A method and system for processing medical data

Abstract

The application discloses a medical data processing method and system, comprising feature recognition on the image collected by an endoscope, to obtain feature information included in the image; wherein the feature information includes multiple target features and feature position coordinates corresponding to each target feature, and one target feature corresponds to one feature position coordinate; determining the feature type of each target feature in the feature information; wherein the feature type at least includes organ type and lesion type; determining the standard image at the current position of the endoscope according to the organ type and the feature position coordinates corresponding to the target feature of the organ type; comparing the feature information with the standard image to obtain error feature information; wherein the error feature information includes at least one error feature and error feature position coordinates corresponding to the error feature; deleting the error feature information from the feature information to obtain correct feature information. The application can improve the accuracy of the examination result based on the endoscope.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a medical data processing method and system. BACKGROUND

[0002] At present, using an endoscope to examine a patient is a relatively common examination method. Since the environment in the patient's body is relatively complex, and the images collected by the endoscope are usually also relatively similar, the doctor using the endoscope is likely to ignore the lesions in the patient's body or identify normal organs as lesions, thereby causing the examination result based on the endoscope to be inaccurate.

[0003] At present, no effective solution has been proposed for the above problems. SUMMARY

[0004] The embodiments of the present application provide a medical data processing method and system, which can correctly identify the target collected by the endoscope, so as to improve the accuracy of the examination result based on the endoscope.

[0005] According to an aspect of an embodiment of the present application, a medical data processing method is provided, comprising:

[0006] performing feature recognition on an image collected by an endoscope to obtain feature information included in the image; wherein the feature information includes a plurality of target features and a feature position coordinate corresponding to each target feature, and one target feature corresponds to one feature position coordinate;

[0007] determining a feature type of each target feature in the feature information; wherein the feature type at least includes an organ type and a lesion type;

[0008] determining a standard image at a current position of the endoscope according to the organ type and the feature position coordinate corresponding to the target feature of the organ type;

[0009] comparing the feature information with the standard image to obtain error feature information; wherein the error feature information includes at least one error feature and an error feature position coordinate corresponding to the error feature;

[0010] deleting the error feature information from the feature information to obtain correct feature information.

[0011] As an optional implementation, the determination of the standard image at the current position of the endoscope according to the organ type and the feature position coordinate corresponding to the target feature of the organ type comprises:

[0012] determining an organ target feature of a feature type being the organ type from the plurality of target features;

[0013] obtaining a confidence degree of the organ target feature;

[0014] obtaining a standard organ target feature with the confidence degree being greater than a confidence degree threshold from the organ target features;

[0015] determining position relationship information between any two of the standard organ target features;

[0016] determining a standard image at a current position of the endoscope according to the feature position coordinates corresponding to the standard organ target features and the position relationship information.

[0017] As an optional implementation, the comparing the feature information with the standard image to obtain error feature information comprises:

[0018] obtaining standard organ information from the standard image; wherein the standard organ information at least includes a standard organ type and a coordinate region corresponding to the standard organ type;

[0019] if an organ type of a certain organ target feature is different from the standard organ type, determining the certain organ target feature as a first error organ target feature;

[0020] if an organ type of a certain organ target feature is the same as the standard organ type, determining position relationship information between a target feature position coordinate corresponding to the certain organ target feature and a target coordinate region corresponding to the organ type of the certain organ target feature;

[0021] if the position relationship information indicates that the target feature position coordinate is located outside the target coordinate region, determining the certain organ target feature as a second error organ target feature;

[0022] determining error feature information according to the first error organ target feature and the second error organ target feature.

[0023] As an optional implementation, after the determining error feature information according to the first error organ target feature and the second error organ target feature, the method further comprises:

[0024] determining a current lesion organ where the lesion target feature is located according to a feature position coordinate corresponding to the lesion target feature of the lesion type;

[0025] obtaining a standard lesion organ matching the lesion type of the lesion target feature;

[0026] determining the lesion target feature as a false lesion target feature if the current lesion organ does not match the standard lesion organ;

[0027] adding the false lesion target feature and the feature position coordinates corresponding to the false lesion target feature into the false feature information.

[0028] As an optional implementation, after the correct feature information is obtained, the method further comprises:

[0029] determining a position region where each correct target feature in the correct feature information is located in the to-be-processed image; wherein one correct target feature corresponds to one position region;

[0030] determining an enclosing box corresponding to each correct target feature according to the position region corresponding to each correct target feature; wherein one correct target feature corresponds to one enclosing box;

[0031] outputting each enclosing box and the feature information of the correct target feature corresponding to each enclosing box in the to-be-processed image.

[0032] According to another aspect of the embodiment of the present application, a medical data processing system is further provided, comprising:

[0033] a recognition unit configured to perform feature recognition on a to-be-processed image collected by an endoscope, to obtain feature information included in the to-be-processed image; wherein the feature information includes a plurality of target features and feature position coordinates corresponding to each target feature, and one target feature corresponds to one feature position coordinate;

[0034] a first determination unit configured to determine a feature type of each target feature in the feature information; wherein the feature type at least includes an organ type and a lesion type;

[0035] a second determination unit configured to determine a standard image at a current position of the endoscope according to the organ type and the feature position coordinates corresponding to the target feature of the organ type;

[0036] a comparison unit configured to compare the feature information with the standard image, to obtain false feature information; wherein the false feature information includes at least one false feature and false feature position coordinates corresponding to the false feature;

[0037] a deletion unit configured to delete the false feature information from the feature information, to obtain correct feature information.

[0038] As an optional implementation, the second determining unit determines the standard image at the current position of the endoscope according to the organ type and the feature position coordinates corresponding to the target feature of the organ type, and the manner is specifically as follows:

[0039] determining an organ target feature of the organ type from the target features;

[0040] obtaining a confidence degree of the organ target feature;

[0041] obtaining a standard organ target feature with a confidence degree greater than a confidence degree threshold from the organ target features;

[0042] determining position relationship information between any two standard organ target features;

[0043] determining the standard image at the current position of the endoscope according to the feature position coordinates corresponding to the standard organ target features and the position relationship information.

[0044] As an optional implementation, the comparison unit compares the feature information with the standard image to obtain the error feature information, and the manner is specifically as follows:

[0045] obtaining standard organ information from the standard image; wherein the standard organ information at least includes a standard organ type and a coordinate region corresponding to the standard organ type;

[0046] if the organ type of a certain organ target feature is different from the standard organ type, determining the certain organ target feature as a first error organ target feature;

[0047] if the organ type of a certain organ target feature is the same as the standard organ type, determining position relationship information between a target feature position coordinate corresponding to the certain organ target feature and a target coordinate region corresponding to the organ type of the certain organ target feature;

[0048] if the position relationship information indicates that the target feature position coordinate is located outside the target coordinate region, determining the certain organ target feature as a second error organ target feature;

[0049] determining the error feature information according to the first error organ target feature and the second error organ target feature.

[0050] As an optional implementation, the comparison unit is further configured to:

[0051] determine a current lesion organ where the lesion target feature is located according to the feature position coordinates corresponding to the lesion target feature of the lesion type;

[0052] obtain a standard lesion organ matching the lesion type of the lesion target feature;

[0053] if the current lesion organ does not match the standard lesion organ, determine the lesion target feature as a false lesion target feature;

[0054] add the false lesion target feature and the feature position coordinates corresponding to the false lesion target feature to the false feature information.

[0055] As an optional implementation, the deleting unit is further configured to:

[0056] after obtaining the correct feature information, determine a location area where each correct target feature in the correct feature information is located in the image to be processed; wherein one correct target feature corresponds to one location area;

[0057] determine an enclosing box corresponding to each correct target feature according to the location area corresponding to each correct target feature; wherein one correct target feature corresponds to one enclosing box;

[0058] output each enclosing box and the feature information of the correct target feature corresponding to each enclosing box in the image to be processed.

[0059] According to still another aspect of the embodiments of the present application, a computing device is provided, which comprises at least one processor, a memory and an input-output unit; wherein the memory is configured to store a computer program, and the processor is configured to invoke the computer program stored in the memory to execute the medical data processing method.

[0060] According to still another aspect of the embodiments of the present application, a computer readable storage medium is provided, which comprises instructions, when executed on a computer, cause the computer to execute the medical data processing method.

[0061] In the embodiment of the present application, the feature recognition can be performed on the image to be processed collected by the endoscope, the feature information including the target feature and the corresponding feature position coordinates of the target feature can be obtained, the standard image at the current position of the endoscope can be obtained according to the organ type of the target feature and the corresponding feature position coordinates of the target feature, the error feature information can be determined from the feature information by comparing the standard image with the obtained feature information, and the error feature information can be deleted from the feature information to obtain the correct feature information. It can be seen that the embodiment of the present application can perform the correctness recognition on the target collected by the endoscope, so as to improve the accuracy of the examination result obtained based on the endoscope. BRIEF DESCRIPTION OF DRAWINGS

[0062] The accompanying drawings, which are included to provide a further understanding of the present application and are incorporated in and constitute a part of this application, illustrate embodiments of the present application and serve to explain the principles of the present application, and do not limit the present application in any manner. In the drawings:

[0063] Figure 1 is a flowchart of an optional medical data processing method according to an embodiment of the present application;

[0064] Figure 2 is a flowchart of a determination method of a standard image at a current position of an endoscope according to an embodiment of the present application;

[0065] Figure 3 is a flowchart of a determination method of error feature information according to an embodiment of the present application;

[0066] Figure 4 is a flowchart of an output method of a bounding box and correct target feature information according to an embodiment of the present application;

[0067] Figure 5 is a structural diagram of an optional medical data processing system according to an embodiment of the present application;

[0068] Figure 6 is a structural diagram of a medium according to an embodiment of the present application;

[0069] Figure 7 is a structural diagram of a computing device according to an embodiment of the present application. DETAILED DESCRIPTION

[0070] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work should belong to the protection scope of the present application.

[0071] It should be noted that the terms "first", "second" and the like in the description and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in other than the order illustrated or described herein. In addition, the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0072] The following will be described with reference to the drawings Figure 1 , Figure 1 The flowchart of the medical data processing method provided by an embodiment of the present application is shown. It should be noted that the embodiments of the present application can be applied to any applicable scenario.

[0073] Figure 1 The flowchart of the medical data processing method provided by an embodiment of the present application is shown. It should be noted that the embodiments of the present application can be applied to any applicable scenario.

[0074] In step S101, feature recognition is performed on the image collected by the endoscope to obtain feature information included in the image.

[0075] In the embodiment of the present application, the feature information includes a plurality of target features and a feature position coordinate corresponding to each target feature, and one target feature corresponds to one feature position coordinate.

[0076] In the embodiment of the present application, the feature recognition model can be a deep learning model obtained by pre-training. The feature recognition model can be used to recognize the internal organs, internal tissues and lesions in the image, and obtain a plurality of target features included in the image.

[0077] In the embodiment of the present application, the image to be processed can be an image of a patient's body collected by an endoscope. The feature position coordinates can be center point coordinates of target features, and a rectangular coordinate system can be established on the image to be processed. The origin of the rectangular coordinate system can be any point in the image to be processed.

[0078] In step S102, the feature type of each target feature in the feature information is determined.

[0079] In the embodiment of the present application, the feature type includes at least an organ type and a lesion type.

[0080] In step S103, a standard image at the current position of the endoscope is determined according to the organ type and the feature position coordinates of the target features corresponding to the organ type.

[0081] In the embodiment of the present application, the area that can be reached by the endoscope can be divided, and the organ or tissue features that can be simultaneously collected by the endoscope in each area can be determined in advance.

[0082] For example, if an ultrasonic endoscope is used to examine a standard mediastinum, it can be determined that the ultrasonic endoscope can reach five areas: a right heart area, a subcarinal space area, a descending aorta and umbilical vein area, a pulmonary aortic window area, and a neck blood vessel area. The organ or tissue features that can be simultaneously collected in the above five areas can also be set, for example, the ultrasonic endoscope can collect a standard image in which the inferior vena cava feature, the superior vena cava feature, and the right atrium feature exist at the same time in the right heart area. Furthermore, the center point of the inferior vena cava feature, the center point of the superior vena cava feature, and the center point of the right atrium feature are determined to have coordinates in the standard image, and the rectangular coordinate system in the standard image has the same setting mode as the rectangular coordinate system in the image to be processed. The first positional relationship between the center point of the inferior vena cava feature and the center point of the superior vena cava feature, the second positional relationship between the center point of the inferior vena cava feature and the center point of the right atrium feature, and the third positional relationship between the center point of the superior vena cava feature and the center point of the right atrium feature can also be determined and embodied in the standard image.

[0083] In the embodiment of the present application, a target area including target features of all organ types can be determined, and the positional relationship between any two organ features included in the target area can also be obtained. The positional relationship between any two organ features is compared with the positional relationship between the two target features corresponding to the two organ features. If the positional relationship matches, it can be considered that the image to be processed collected by the endoscope is the image of the target area, that is, the current position of the endoscope is the target area, and the standard image corresponding to the target area can be determined as the standard image at the current position of the endoscope.

[0084] In another embodiment of the present application, in order to accurately determine the standard image at the current position and improve the accuracy of the standard image, a standard organ target feature can be determined from the recognized organ target features according to the confidence, and the positional relationship between any two standard organ target features can be determined; then the current position of the endoscope in the patient's body can be determined based on the feature position coordinates corresponding to the standard organ target features and the positional relationship between any two standard organ target features, as shown in the following formula: Figure 2 As shown in the figure, the step S103 is replaced by the following steps S201 to S205:

[0085] Step S201: organ target features of the organ type are determined from the plurality of target features.

[0086] Step S202: the confidence of the organ target features is obtained.

[0087] Step S203: standard organ target features with a confidence greater than a confidence threshold are obtained from the organ target features.

[0088] Step S204: the positional relationship information between any two standard organ target features is determined.

[0089] Step S205: the standard image at the current position of the endoscope is determined according to the feature position coordinates corresponding to the standard organ target features and the positional relationship information.

[0090] The steps S201 to S205 are implemented, the standard organ target features can be determined from the recognized organ target features according to the confidence, and the positional relationship between any two standard organ target features can be determined; then the current position of the endoscope in the patient's body can be determined based on the feature position coordinates corresponding to the standard organ target features and the positional relationship between any two standard organ target features, so that the standard image at the current position is accurately determined, and the accuracy of the standard image is improved.

[0091] In the embodiment of the present application, the confidence of the organ target feature can be the credibility of the organ target feature output by the feature recognition model.

[0092] Step S104: the feature information is compared with the standard image to obtain error feature information.

[0093] In the embodiment of the present application, the error feature information includes at least one error feature and the error feature position coordinates corresponding to the error feature.

[0094] In another embodiment of the present application, in order to improve the accuracy of the error feature information, the standard organ type and the coordinate region corresponding to the standard organ type can be determined from the standard image, and the organ target feature with an organ type different from the standard organ type can be determined as an error organ target feature; and the organ target feature with a feature position coordinate not in the coordinate region corresponding to its organ type can also be determined as an error organ target feature, as shown in the following table: Figure 3 As shown in the table, the step S104 is replaced by the following steps S301-S305:

[0095] In the step S301, the standard organ information is obtained from the standard image.

[0096] In the embodiment of the present application, the standard organ information at least includes the standard organ type and the coordinate region corresponding to the standard organ type.

[0097] In the step S302, if the organ type of a certain organ target feature is different from the standard organ type, the certain organ target feature is determined as a first error organ target feature.

[0098] In the step S303, if the organ type of a certain organ target feature is the same as the standard organ type, the position relationship information between the target feature position coordinate of the certain organ target feature and the target coordinate region corresponding to the organ type of the certain organ target feature is determined.

[0099] In the step S304, if the position relationship information indicates that the target feature position coordinate is located outside the target coordinate region, the certain organ target feature is determined as a second error organ target feature.

[0100] In the step S305, the error feature information is determined according to the first error organ target feature and the second error organ target feature.

[0101] The steps S301-S305 are implemented, the standard organ type and the coordinate region corresponding to the standard organ type can be determined from the standard image, and the organ target feature with an organ type different from the standard organ type can be determined as an error organ target feature; and the organ target feature with a feature position coordinate not in the coordinate region corresponding to its organ type can also be determined as an error organ target feature, thereby improving the accuracy of the error feature information.

[0102] As an optional implementation, after the step S305 of determining the error feature information according to the first error organ target feature and the second error organ target feature, the following steps can also be performed:

[0103] determine a current lesion organ where the lesion target feature is located according to a feature position coordinate corresponding to a lesion target feature of the lesion type;

[0104] obtain a standard lesion organ matching the lesion type of the lesion target feature;

[0105] if the current lesion organ does not match the standard lesion organ, determine the lesion target feature as an error lesion target feature;

[0106] add the error lesion target feature and the feature position coordinate corresponding to the error lesion target feature to the error feature information.

[0107] According to the position where the lesion target feature is located, the current lesion organ where the lesion target feature appears can be determined, the labeled lesion organ where the lesion target feature usually appears can be obtained, whether the current lesion organ can appear the lesion target feature can be determined, and the lesion target feature that cannot appear on the current lesion organ can be determined as an error lesion target feature, so that the comprehensiveness of the error feature information is improved.

[0108] Step S105, delete the error feature information from the feature information to obtain correct feature information.

[0109] In another embodiment of the present application, in order to more intuitively identify the correct target feature in the to-be-processed image, the bounding box and the feature information of the correct target feature corresponding to each bounding box can be output in the to-be-processed image, as shown in FIG. 5. Figure 4 After the above step S105, the following steps S401-S403 can be further included:

[0110] Step S401, determine the position region of each correct target feature in the to-be-processed image in the correct feature information.

[0111] In the embodiment of the present application, one correct target feature corresponds to one position region.

[0112] Step S402, determine the bounding box corresponding to each correct target feature according to the position region corresponding to each correct target feature.

[0113] In the embodiment of the present application, one correct target feature corresponds to one bounding box.

[0114] Step S403, output each bounding box and the feature information of the correct target feature corresponding to each bounding box in the to-be-processed image.

[0115] By implementing the steps S401-S403, the bounding box and the feature information of the correct target feature corresponding to each bounding box can be output in the to-be-processed image, so that the correct target feature can be more intuitively identified in the to-be-processed image.

[0116] In the embodiment of the present application, the position region of the correct target feature in the to-be-processed image can be entirely in the bounding box corresponding to the correct target feature, and the shape of the bounding box can be circular, rectangular, polygonal, etc., which is not limited in the embodiment of the present application.

[0117] The present application can correctly identify the target collected by the endoscope, so as to improve the accuracy of the examination result based on the endoscope. In addition, the present application can also improve the accuracy of the standard image. In addition, the present application can also improve the accuracy of the error feature information. In addition, the present application can also improve the comprehensiveness of the error feature information. In addition, the present application can also more intuitively identify the correct target feature in the to-be-processed image.

[0118] After introducing the method of the exemplary embodiment of the present application, next, with reference to Figure 5 A medical data processing system of an exemplary embodiment of the present application is described, which comprises:

[0119] The identification unit 501 is configured to perform feature identification on the to-be-processed image collected by the endoscope, to obtain feature information included in the to-be-processed image. The feature information includes a plurality of target features and feature position coordinates corresponding to each target feature, and one target feature corresponds to one feature position coordinate.

[0120] The first determination unit 502 is configured to determine the feature type of each target feature in the feature information obtained by the identification unit 501. The feature type at least includes an organ type and a lesion type.

[0121] The second determination unit 503 is configured to determine a standard image at a current position of the endoscope according to the organ type determined by the first determination unit 502 and the feature position coordinates of the target feature corresponding to the organ type.

[0122] The comparison unit 504 is configured to compare the feature information obtained by the identification unit 501 with the standard image obtained by the second determination unit 503, to obtain error feature information. The error feature information includes at least one error feature and error feature position coordinates corresponding to the error feature.

[0123] The deletion unit 505 is configured to delete the error feature information obtained by the comparison unit 504 from the feature information obtained by the identification unit 501, to obtain correct feature information.

[0124] As an optional implementation, the second determining unit 503 determines the standard image at the current position of the endoscope in the following manner according to the organ type and the feature position coordinates corresponding to the target feature of the organ type:

[0125] determining an organ target feature of the feature type from the plurality of target features;

[0126] obtaining a confidence of the organ target feature;

[0127] obtaining a standard organ target feature with a confidence greater than a confidence threshold from the organ target feature;

[0128] determining position relationship information between any two of the standard organ target features;

[0129] determining the standard image at the current position of the endoscope according to the feature position coordinates corresponding to the standard organ target feature and the position relationship information.

[0130] In this implementation, the standard organ target feature can be determined from the recognized organ target features according to the confidence, and the position relationship between any two standard organ target features can be determined. Then, the current position of the endoscope in the patient can be determined based on the feature position coordinates corresponding to the standard organ target feature and the position relationship between any two standard organ target features, so that the standard image at the current position can be accurately determined, and the accuracy of the standard image is improved.

[0131] As an optional implementation, the comparison unit 504 compares the feature information with the standard image in the following manner to obtain the error feature information:

[0132] obtaining standard organ information from the standard image; wherein the standard organ information at least includes a standard organ type and a coordinate region corresponding to the standard organ type;

[0133] if the organ type of a certain organ target feature is different from the standard organ type, determining the certain organ target feature as a first error organ target feature;

[0134] if the organ type of a certain organ target feature is the same as the standard organ type, determining the position relationship information between the target feature position coordinates corresponding to the certain organ target feature and the target coordinate region corresponding to the organ type of the certain organ target feature;

[0135] If the position relation information indicates that the target feature position coordinate is located outside the target coordinate region, the certain organ target feature is determined as a second error organ target feature;

[0136] The error feature information is determined according to the first error organ target feature and the second error organ target feature.

[0137] In this embodiment, the standard organ type and the coordinate region corresponding to the standard organ type can be determined from the standard image, and the organ target feature with an organ type different from the standard organ type can be determined as an error organ target feature; and the organ target feature with a feature position coordinate not in the coordinate region corresponding to the organ type of the organ target feature can also be determined as an error organ target feature, thereby improving the accuracy of the error feature information.

[0138] As an optional embodiment, the comparison unit 504 is further configured to:

[0139] After the error feature information is determined according to the first error organ target feature and the second error organ target feature, a current lesion organ in which the lesion target feature is located is determined according to the feature position coordinate corresponding to the lesion target feature of the lesion type.

[0140] A standard lesion organ matching the lesion type of the lesion target feature is obtained.

[0141] If the current lesion organ does not match the standard lesion organ, the lesion target feature is determined as an error lesion target feature.

[0142] The error lesion target feature and the feature position coordinate corresponding to the error lesion target feature are added to the error feature information.

[0143] In this embodiment, the current lesion organ in which the lesion target feature is located can be determined according to the position of the lesion target feature, and whether the current lesion organ can appear the lesion target feature can be determined by obtaining the labeled lesion organ in which the lesion target feature usually appears; and the lesion target feature that cannot appear in the current lesion organ is determined as an error lesion target feature, thereby improving the comprehensiveness of the error feature information.

[0144] As an optional embodiment, the deletion unit 505 is further configured to:

[0145] After the correct feature information is obtained, a position region in which each correct target feature in the correct feature information is located in the image to be processed is determined; one correct target feature corresponds to one position region.

[0146] Determine the bounding box corresponding to each correct target feature according to the position region corresponding to each correct target feature respectively; wherein one correct target feature corresponds to one bounding box;

[0147] Output each bounding box and the feature information of the correct target feature corresponding to each bounding box in the image to be processed.

[0148] The implementation of the embodiment can output the bounding box and the feature information of the correct target feature corresponding to each bounding box in the image to be processed, so that the correct target feature can be more intuitively identified in the image to be processed.

[0149] After introducing the method and system of the exemplary embodiments of the present application, next, with reference to Figure 6 The computer readable storage medium of the exemplary embodiments of the present application is described, please refer to Figure 6 The computer readable storage medium shown is an optical disc 60, and a computer program (i.e. program product) is stored on the optical disc 60. When the computer program is run by a processor, each step described in the above method embodiment is implemented, for example, feature recognition is performed on the image to be processed collected by an endoscope to obtain feature information included in the image to be processed; wherein the feature information includes a plurality of target features and a feature position coordinate corresponding to each target feature, and one target feature corresponds to one feature position coordinate; the feature type of each target feature in the feature information is determined; wherein the feature type at least includes an organ type and a lesion type; a standard image at a current position of the endoscope is determined according to the organ type and the feature position coordinate corresponding to the target feature of the organ type; the feature information is compared with the standard image to obtain error feature information; wherein the error feature information includes at least one error feature and an error feature position coordinate corresponding to the error feature; the error feature information is deleted from the feature information to obtain correct feature information; the specific implementation of each step is not repeated here.

[0150] It should be noted that examples of the computer readable storage medium can also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical, magnetic storage medium, which will not be repeated here.

[0151] After introducing the method, system and medium of the exemplary embodiments of the present application, next, with reference to Figure 7 The computer device for processing medical data of the exemplary embodiments of the present application.

[0152] Figure 7 A block diagram is shown of an exemplary computing device 70 suitable for implementing embodiments of the present invention. The computing device 70 may be a computer system or a server. Figure 7 The computing device 70 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0153] like Figure 7 As shown, the components of the computing device 70 may include, but are not limited to: one or more processors or processing units 701, system memory 702, and bus 703 connecting different system components (including system memory 702 and processing unit 701).

[0154] The computing device 70 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computing device 70, including volatile and non-volatile media, removable and non-removable media.

[0155] System memory 702 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 7021 and / or cache memory 7022. Computing device 70 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, ROM 7023 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 7 Not shown in the image (usually referred to as a "hard drive"). Although not shown in Figure 7 The diagram illustrates that disk drives for reading and writing to removable non-volatile disks (e.g., "floppy disks") and optical disc drives for reading and writing to removable non-volatile optical discs (e.g., CD-ROMs, DVD-ROMs, or other optical media) can be provided. In these cases, each drive can be connected to bus 703 via one or more data media interfaces. System memory 702 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0156] A program / utility 7025 having a set (at least one) of program modules 7024 may be stored, for example, in system memory 702, and such program modules 7024 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment. Program modules 7024 typically perform the functions and / or methods described in the embodiments of the present invention.

[0157] The computing device 70 can also communicate with one or more external devices 704 such as a keyboard, a pointing device, a display, etc. via an input / output (I / O) interface 705. Further, the computing device 70 can communicate with one or more networks such as a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet via a network adapter 706. As Figure 7 illustrated, the network adapter 706 is in communication with the other modules of the computing device 70 such as the processing unit 701 via the bus 703. It should be appreciated that the computing device 70 can also be connected to other devices not shown in FIG. 7, via the I / O interface 705 and / or the network adapter 706. ​ It should be appreciated that other hardware and / or software modules can be used in connection with the computing device 70, as is not shown in FIG. 7.

[0158] The processing unit 701 performs various function applications and data processing by running programs stored in the system memory 702, such as feature recognition on a to-be-processed image collected by an endoscope to obtain feature information included in the to-be-processed image, wherein the feature information includes a plurality of target features and a feature position coordinate corresponding to each target feature, one target feature corresponds to one feature position coordinate, determines a feature type of each target feature in the feature information, wherein the feature type at least includes an organ type and a lesion type, determines a standard image at a current position of the endoscope according to the organ type and the feature position coordinate corresponding to the target feature of the organ type, compares the feature information with the standard image to obtain error feature information, wherein the error feature information includes at least one error feature and an error feature position coordinate corresponding to the error feature, and deletes the error feature information from the feature information to obtain correct feature information. The specific implementation of each step will not be repeated here. It should be noted that although several units / modules or sub-units / sub-modules of the medical data processing system are mentioned in the foregoing detailed description, such division is only exemplary and not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into a plurality of units / modules.

[0159] In the description of the present application, it should be noted that the terms "first", "second", "third" are only for the purpose of description and cannot be understood as indicating or implying relative importance.

[0160] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0161] In several embodiments provided by the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. The described device embodiments are merely schematic, and for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation; for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, or the among different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electric, mechanical or other forms.

[0162] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. In actual implementation, some or all of the units can be selected according to the actual needs to achieve the purposes of the embodiments of the present application.

[0163] In addition, each function unit in the various embodiments of the present application can be integrated in one processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit.

[0164] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0165] Finally, it should be noted that the above-described embodiments are merely intended for describing the technical solutions of the present application, but not intended to limit the present application, and the protection scope of the present application is not limited to this. Although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that any modification or easy-to-think change, or equivalent replacement of part of the technical features of the technical solutions recorded in the foregoing embodiments can be made within the technical range disclosed by the present application by any person skilled in the art. The modification, change or replacement does not make the corresponding technical solution deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0166] In addition, although the operations of the method of the present application are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in this particular order, or that all of the illustrated operations must be performed to achieve the desired result. Additionally or alternatively, certain steps can be omitted, combined into a single step, and / or divided into multiple steps.

Claims

1. A medical data processing method, comprising: performing feature recognition on an image to be processed collected by an endoscope to obtain feature information included in the image to be processed; wherein the feature information includes a plurality of target features and a feature position coordinate corresponding to each target feature, and one target feature corresponds to one feature position coordinate; determining a feature type of each target feature in the feature information; wherein the feature type at least includes an organ type and a lesion type; determining a standard image at a current position of the endoscope according to the organ type and the feature position coordinate corresponding to the target feature of the organ type; wherein it includes: determining an organ target feature of the organ type from the plurality of target features; obtaining a confidence degree of the organ target feature; obtaining a standard organ target feature with a confidence degree greater than a confidence degree threshold from the organ target feature; determining position relationship information between any two standard organ target features; and determining a standard image at a current position of the endoscope according to the feature position coordinate corresponding to the standard organ target feature and the position relationship information; comparing the feature information with the standard image to obtain error feature information; wherein the error feature information includes at least one error feature and an error feature position coordinate corresponding to the error feature; deleting the error feature information from the feature information to obtain correct feature information.

2. The medical data processing method of claim 1, wherein the comparing the feature information with the standard image to obtain error feature information comprises: obtaining standard organ information from the standard image; wherein the standard organ information at least includes a standard organ type and a coordinate region corresponding to the standard organ type; if an organ type of a certain organ target feature is different from the standard organ type, determining the certain organ target feature as a first error organ target feature; if an organ type of a certain organ target feature is the same as the standard organ type, determining position relationship information between a target feature position coordinate corresponding to the certain organ target feature and a target coordinate region corresponding to the organ type of the certain organ target feature; if the position relationship information indicates that the target feature position coordinate is located outside the target coordinate region, determining the certain organ target feature as a second error organ target feature; determining error feature information according to the first error organ target feature and the second error organ target feature.

3. The medical data processing method of claim 2, wherein after the determining error feature information according to the first error organ target feature and the second error organ target feature, the method further comprises: determining a current lesion organ where a lesion target feature of the lesion type is located according to a feature position coordinate corresponding to the lesion target feature of the lesion type; obtaining a standard lesion organ matching a lesion type of the lesion target feature; if the current lesion organ does not match the standard lesion organ, determining the lesion target feature as an error lesion target feature. Add the error lesion target feature and the feature position coordinates corresponding to the error lesion target feature to the error feature information.

4. The medical data processing method according to any one of claims 1-3, after the correct feature information is obtained, the method further comprises: determining a position region where each correct target feature in the correct feature information is located in the image to be processed; wherein one correct target feature corresponds to one position region; determining a bounding box corresponding to each correct target feature according to the position region corresponding to each correct target feature; wherein one correct target feature corresponds to one bounding box; outputting each bounding box and the feature information of the correct target feature corresponding to each bounding box in the image to be processed.

5. A medical data processing system, comprising: a recognition unit configured to perform feature recognition on an image to be processed collected by an endoscope to obtain feature information included in the image to be processed; wherein the feature information includes a plurality of target features and feature position coordinates corresponding to each target feature, and one target feature corresponds to one feature position coordinate; a first determination unit configured to determine a feature type of each target feature in the feature information; wherein the feature type at least includes an organ type and a lesion type; a second determination unit configured to determine a standard image at a current position of the endoscope according to the organ type and the feature position coordinates corresponding to the target features of the organ type; wherein the second determination unit comprises: determining organ target features of the organ type from the plurality of target features; obtaining a confidence degree of the organ target features; obtaining standard organ target features with a confidence degree greater than a confidence degree threshold from the organ target features; determining position relationship information between any two standard organ target features; and determining a standard image at a current position of the endoscope according to the feature position coordinates corresponding to the standard organ target features and the position relationship information; a comparison unit configured to compare the feature information with the standard image to obtain error feature information; wherein the error feature information includes at least one error feature and error feature position coordinates corresponding to the error feature; a deletion unit configured to delete the error feature information from the feature information to obtain correct feature information.

6. The medical data processing system according to claim 5, the second determination unit determines the standard image at the current position of the endoscope according to the organ type and the feature position coordinates corresponding to the target features of the organ type in the following manner: determining organ target features of the organ type from the plurality of target features; obtaining a confidence degree of the organ target features; obtaining standard organ target features with a confidence degree greater than a confidence degree threshold from the organ target features; determining position relationship information between any two standard organ target features; and determining a standard image at a current position of the endoscope according to the feature position coordinates corresponding to the standard organ target features and the position relationship information. According to the feature position coordinates corresponding to the standard organ target feature and the position relationship information, a standard image at a current position of the endoscope is determined. 7.The medical data processing system of claim 6, wherein the comparison unit compares the feature information with the standard image to obtain the error feature information in the following manner: the standard organ information comprises at least a standard organ type and a coordinate region corresponding to the standard organ type; if an organ type of a certain organ target feature is different from the standard organ type, the certain organ target feature is determined as a first error organ target feature; if the organ type of the certain organ target feature is the same as the standard organ type, position relationship information between a target feature position coordinate corresponding to the certain organ target feature and a target coordinate region corresponding to the organ type of the certain organ target feature is determined; if the position relationship information indicates that the target feature position coordinate is located outside the target coordinate region, the certain organ target feature is determined as a second error organ target feature; and the first error organ target feature and the second error organ target feature are used to determine the error feature information. acquiring standard organ information from the standard image; wherein 8.The medical data processing system of claim 7, wherein the comparison unit is further configured to: after determining the error feature information according to the first error organ target feature and the second error organ target feature, determine a current lesion organ in which a lesion target feature corresponding to a feature position coordinate of the lesion target feature of the lesion type is located; obtain a standard lesion organ matching the lesion type of the lesion target feature; if the current lesion organ does not match the standard lesion organ, determine the lesion target feature as an error lesion target feature; and add the error lesion target feature and the feature position coordinate corresponding to the error lesion target feature to the error feature information. 9.The medical data processing system of claim 5 or 6 or 7 or 8, wherein the deletion unit is further configured to: one correct target feature corresponds to one position region; determine a bounding box corresponding to each correct target feature according to the position region corresponding to each correct target feature; and output each bounding box and feature information of the correct target feature corresponding to each bounding box in the image to be processed. ​ ​ ​ ​ ​ ​ ​ ​ ​ After obtaining the correct feature information, the position region where each correct target feature in the correct feature information is located in the image to be processed is determined; wherein, ​ ​ ​

Images