtained by training based on the training image of the preset section type corresponding to the disease category.

[0087] For multiple preset section types corresponding to disease categories, the target section image of the preset section type can be input into the first detection model corresponding to the preset section type to obtain the confidence of the first detection model that the category of the target section image is the disease category. The confidence can be used as the probability that the target fetus suffers from the disease of the disease category.

[0088] According to the above scheme of the embodiment of the present invention, the target section image of the preset section type can be input into the first detection model corresponding to the preset section type to obtain the first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type. The above scheme can use the first detection model to generate the first probability to improve the accuracy and robustness of the first probability obtained, thereby improving the accuracy of the probability of illness obtained based on the first probability, which is beneficial to improving the accuracy of the output standard disease name. In addition, the above scheme can realize targeted detection of disease categories, which is also beneficial to improving the accuracy of the final output standard disease name.

[0089] Exemplarily, the first detection model is trained by the following steps: for each disease category in each preset gestational stage, a training image of a baseline fetus of a preset section type corresponding to the disease category at the preset gestational stage in the training data set is input into the first initial model for training, and the model parameters of the first initial model are adjusted based on the first label of the training image to obtain the first detection model.

[0090] The first label can be used to indicate whether the actual disease category of the training image is the disease category corresponding to the training image. For example, for each disease category, if the disease category corresponds to preset section type 1 and preset section type 2, multiple training images of preset section type 1 corresponding to the disease category and multiple training images of preset section type 2 corresponding to the disease category can be sequentially input into the first initial model for training. The first initial model can generate a predicted label based on the training image. The model parameters of the first initial model can be adjusted based on the difference between the first label and the predicted label of the training image to iteratively reduce the difference between the first label and the predicted label, so as to continuously improve the accuracy of the predicted label output by the first initial model, and then obtain the first detection model. It should be understood that the first initial model can also use the loss function required in the relevant technology to determine the loss value to adjust the model parameters, and the embodiment of the present invention is not limited here.

[0091] According to the above scheme of the embodiment of the present invention, for each disease category in each preset gestational age stage, a training image of a reference fetus of a preset section type corresponding to the disease category at the preset gestational age stage in the training data set can be input into the first initial model for training, and the model parameters of the first initial model can be adjusted based on the first label of the training image to obtain a first detection model. The first detection model trained by the above scheme can realize targeted detection of disease categories, improve the accuracy of the first probability obtained, and further improve the accuracy of the probability of illness obtained based on the first probability, which is conducive to improving the accuracy of the output standard disease name.

[0092] Exemplarily, the preset first algorithm includes a second detection model.

[0093] Correspondingly, in step S131c1, the target section image of the preset section type is input into a preset first algorithm to calculate the first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type, including: inputting the target section image of the preset section type into a second detection model corresponding to the preset section type to obtain the first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type.

[0094] For multiple preset section types corresponding to disease categories, the second detection model corresponding to the preset section type is obtained by training based on multiple training images corresponding to each disease category including at least the preset section type. In one example, if disease category A, disease category B, and disease category C all include the preset section type, the second detection model can be trained based on the training images corresponding to each of disease category A, disease category B, and disease category C in the preset section type. In another example, referring to the above example, the second detection model is trained based on the training images corresponding to each of disease category A, disease category B, and disease category C. Compared with the second detection model obtained in the previous example, the training method of the second detection model obtained in this example is simpler, and the total number of second detection models actually used is also less. Developers can train the second detection model in a corresponding manner according to actual conditions, and the embodiments of the present invention are not limited here.

[0095] According to the above scheme of the embodiment of the present invention, the target section image of the preset section type can be input into the second detection model corresponding to the preset section type to obtain the first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type. The second detection model corresponding to the above preset section type can be trained based on a plurality of training images corresponding to the disease categories of at least the preset section type, so its training method is simpler than the first detection model, and the total number of detection models required is also less, which can improve the generation efficiency of the first probability, and then improve the generation efficiency of the probability of illness, which is conducive to improving the efficiency of outputting standard disease names.

[0096] Exemplarily, the second detection model is trained by the following steps: For each disease category in each preset gestational stage, a training image of a baseline fetus of a preset section type corresponding to the disease category at the preset gestational stage in the training data set is input into the second initial model for training, and the model parameters of the second initial model are adjusted based on the second label of the training image to obtain a second detection model.

[0097] The second label is used to represent the actual disease category of the training image to adapt to the multi-classification scenario of the second initial model. It should be understood that for multiple preset section types corresponding to the disease category, the second detection model corresponding to the preset section type is trained based on multiple training images corresponding to each disease category including at least the preset section type. For example, in one example, if disease category A, disease category B, and disease category C all include preset section type 1, a second detection model can be trained based on the training images corresponding to disease category A, disease category B, and disease category C in preset section type 1. The output result of the second detection model may include the first probability that the target fetus suffers from diseases of disease category A, disease category B, and disease category C in the target section image of preset section type 1. If disease category A and disease category B both include preset section type 2, another second detection model can be trained based on the training images corresponding to disease category A and disease category B in preset section type 2. The output result of the second detection model may include the first probability that the target fetus suffers from diseases of disease category A and disease category B in the target section image of preset section type 2. The second initial model can generate a predicted label based on the training image. The model parameters of the second initial model can be adjusted based on the difference between the second label and the predicted label of the training image to iteratively reduce the difference between the second label and the predicted label, thereby continuously improving the accuracy of the predicted label output by the second initial model, and then obtaining a second detection model. It should be understood that the second initial model can also use the loss function required in the relevant technology to determine the loss value to adjust the model parameters, and the embodiment of the present invention is not limited here.

[0098] According to the above scheme of the embodiment of the present invention, for each disease category in each preset gestational age stage, the training image of the reference fetus of the preset section type corresponding to the disease category at the preset gestational age stage in the training data set can be input into the second initial model for training, and the model parameters of the second initial model can be adjusted based on the second label of the training image to obtain the second detection model. The total amount of the second detection model trained by the above scheme is less than the total amount of the first detection model, which can improve the generation efficiency of the first probability, and further improve the output efficiency of the standard disease name.

[0099] Exemplarily, in step S140, based on each disease probability, a final disease category is determined from multiple disease categories corresponding to the target gestational age stage, including: determining the disease category corresponding to the maximum disease probability among all disease probabilities as the final disease category.

[0100] For example, the multiple disease categories corresponding to the target gestational age stage include disease category A, disease category B, and disease category C. The probability of disease category A is 0.5, the probability of disease category B is 0.6, and the probability of disease category C is 0.7. Then, the disease category C corresponding to the largest 0.7 can be used as the final disease category.

[0101] According to the above solution of the embodiment of the present invention, the disease category corresponding to the maximum disease probability among all disease probabilities can be directly determined as the final disease category. The implementation of the above solution is relatively simple, and its computing power requirement is also low. It can be adapted to processors of different models, and can also improve the output efficiency of standard disease names.

[0102] Exemplarily, outputting the corresponding standard disease name in step S140 may include step S141 and step S142.

[0103] In step S141, the standard disease name corresponding to the final disease category is displayed in the first interface.

[0104] The first interface may be displayed on a display screen. The first interface may include a test report of the target fetus. Specifically, the test report may be a test report to be filled in the test phase, or may be a report that has been fully filled in the report maintenance phase.

[0105] In step S142, in response to the user selecting a standard disease name in the first interface, the standard disease name is filled into the area corresponding to the disease category displayed in the test report.

[0106] The user can select a standard disease name through the input device to fill the standard disease name into the above area. The area corresponding to the disease category in the test report can be used to fill in the relevant text of the disease category.

[0107] According to the above scheme of the embodiment of the present invention, the standard disease name corresponding to the final disease category can be displayed in the first interface. Then, in response to the user's operation of selecting the standard disease name in the first interface, the standard disease name is filled into the area corresponding to the disease category displayed in the test report. The above scheme can assist the user in updating the test report. If the actual diseases suffered by different fetuses are the same, the text related to the disease category filled in the test report of the above different fetuses is also the same, which is conducive to reducing the difficulty of implementing subsequent medical data statistics.

[0108] An embodiment of the present invention also provides a medical device. Figure 2 FIG. 2 shows a schematic block diagram of a medical device 200 according to an embodiment of the present invention. Figure 2As shown, the medical device 200 may include a signal generating device 210 and a processor 220 for executing the data processing method in any of the above embodiments.

[0109] The signal generating device 210 is used to transmit a preset signal and receive an echo signal generated by the preset signal acting on the target fetus. The signal generating device can be an ultrasound probe or a magnetic resonance generator. The preset signal can be an ultrasound signal or a magnetic resonance signal.

[0110] The processor 220 is further configured to generate a target section image of the target fetus based on the echo signal.

[0111] In an example, the processor may also be connected to the display screen and input device mentioned above to implement the data processing methods of the various embodiments mentioned above, and the embodiments of the present invention will not be described in detail here.

[0112] In addition, according to another aspect of the present invention, a storage medium is also provided, on which computer program instructions are stored, and when the computer program instructions are executed by a computer or a processor, the computer or the processor performs the corresponding steps related to the above-mentioned data processing method of the embodiment of the present invention, and is used to implement the corresponding modules in the above-mentioned medical device according to the embodiment of the present invention. The storage medium may include, for example, a memory card of a smart phone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disk read-only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media.

[0113] According to yet another aspect of the present invention, a computer program product is provided, comprising computer program instructions, which, when executed by a computer or a processor, enable the computer or the processor to execute corresponding steps related to the above data processing method.

[0114] A person of ordinary skill in the art can understand the specific implementation schemes of the above-mentioned medical device, storage medium and computer program product by reading the above-mentioned description of the data processing method, which will not be described here for the sake of brevity.

[0115] Although example embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above example embodiments are merely exemplary and are not intended to limit the scope of the present invention thereto. Various changes and modifications may be made therein by one of ordinary skill in the art without departing from the scope and spirit of the present invention. All such changes and modifications are intended to be included within the scope of the present invention as required by the appended claims.

[0116] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0117] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0118] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0119] Similarly, it should be understood that in order to streamline the present invention and help understand one or more of the various inventive aspects, in the description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, the method of the present invention should not be interpreted as reflecting the following intention: the claimed invention requires more features than the features explicitly stated in each claim. More specifically, as reflected in the corresponding claims, the inventive point is that the corresponding technical problem can be solved with less than all the features of a single disclosed embodiment. Therefore, the claims following the specific embodiment are hereby expressly incorporated into the specific embodiment, wherein each claim itself serves as a separate embodiment of the present invention.

[0120] Those skilled in the art will understand that, except for mutually exclusive features, all features disclosed in this specification (including the accompanying claims, abstract and drawings) and all processes or units of any method or device disclosed in this specification may be combined in any combination. Unless otherwise explicitly stated, each feature disclosed in this specification (including the accompanying claims, abstract and drawings) may be replaced by an alternative feature that provides the same, equivalent or similar purpose.

[0121] In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims, any one of the claimed embodiments may be used in any combination.

[0122] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all of the functions of some modules in a medical device according to an embodiment of the present invention. The present invention may also be implemented as a device program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0123] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0124] The above is only a specific embodiment of the present invention or an explanation of a specific embodiment. The protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A data processing method, characterized in that: The method comprises: Obtaining a target gestational age of the target fetus, and based on the target gestational age, determining multiple disease categories corresponding to the target gestational age in a disease database, wherein the disease database includes multiple preset gestational age and multiple disease categories corresponding to each preset gestational age, and each disease category corresponds to a standard disease name; Acquiring disease representation information of the target fetus, wherein the disease representation information includes one or more of a target section image of the target fetus and a disease description text; For each disease category corresponding to the target gestational age stage, based on the disease representation information of the target fetus, calculating the probability that the target fetus suffers from a disease of the disease category; Based on each of the disease probabilities, a final disease category is determined from a plurality of disease categories corresponding to the target gestational age stage, and the corresponding standard disease name is output.

2. The method according to claim 1, characterized in that The disease representation information includes a text describing the disease of the target fetus; Correspondingly, for each disease category corresponding to the target gestational age stage, based on the disease representation information of the target fetus, calculating the probability that the target fetus suffers from a disease of the disease category includes: Acquire first text information from the disease description text; Compare the first text information with the first preset information corresponding to each disease category corresponding to the target gestational age stage, so as to obtain a first similarity between the first text information and each first preset information; Each of the disease probabilities is determined based on each of the first similarities.

3. The method according to claim 2, characterized in that The disease representation information of the target fetus also includes a target section image of the target fetus; Correspondingly, the determining of each disease probability based on each first similarity includes, for each disease category corresponding to the target gestational age stage, if its corresponding first similarity is less than a preset similarity threshold, inputting its corresponding target section image into a preset first algorithm to calculate its corresponding disease probability; otherwise, determining its corresponding disease probability based on its corresponding first similarity.

4. The method according to claim 2, characterized in that After obtaining the first similarity between the first text information and each first preset information, the method further includes: If each of the first similarities is less than a preset similarity threshold, obtaining second text information from the disease description text; For each second preset information of each disease category corresponding to the target gestational age stage, the second text information is compared with the second preset information to obtain a second similarity between the second text information and the second preset information; The step of determining each of the disease probabilities based on each of the first similarities is replaced by determining each of the disease probabilities based on each of the second similarities.

5. The method according to claim 1, characterized in that The disease representation information of the target fetus includes a target section image of the target fetus; Correspondingly, for each disease category corresponding to the target gestational age stage, the probability of the target fetus suffering from a disease of that disease category is calculated based on the disease characterization information of the target fetus, including: for each disease category corresponding to the target gestational age stage, inputting the corresponding target section image into a preset first algorithm to calculate the corresponding disease probability.

6. The method according to claim 3 or 5, characterized in that The target section image includes a plurality of target section images belonging to a plurality of preset section types, and the disease database further includes: for each disease category, at least one preset section type corresponding to the disease category and a section weight of each preset section type corresponding to the disease category, wherein each section weight is used to indicate the importance of the target section image of the preset section type as a basis for calculating the probability of illness of the disease category; Correspondingly, for each disease category corresponding to the target gestational age stage, the corresponding target section image is input into a preset first algorithm to calculate the corresponding disease probability, including: For each preset section type corresponding to each disease category corresponding to the target gestational age stage, inputting the target section image of the preset section type into a preset first algorithm, and calculating a first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type; Based on the respective section weights of each preset section type corresponding to the disease category and each of the first probabilities of the disease category, the probability of the target fetus suffering from the disease of the disease category is determined.

7. The method according to claim 6, characterized in that The preset first algorithm includes a first detection model; Correspondingly, the target section image of the preset section type is input into a preset first algorithm to calculate the first probability that the target fetus suffers from the disease of the disease category in the target section image of the preset section type, including: inputting the target section image of the preset section type into a first detection model corresponding to the preset section type to obtain the first probability that the target fetus suffers from the disease of the disease category in the target section image of the preset section type, wherein the first detection model corresponding to the preset section type is trained based on the training image corresponding to the disease category, and the first detection model corresponding to the preset section type is different in different disease categories.

8. The method according to claim 7, characterized in that The first detection model is trained by the following steps: For each disease category in each preset gestational age stage, a training image of a baseline fetus of a preset section type corresponding to the disease category at the preset gestational age stage in a training data set is input into a first initial model for training, and the model parameters of the first initial model are adjusted based on a first label of the training image to obtain a first detection model, wherein the first label is used to indicate whether the actual disease category of the training image is the disease category corresponding to the training image.

9. The method according to claim 6, characterized in that The preset first algorithm includes a second detection model; Correspondingly, the step of inputting the target section image of the preset section type into a preset first algorithm to calculate a first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type includes: The target section image of the preset section type is input into a second detection model corresponding to the preset section type to obtain a first probability that the target fetus suffers from a disease of the disease category in the target section image of the preset section type, wherein the second detection model corresponding to the preset section type is trained based on multiple training images corresponding to each disease category including at least the preset section type.

10. The method according to claim 9, characterized in that The second detection model is trained by the following steps: For each disease category in each preset gestational age stage, a training image of a baseline fetus of a preset section type corresponding to the disease category at the preset gestational age stage in the training data set is input into a second initial model for training, and the model parameters of the second initial model are adjusted based on a second label of the training image to obtain a second detection model, wherein the second label is used to represent the actual disease category of the training image.

11. The method according to claim 3 or 5, characterized in that: The method further includes: calculating the target gestational age stage based on the target section image according to a preset second algorithm.

12. The method according to claim 1, characterized in that Determining a final disease category from a plurality of disease categories corresponding to the target gestational age stage based on each of the disease probabilities includes: The disease category corresponding to the largest disease probability among all the disease probabilities is determined as the final disease category.

13. The method according to claim 1, characterized in that The output corresponds to the standard disease name, including: Displaying the standard disease name corresponding to the final disease category in the first interface, wherein the first interface also includes the test report of the target fetus; In response to the user selecting the standard disease name in the first interface, the standard disease name is filled into the area corresponding to the disease category displayed in the test report.

14. A medical device, characterized in that: The medical device comprises a signal generating device and a processor for executing the data processing method according to any one of claims 1 to 13; The signal generating device is used to transmit a preset signal and receive an echo signal generated by the preset signal acting on the target fetus; The processor is further configured to generate a target section image of the target fetus based on the echo signal.

15. A storage medium storing computer program instructions, characterized in that: The computer program instructions are used to execute the data processing method according to any one of claims 1 to 13 when executed.

16. A computer program product comprising computer program instructions, characterized in that The computer program instructions are used to perform the data processing method according to any one of claims 1 to 13 when executed by a processor.

Images