Method and apparatus for controlling image interpretation for medical examination, and picture archiving and communication system

Abstract

The present disclosure relates to the field of image processing, and provides a method and apparatus for controlling image interpretation for medical examination, and a picture archiving and communication system. The method includes: adding a pathological detection result to a queue, wherein when remaining image interpretation completion durations corresponding to any two pathological detection results in the queue are greater than or equal to a first threshold, a pathological detection result hitting more key pathologies has a higher priority; and generating and sending a label to a picture archiving and communication system, so that the system determines an image interpretation sequence corresponding to a subject based on the label, displays an image and the label of the subject based on the image interpretation sequence, and updates an image interpretation sequence corresponding to another subject.

Description

Docket No. 701818-WO-2METHOD AND APPARATUS FOR CONTROLLING IMAGE INTERPRETATION FOR MEDICAL EXAMINATION, AND PICTURE ARCHIVING AND COMMUNICATIONSYSTEMCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is based on and claims priority to Chinese Application No.202411242641.8, filed on September 5, 2024, the entire contents of which is herein incorporated by reference.TECHNICAL FIELD

[0002] The present disclosure relates to the field of image processing, and in particular, to a method and apparatus for controlling image interpretation for medical examination, and a picture archiving and communication system.BACKGROUND

[0003] Performing medical imaging, e.g. radiological imaging, on a patient to obtain a medical image of the patient, and analyzing the medical image, is one of the most effective means for evaluating a condition of the patient. Medical image analysis may be divided into two phases: a pathological detection phase and an image interpretation phase. Pathological detection is usually completed by a computer. In the pathological detection phase, the computer analyzes the medical image to obtain a pathological detection result of the patient. In the image interpretation phase, the condition of the patient is usually analyzed by a doctor with reference to the medical image and the pathological detection result (namely, image interpretation), and a report recording the condition of the patient is submitted. Medical image analysis is considered to be completed when the report is submitted.

[0004] The time consumed by actual medical image analysis is equal to the total time consumed by pathological detection and image interpretation, and the time consumed by image interpretation further includes time spent waiting for the doctor to perform image interpretation.Docket No. 701818-WO-2A continuous increase in patients, lack of experienced doctors, and wide application of telemedicine will lead to an increase in time spent waiting for a doctor to perform image interpretation. Thus, the time consumed by actual analysis is also increased. Medical image analysis is presented to a doctor in the form of a medical examination. If image interpretation sequences for medical examinations are determined based on the chronological order of imaging of medical images of a plurality of patients, some patients may be in danger of losing their lives due to not receiving timely treatment. Therefore, a technical problem to be urgently resolved in the art is how to help doctors to determine priorities of patients, to reduce the time consumed by image interpretation for a patient with an urgent condition, and reduce the probability of the patient becoming critically ill due to the time consumed by image interpretation for medical examination being excessively high.SUMMARY OF THE INVENTION

[0005] In view of this, the present disclosure provides a method and apparatus for controlling image interpretation for medical examination, and a picture archiving and communication system. For different subjects for which the remaining duration for completing image interpretation is sufficient, according to the method of the present disclosure, a higher priority is set for examination of a subject suspected of having a critical pathology, so as to speed up image interpretation for critical pathological examination, thereby reducing the time consumed by image interpretation for medical examination of the subject suspected of having a critical pathology, and reducing the occurrence of acute and critical illness deterioration caused by the time consumed by image interpretation being excessively high.

[0006] According to one aspect of the present disclosure, provided is a method for controlling image interpretation for medical examination. The method comprises: inputting an image of a current subject into a pathological detection model, the pathological detection model outputting a pathological detection result of the current subject; determining a hit situation of the pathological detection result of the current subject for a critical pathology based on a preset critical pathology list; adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and a remaining image interpretation completion duration corresponding to the pathological detection result of the current subject, wherein whenDocket No. 701818-WO-2 remaining image interpretation completion durations corresponding to any two pathological detection results in the priority sorting queue are greater than or equal to a first threshold, a pathological detection result hitting more critical pathologies has a higher priority; generating a first label, a time label, and a priority label of the current subject, wherein the first label records a critical pathology hit by the pathological detection result of the current subject, the time label records examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in the priority sorting queue; and sending the first label, the time label, and the priority label to a picture archiving and communication system, the system determining an image interpretation sequence corresponding to the current subject based on the priority label and the time label, displaying the image and the first label of the current subject based on the image interpretation sequence, and updating an image interpretation sequence corresponding to another subject.

[0007] In a possible implementation, the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is equal to the difference between expected image interpretation completion time of the current subject and the current time, and the expected image interpretation completion time of the current subject is equal to the sum of the examination completion time of the current subject and an expected image interpretation queue time consumption value.

[0008] In a possible implementation, when the quantities of critical pathologies hit by any two pathological detection results in the priority sorting queue are the same, a pathological detection result corresponding to a shorter remaining image interpretation completion duration has a higher priority.

[0009] In a possible implementation, in the priority sorting queue, the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is less than the first threshold is higher than the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is greater than or equal to the first threshold.

[0010] In a possible implementation, the priority sorting queue comprises a first sub-queue, a second sub-queue and a third sub-queue, the priority of the first sub-queue is higher than the priority of the second sub-queue, the priority of the second sub-queue is higher than the priorityDocket No. 701818-WO-2 of the third sub-queue, and said adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject comprises: adding the pathological detection result of the current subject to the second sub-queue when the quantity of critical pathologies hit by the pathological detection result of the current subject is greater than 0, wherein in the second sub-queue, a pathological detection result hitting more critical pathologies critical pathologies has a higher priority, and when the quantities of critical pathologies hit by pathological detection results are the same, a pathological detection result corresponding to an earlier examination completion time has a higher priority; adding the pathological detection result of the current subject to the third sub-queue when the quantity of critical pathologies hit by the pathological detection result of the current subject is equal to 0, wherein in the third sub-queue, a pathological detection result corresponding to an earlier examination completion time has a higher priority; and monitoring remaining image interpretation completion durations corresponding to pathological detection results in the second sub-queue and the third sub-queue, and when a remaining image interpretation completion duration corresponding to any pathological detection result is less than the first threshold, moving the pathological detection result from a current sub-queue to the first sub-queue, wherein in the first sub-queue, a pathological detection result corresponding to a smaller remaining image interpretation completion duration value has a higher priority.

[0011] In a possible implementation, the method further comprises: determining a hit situation of the pathological detection result of the current subject for an incidental pathology based on a preset incidental pathology list, wherein an urgency level of the incidental pathology is less than an urgency level of the critical pathology; generating a second label of the current subject when the quantity of incidental pathologies hit by the pathological detection result of the current subject is greater than 0, wherein the second label records an incidental pathology hit by the pathological detection result of the current subject; and sending the second label to the picture archiving and communication system, the picture archiving and communication system being further configured to display the second label, and the second label being displayed based on the same image interpretation sequence as the first label.

[0012] In a possible implementation, the method further comprises: acquiring a submission time of a report of the current subject from the picture archiving and communication system;Docket No. 701818-WO-2 determining image analysis time consumption of the current subject based on the difference between the submission time of the report of the current subject and the examination completion time of the current subject; and based on a size relationship between the image analysis time consumption and the expected image interpretation queue time consumption value, adjusting one or a plurality of a parameter of the pathological detection model, the expected image interpretation queue time consumption value or the first threshold.

[0013] According to another aspect of the present disclosure, provided is a system for controlling image interpretation for medical examination. The system comprises: a pathological detection module, used to input an image of a current subject into a pathological detection model, the pathological detection model outputting a pathological detection result of the current subject; a first determining module, used to determine a hit situation of the pathological detection result of the current subject for a critical pathology based on a preset critical pathology list; a sorting module, used to add the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject, wherein when remaining image interpretation completion durations corresponding to any two pathological detection results in the priority sorting queue are greater than or equal to a first threshold, a pathological detection result hitting more critical pathologies has a higher priority; a first generation module, used to generate a first label, a time label, and a priority label of the current subject, wherein the first label records a critical pathology hit by the pathological detection result of the current subject, the time label records examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in the priority sorting queue; and a first sending module, used to send the first label, the time label, and the priority label to a picture archiving and communication system, the system determining an image interpretation sequence corresponding to the current subject based on the priority label and the time label, displaying the image and the first label of the current subject based on the image interpretation sequence, and updating an image interpretation sequence corresponding to another subject.

[0014] In a possible implementation, the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is equal to the difference between expected image interpretation completion time of the current subject and the currentDocket No. 701818-WO-2 time, and the expected image interpretation completion time of the current subject is equal to the sum of the examination completion time of the current subject and an expected image interpretation queue time consumption value.

[0015] In In a possible implementation, when the quantities of critical pathologies hit by any two pathological detection results in the priority sorting queue are the same, a pathological detection result corresponding to a shorter remaining image interpretation completion duration has a higher priority.

[0016] In a possible implementation, in the priority sorting queue, the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is less than the first threshold is higher than the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is greater than or equal to the first threshold.

[0017] In a possible implementation, the priority sorting queue comprises a first sub-queue, a second sub-queue and a third sub-queue, the priority of the first sub-queue is higher than the priority of the second sub-queue, the priority of the second sub-queue is higher than the priority of the third sub-queue, and said adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject comprises: adding the pathological detection result of the current subject to the second sub-queue when the quantity of critical pathologies hit by the pathological detection result of the current subject is greater than 0, wherein in the second sub-queue, a pathological detection result hitting more critical pathologies has a higher priority, and when the quantities of critical pathologies hit by pathological detection results are the same, a pathological detection result corresponding to earlier examination completion time has a higher priority; adding the pathological detection result of the current subject to the third sub-queue when the quantity of critical pathologies hit by the pathological detection result of the current subject is equal to 0, wherein in the third subqueue, a pathological detection result corresponding to an earlier examination completion time has a higher priority; and monitoring remaining image interpretation completion durations corresponding to pathological detection results in the second sub-queue and the third sub-queue, and when a remaining image interpretation completion duration corresponding to anyDocket No. 701818-WO-2 pathological detection result is less than the first threshold, moving the pathological detection result from a current sub-queue to the first sub-queue, wherein in the first sub-queue, a pathological detection result corresponding to a smaller remaining image interpretation completion duration value has a higher priority.

[0018] In a possible implementation, the system further comprises: a second determining module, used to determine a hit situation of the pathological detection result of the current subject for an incidental pathology based on a preset incidental pathology list, wherein an urgency level of the incidental pathology is less than an urgency level of the critical pathology; a second generation module, used to generate a second label of the current subject when the quantity of incidental pathologies hit by the pathological detection result of the current subject is greater than 0, wherein the second label records an incidental pathology hit by the pathological detection result of the current subject; and a second sending module, used to send the second label to the picture archiving and communication system, the picture archiving and communication system being further used to display the second label, and the second label being displayed based on the same image interpretation sequence as the first label.

[0019] In a possible implementation, the system further comprises: an acquisition module, used to acquire submission time of a report of the current subject from the picture archiving and communication system; a third determining module, used to determine image analysis time consumption of the current subject based on the difference between the submission time of the report of the current subject and the examination completion time of the current subject; and an adjustment module, used to adjust one or a plurality of a parameter of the pathological detection model, the expected image interpretation queue time consumption value, or the first threshold based on a size relationship between the image analysis time consumption and the expected image interpretation queue time consumption value.

[0020] According to another aspect of the present disclosure, provided is a picture archiving and communication system. The system is used to: receive an image of a current subject; receive a first label, a time label, and a priority label of the current subject, wherein the first label records a critical pathology hit by a pathological detection result of the current subject, the time label records examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in a priority sorting queue;Docket No. 701818-WO-2 determine, based on the time label, whether a remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to a first threshold; when the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to the first threshold, determine a first image interpretation sequence corresponding to the current subject based on the priority label of the current subject, display the image and the first label of the current subject based on the first image interpretation sequence, and update an image interpretation sequence corresponding to another subject whose image interpretation sequence is lower than or equal to the first image interpretation sequence; and when the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is less than the first threshold, determine a second image interpretation sequence corresponding to the current subject based on the remaining image interpretation completion duration, display the image and the first label of the current subject based on the second image interpretation sequence, and update an image interpretation sequence corresponding to another subject whose image interpretation sequence is lower than or equal to the second image interpretation sequence, wherein the second image interpretation sequence is higher than the first image interpretation sequence.

[0021] In a possible implementation, the system is further used to: receive a second label of the current subject, wherein the second label records an incidental pathology hit by the pathological detection result of the current subject; and display the second label based on the same image interpretation sequence as the first label.

[0022] According to another aspect of the present disclosure, provided is an apparatus for controlling image interpretation for medical examination. The apparatus comprises: a processor; and a memory, used to store instructions executable by the processor, wherein the processor is configured to implement the above method when executing the instructions stored in the memory.

[0023] According to another aspect of the present disclosure, provided is a non-volatile computer-readable storage medium, having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the above method.Docket No. 701818-WO-2

[0024] According to another aspect of the present disclosure, provided is a computer program product, comprising computer-readable code or a non-volatile computer-readable storage medium bearing computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above method.

[0025] According to the method for controlling image interpretation for medical examination in the embodiments of the present disclosure, the image of the current subject is input into the pathological detection model, and the pathological detection model outputs the pathological detection result of the current subject, to complete the work of a pre-examination phase. The hit situation of the pathological detection result of the current subject for the critical pathology is determined based on the preset critical pathology list; the pathological detection result of the current subject is added to the priority sorting queue based on the hit situation and the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject, wherein when remaining image interpretation completion durations corresponding to any two pathological detection results in the priority sorting queue are greater than or equal to the first threshold, a pathological detection result hitting more critical pathologies has a higher priority. In this case, for different subjects for which the remaining duration for completing image interpretation is sufficient, a higher priority may be set for an examination of a subject suspected of having a critical pathology. The first label, the time label, and the priority label of the current subject are generated, wherein the first label records the critical pathology hit by the pathological detection result of the current subject, the time label records the examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in the priority sorting queue. The first label, the time label, and the priority label are sent to the picture archiving and communication system, the system determines the image interpretation sequence corresponding to the current subject based on the time label and the priority label, displays the image and the first label of the current subject based on the image interpretation sequence, and updates the image interpretation sequence corresponding to the another subject. In this way, the picture archiving and communication system can display medical examination information of the current subject and automatically perform image interpretation sorting on each subject, so that a doctor can see, in a work list, an image interpretation sequence of each subject that is updated in realDocket No. 701818-WO-2 time. In this case, the doctor can start image interpretation for examination of the subject suspected of having a critical pathology earlier, thereby reducing the time consumed by image interpretation for medical examination of the subject suspected of having a critical pathology, and reducing the occurrence of acute and critical illness deterioration caused by the time consumed by image interpretation being excessively high.

[0026] Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments provided with reference to the accompanying drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, incorporated in and constituting a part of the specification, illustrate exemplary embodiments, features, and aspects of the present disclosure together with the specification, and serve to explain the principles of the present disclosure.

[0028] FIG. 1 shows an exemplary application scenario of a method for controlling image interpretation for medical examination according to an embodiment of the present disclosure;

[0029] FIG. 2 is a schematic flowchart of a method for controlling image interpretation for medical examination according to an embodiment of the present disclosure;

[0030] FIG. 3 shows an example of a pathological detection model according to an embodiment of the present disclosure;

[0031] FIG. 4a is a schematic diagram of a structure of a priority sorting queue according to an embodiment of the present disclosure;

[0032] FIG. 4b is a schematic diagram of a structure of a priority sorting queue according to an embodiment of the present disclosure;

[0033] FIG. 5 is a schematic diagram of a structure of a priority sorting queue according to an embodiment of the present disclosure;

[0034] FIG. 6 is a schematic flowchart of a method for controlling image interpretation for medical examination according to an embodiment of the present disclosure;Docket No. 701818-WO-2

[0035] FIG. 7 is a schematic diagram of a workflow of a picture archiving and communication system according to the present disclosure;

[0036] FIG. 8 is a schematic diagram of a work list of a picture archiving and communication system according to an embodiment of the present disclosure;

[0037] FIG. 9a is a schematic diagram of a work list of a picture archiving and communication system according to an embodiment of the present disclosure;

[0038] FIG. 9b is a schematic diagram of a work list of a picture archiving and communication system according to an embodiment of the present disclosure;

[0039] FIG. 10 is a schematic diagram of a workflow of a picture archiving and communication system according to the present disclosure;

[0040] FIG. 11 is a schematic diagram of a structure of a system for controlling image interpretation for medical examination according to an embodiment of the present disclosure; and

[0041] FIG. 12 is a block diagram of an apparatus 1900 according to an embodiment of the present disclosure.DETAILED DESCRIPTION

[0042] Various exemplary embodiments, features, and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the drawings represent elements having the same or similar functions. While various aspects of the embodiments are illustrated in the accompanying drawing, the accompanying drawings are not necessarily drawn to scale unless specifically stated.

[0043] The word “exemplary” dedicated herein means “used as an example or an embodiment, or illustrative”. Any of the embodiments illustrated herein as “exemplary” is not necessarily interpreted as being superior to or better than other embodiments.

[0044] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the detailed description below. It should be understood by those skilled in the art that the present disclosure can be implemented without some of the specific details. InDocket No. 701818-WO-2 some examples, the methods, means, elements, and circuits well known to those skilled in the art are not described in detail in order to highlight the main idea of the present disclosure.

[0045] Radiological examination is used as an example of medical examination below. A person skilled in the art should understand that a method for controlling image interpretation for medical examination of the present disclosure is also applicable to medical examinations such as ultrasonic examination and magnetic resonance examination. The specific type of medical examination is not limited in the present disclosure.

[0046] For radiological examination, an expected image interpretation queue time consumption value may be preset, and the proportion of radiological examinations, whose image interpretation time consumption exceeds the expected image interpretation queue time consumption value, in all radiological examinations is determined based on image interpretation time consumption of radiological examinations whose image interpretation has already been completed. The efficiency of radiological examination may be evaluated based on said proportion. The time consumed by image interpretation includes time spent waiting for a doctor to complete image interpretation of another examination with a higher priority. In the prior art, the time consumed by image interpretation for most radiological examinations is high. The reasons are as follows:

[0047] 1. Because radiological examination is widely applied, the quantity of patients who need to undergo radiological examination offline in each hospital is large. For a patient with a low image interpretation sequence, time spent by the patient waiting for a doctor to perform image interpretation is increased, and therefore, the time consumed by image interpretation is also increased.

[0048] 2. Differences in medical treatment between regions are inevitable, and it is difficult to ensure that each hospital can make diagnoses for complex conditions. Based on this, some hospitals use an “Internet + medical treatment” telemedicine mode, enabling a hospital with strong medical treatment capabilities to provide medical support to a hospital with weak medical treatment capabilities. For example, after a radiological imaging device of any hospital supporting “Internet + medical treatment” generates a radiological image for a patient offline, the image may be transmitted to another hospital that supports “Internet + medical treatment” and that has stronger medical treatment capabilities, and the other hospital completes pathologicalDocket No. 701818-WO-2 detection and image interpretation, and returns a report to the hospital that generated the image. In this case, an offline patient of the hospital with weak medical treatment capabilities also serves as an online patient of the hospital with strong medical capabilities. Moreover, the quantity of offline patients of the hospital with strong medical capabilities is not reduced. Consequently, the quantity of patients of the hospital with strong medical capabilities is larger, a patient with a low image interpretation sequence needs to spend more time on waiting for a doctor to perform image interpretation, resulting in greater time consumed by image interpretation.

[0049] 3. A pathological detection result obtained by a computer can only be used as a reference, and the step of completing image interpretation manually by a doctor is indispensable. In addition, radiological image analysis is highly dependent on the experience of a doctor. Most primary medical staff find that it is difficult to accurately complete this task. Consequently, the quantity of doctors who can undertake image interpretation tasks is small. This further increases the time consumed by image interpretation.

[0050] Radiological image analysis is presented to the doctor in the form of radiological examination. When the time consumed by image interpretation for radiological examination is high, if image interpretation sequences of radiological examinations are determined based on imaging sequences of radiological images of a plurality of patients, some patients in urgent need of treatment may miss the golden time for emergency rescue due to not receiving prompt image interpretation, leading to condition deterioration. Therefore, a technical problem to be urgently resolved in the art is how to help doctors to determine an image interpretation sequence of radiological examination to reduce the time consumed by image interpretation for a patient with an urgent condition, and reduce the probability of the patient becoming critically ill due to the time consumed by image interpretation for medical examination being excessively high.

[0051] In addition, a condition of a patient who does not urgently need treatment may become more severe as time goes by. Even if the patient's health condition is good, if the time consumed by image interpretation is excessively high, the examination experience of the patient may be degraded. Therefore, it is also necessary to set a higher priority for a patient who spends more time on image interpretation, so as to prevent a condition of the patient from deteriorating and improve the examination experience of the patient.Docket No. 701818-WO-2

[0052] In view of this, the present disclosure provides a method and apparatus for controlling image interpretation for medical examination and a picture archiving and communication system. For different subjects for which the remaining duration for completing image interpretation is sufficient, according to the method of the present disclosure, a higher priority is set for examination of a subject suspected of having a critical pathology, so as to speed up image interpretation for critical pathological examination, thereby reducing the time consumed by image interpretation for medical examination of the subject suspected of having a critical pathology, and reducing the occurrence of acute and critical illness deterioration caused by the time consumed by image interpretation for medical examination being excessively high.

[0053] Further, for different subjects for which the remaining duration for completing image interpretation is sufficient and for which the remaining duration for completing image interpretation is insufficient, according to the method of the present disclosure, a higher priority may be set for a subject for which the remaining duration for completing image interpretation is insufficient, thereby raising an image interpretation sequence of such a subject, reducing the time consumed by image interpretation for such a subject, preventing a subject to be examined from waiting for an excessively long time, and improving the examination experience of the subject to be examined.

[0054] FIG. 1 shows an exemplary application scenario of a method for controlling image interpretation for medical examination according to an embodiment of the present disclosure.

[0055] As shown in FIG. 1, the method for controlling image interpretation for medical examination in this embodiment of the present disclosure may be applied to a system for controlling image interpretation for medical examination. The system for controlling image interpretation for medical examination may be disposed on a first device. A pathological detection model (not shown) may be further disposed on the first device. The pathological detection model may be disposed in or outside the system for controlling image interpretation for medical examination. This is not limited in the present disclosure. The first device, a second device, and a third device may communicate with each other. The second device may be a medical imaging device, used to generate one of a radiological image (for example, an X-ray image), an ultrasound image, a magnetic resonance image, or the like. A picture archiving and communication system is disposed on the third device.Docket No. 701818-WO-2

[0056] For example, the second device generates a radiological image. Each time the second device collects an image of a subject (that is, the patient described above), the second device may transmit the image to the system for controlling image interpretation for medical examination on the first device and the picture archiving and communication system on the third device.

[0057] When the system for controlling image interpretation for medical examination of the present disclosure performs the method for controlling image interpretation for medical examination, the image may be input into the pathological detection model, and the pathological detection model completes the work of a pathological detection phase of radiological examination to obtain a pathological detection result. Based on the pathological detection result, the system for controlling image interpretation for medical examination may sort all subjects for which pathological detection has already been completed but reports have not been submitted, to obtain a priority label of each subject, and transmit the priority label, the pathological detection result and the like to the picture archiving and communication system on the third device.

[0058] The picture archiving and communication system may determine an image interpretation sequence corresponding to each subject based on the priority label, to display the image and the pathological detection result of each subject in a work list based on the determined image interpretation sequence. Different subjects correspond to different image interpretation sequences. A higher image interpretation sequence indicates that a display position of an image and a pathological detection result is put further forward.

[0059] A person skilled in the art should understand that the system for controlling image interpretation for medical examination and the picture archiving and communication system may alternatively be disposed on the same device. Whether the system for controlling image interpretation for medical examination and the picture archiving and communication system are disposed on the same device is not limited in the embodiments of the present disclosure.

[0060] A doctor can check the image and the pathological detection result of each subject on the picture archiving and communication system, and submits a report after completing image interpretation. After a report of any subject is submitted, radiological image analysis of the subject is completed.Docket No. 701818-WO-2

[0061] FIG. 2 is a schematic flowchart of a method for controlling image interpretation for medical examination according to an embodiment of the present disclosure.

[0062] As shown in FIG. 2, in a possible implementation, the method includes the following steps:

[0063] Step S21 : Inputting an image of a current subject into a pathological detection model, the pathological detection model outputting a pathological detection result of the current subject;

[0064] Step S22: Determining a hit situation of the pathological detection result of the current subject for a critical pathology based on a preset critical pathology list;

[0065] Step S23: Adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and a remaining image interpretation completion duration corresponding to the pathological detection result of the current subject, wherein when remaining image interpretation completion durations corresponding to any two pathological detection results in the priority sorting queue are greater than or equal to a first threshold, a pathological detection result hitting more critical pathologies has a higher priority;

[0066] Step S24: Generating a first label, a time label, and a priority label of the current subject, wherein the first label records a critical pathology hit by the pathological detection result of the current subject, the time label records examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in the priority sorting queue;

[0067] Step S25: Sending the first label, the time label, and the priority label to a picture archiving and communication system, the system determining an image interpretation sequence corresponding to the current subject based on the priority label and the time label, displaying the image and the first label of the current subject based on the image interpretation sequence, and updating an image interpretation sequence corresponding to another subject.

[0068] For example, in step S21, the image of the current subject may be input into the pathological detection model, and the pathological detection model may output the pathological detection result of the current subject.Docket No. 701818-WO-2

[0069] The image may be any type of medical image, for example, a radiological image (for example, an X-ray image), an ultrasonic image, or a magnetic resonance image. The radiological image is used as an example for description below.

[0070] The pathological detection model may be a deep convolution neural network (DCNN) model implemented based on existing technology. The pathological detection model may include a plurality of sub-models, each sub-model is used to detect one type of pathology, and different sub-models detect different pathologies. In practical application, images input into the pathological detection model may include images of a plurality of body parts (for example, the abdomen and chest), so that the pathological detection model can detect pathologies of the plurality of body parts (for example, pneumothorax, pneumoperitoneum, and acute pancreatitis). When a data analysis capability of the DCNN model used is strong enough, the pathological detection model may detect up to 127 pathologies. Images of body parts that are included in the images input into the pathological detection model and pathologies of body parts that can be specifically detected by the pathological detection model are not limited in the present disclosure. For clarity of description, an example in which the pathological detection model is used to detect chest pathology is used below.

[0071] FIG. 3 shows an example of a pathological detection model according to an embodiment of the present disclosure.

[0072] As shown in FIG. 3, the pathological detection model includes a sub-model 1, a submodel 2, ..., and a sub-model 12. Each sub-model may be used to detect one type of pathology, and different sub-models detect different pathologies. Therefore, the pathological detection model may detect up to 12 types of pathologies. Each type of pathology represents one cause of a disease. Each sub-model may include two convolutional neural networks (CNN). The two CNNs have different resolutions. A pathological detection result of the sub-model may be obtained after detection results output by the two CNNs are integrated. The detection result output by each CNN may include a positive / negative result indicating whether a corresponding pathology is detected and a confidence level of the positive / negative result. The positive result represents that the pathology is detected, and the negative result represents that the pathology is not detected. The confidence level may indicate the probability that the pathology is positive / negative. Integrating the detection results output by the two CNNs to obtain theDocket No. 701818-WO-2 pathological detection result of the sub-model may be implemented based on existing technology. For example, the weight of each CNN may be determined based on a positive / negative result included in each CNN, and whether the pathological detection result of the sub-model includes a positive result or a negative result and a confidence level of the positive / negative result in the pathological detection result of the sub-model are determined based on a weighted summation result of confidence levels of positive / negative results of the two CNNs. Pathological detection results of the plurality of sub-models are summarized as the pathological detection result of the current subject that is output by the pathological detection model. That is, the pathological detection result output by the pathological detection model includes a pathological detection result of each sub-model.

[0073] The pathology detected by each sub-model is shown in Table 1.Table 1

[0074] A person skilled in the art should understand that the 12 types of pathologies shown in Table 1 are 12 types of the most common chest pathologies. In practical application, the quantity of sub-models and the pathology detected by each sub-model may be changed asDocket No. 701818-WO-2 required. Which pathologies are specifically detected when the pathological detection model is used to detect the chest pathologies is not limited in the embodiments of the present disclosure.

[0075] As shown in FIG. 3, the pathological detection result output by the pathological detection model may include a positive / negative result indicating whether each type of pathology is detected, and a confidence level of the positive / negative result. Only pathologies with positive results, including pneumomediastinum, cardiomegaly, and pleural effusion / abnormality, are shown in FIG. 3. A confidence level of pneumomediastinum is 95%, a confidence level of cardiomegaly is 68%, and a confidence level of pleural effusion / abnormality is 89%.

[0076] The pathological detection model may be pre-trained. Training of the pathological detection model may be implemented based on existing technology. For example, an image sample is input into the pathological detection model to obtain a prediction result, and the pathological detection model is adjusted based on the prediction result and a label until training is completed, wherein the label may represent an actual pathology corresponding to the image sample. A specific training manner of the pathological detection model is not described in detail herein.

[0077] Before the image is input into the pathological detection model, the image may be preprocessed, for example, the image is cropped, scaled down or the like to adjust the size of the image, so that the size of the image is the same as a preset input size of the pathological detection model. A specific adjustment manner of the size of the image and a specific value of the preset input size are not limited in the embodiments of the present disclosure.

[0078] When the pathological detection model is used to detect chest pathologies, only a chest image is valid for the pathological detection model. Therefore, the chest image may be selected by using a classification model (for example, an image projection classifier). Further, chest images at fixed angles, for example, front anterioposterior (AP) and posterioanterior (PA) chest images, may be selected, to make pathological detection easier. The classification model may implement classification based on existing technology. A specific classification manner of the classification model is not described in detail herein again.

[0079] In step S21, a selected image may be input into the pathological detection model. The image may be input into each sub-model separately. After each sub-model separately detects a pathology, pathological detection results are summarized. Exemplary pathologicalDocket No. 701818-WO-2 detection manners of the sub-models and a summarization manner of the pathological detection results of the sub-models are described above, and are not described again here.

[0080] Urgency levels of different pathologies are different. A pathology with a high urgency level that needs to be treated as soon as possible may be marked as a critical pathology. The urgency level of each type of pathology may be set by a doctor to obtain a critical pathology list. Table 2 shows an example of the critical pathology list. An identifier of a critical pathology may be C.Table 2

[0081] A person skilled in the art should understand that whether each type of pathology is a critical pathology may be adjusted at any time based on an application scenario. Which pathologies are critical pathologies is not limited in the present disclosure.

[0082] In step S22, the hit situation of the pathological detection result of the current subject for the critical pathology may be determined based on the preset critical pathology list. The hit situation of the pathological detection result of the current subject for the critical pathology may indicate the quantity of critical pathologies hit by the pathological detection result of the current subject. Optionally, the hit situation may further include a confidence level. The pathological detection result of FIG. 3 is used as an example. The pathological detection result hits one critical pathology (pneumomediastinum), and the confidence level is 95%.

[0083] In step S23, the pathological detection result of the current subject may be added to the priority sorting queue based on the hit situation and the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject. An expected image interpretation queue time consumption value may be preset. A specific value of the expected image interpretation queue time consumption value may be determined asDocket No. 701818-WO-2 required. When radiological examination of the current subject is completed (that is, an imaging scanning process of the current subject ends, and a radiological image of the current subject is generated), the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is largest and equal to the expected image interpretation queue time consumption value. As time goes by, the remaining image interpretation completion duration is gradually shortened. If an image interpretation report of the current subject is not submitted after the remaining image interpretation completion duration is shortened to 0, the remaining image interpretation completion duration becomes a negative value. An exemplary calculation manner of the remaining image interpretation completion duration is provided below.

[0084] Different expected image interpretation queue time consumption values may be preset based on the quantity of critical pathologies hit by the pathological detection result, so that a pathological detection result hitting more critical pathologies corresponds to a smaller expected image interpretation queue time consumption value. Alternatively, only one expected image interpretation queue time consumption value may be preset. Regardless of the quantity of critical pathologies hit by the pathological detection result, the pathological detection result corresponds to the same expected image interpretation queue time consumption value. A correspondence between the quantity of critical pathologies hit by the pathological detection result and the expected image interpretation queue time consumption value is not limited in the embodiments of the present disclosure.

[0085] When the remaining image interpretation completion durations corresponding to any two pathological detection results in the priority sorting queue are greater than or equal to the first threshold, the remaining image interpretation completion durations may be considered to be sufficient. In this case, a pathological detection result hitting more critical pathologies has a higher priority. For example, in the priority sorting queue, a pathological detection result 1 hits two critical pathologies and a corresponding remaining image interpretation completion duration is greater than the first threshold, and a pathological detection result 2 hits three critical pathologies and a corresponding remaining image interpretation completion duration is greater than the first threshold. In this case, the priority of the pathological detection result 2 may be higher than that of the pathological detection result 1. An exemplary sorting manner of the pathological detection results in the priority sorting queue is provided below.Docket No. 701818-WO-2

[0086] The first threshold may be set to any value less than the expected image interpretation queue time consumption value, for example, set to 0. A specific value of the first threshold is not limited in the embodiments of the present disclosure.

[0087] In step S24, the first label, the time label, and the priority label of the current subject may be generated, so that the first label records the critical pathology hit by the pathological detection result of the current subject, the time label records the examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in the priority sorting queue. The examination completion time refers to a time point at which an imaging scanning process of a subject ends and a medical image of the subject is generated.

[0088] The first label, the time label, and the priority label may be digital imaging and communications in medicine (DICOM) labels, and may be encapsulated into a health level seven (HL7) message. An exemplary encapsulation manner is provided below.

[0089] In step S25, the first label, the time label, and the priority label may be sent to the picture archiving and communication system. For example, an encapsulated data packet may be sent to the picture archiving and communication system. The picture archiving and communication system may determine the image interpretation sequence corresponding to the current subject based on the time label and the priority label, so as to display the image and the first label of the current subject based on the determined image interpretation sequence. The picture archiving and communication system may further update the image interpretation sequence of the another subject. Working details of the picture archiving and communication system are provided below.

[0090] According to the method for controlling image interpretation for medical examination in the embodiments of the present disclosure, the image of the current subject is input into the pathological detection model, and the pathological detection model outputs the pathological detection result of the current subject, to complete the work of a pre-examination phase. The hit situation of the pathological detection result of the current subject for the critical pathology is determined based on the preset critical pathology list; the pathological detection result of the current subject is added to the priority sorting queue based on the hit situation and the remaining image interpretation completion duration corresponding to the pathologicalDocket No. 701818-WO-2 detection result of the current subject, wherein when the remaining image interpretation completion durations corresponding to any two pathological detection results in the priority sorting queue are greater than or equal to the first threshold, a pathological detection result hitting more critical pathologies has a higher priority. In this case, for different subjects for which the remaining duration for completing image interpretation is sufficient, a higher priority may be set for an examination of a subject suspected of having a critical pathology. The first label, the time label, and the priority label of the current subject are generated, wherein the first label records the critical pathology hit by the pathological detection result of the current subject, the time label records the examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in the priority sorting queue. The first label, the time label, and the priority label are sent to the picture archiving and communication system, the system determines the image interpretation sequence corresponding to the current subject based on the time label and the priority label, displays the image and the first label of the current subject based on the image interpretation sequence, and updates the image interpretation sequence corresponding to the another subject. In this way, the picture archiving and communication system can display medical examination information of the current subject and automatically perform image interpretation sorting on each subject, so that a doctor can see, in a work list, an image interpretation sequence of each subject that is updated in real time. In this case, the doctor can start image interpretation for examination of the subject suspected of having a critical pathology earlier, thereby reducing the time consumed by image interpretation for medical examination of the subject suspected of having a critical pathology, and reducing the occurrence of acute and critical illness deterioration caused by the time consumed by image interpretation being excessively high.

[0091] In a possible implementation, the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is equal to the difference between expected image interpretation completion time of the current subject and the current time, and the expected image interpretation completion time of the current subject is equal to the sum of the examination completion time of the current subject and an expected image interpretation queue time consumption value.

[0092] For example, the examination completion time of the current subject is acquired while the image of the current subject is acquired. The expected image interpretationDocket No. 701818-WO-2 completion time of the current subject may be determined based on the examination completion time of the current subject and the preset expected image interpretation queue time consumption value. The expected image interpretation completion time of the current subject is equal to the sum of the examination completion time of the current subject and the expected image interpretation queue time consumption value. For example, the examination completion time of the current subject is 9:00, and the expected image interpretation queue time consumption value is 3 hours. Then, the expected image interpretation completion time of the current subject may be 12:00.

[0093] The expected image interpretation queue time consumption value may be customized and adjusted by a user based on an actual situation. For example, when the method of the present disclosure is used in a hospital with a high patient throughput, a large expected image interpretation queue time consumption value may be set. Otherwise, when the method is used in a hospital with a large patient volume, a small expected image interpretation queue time consumption value may be set. The specific manner of setting the expected image interpretation queue time consumption value is not limited in the present disclosure.

[0094] The remaining image interpretation completion duration corresponding to the pathological detection result of the current subject may be calculated based on the expected image interpretation completion time of the current subject and the current time. The remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is equal to the difference between the expected image interpretation completion time of the current subject and the current time. For example, when the expected image interpretation completion time of the current subject is 12:00 and the current time is 10:00, the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject may be 2 hours.

[0095] The priority sorting queue of this embodiment of the present disclosure may also use more sorting rules. Examples of sorting rules used by the priority sorting queue are provided below.

[0096] In a possible implementation, when the quantities of critical pathologies hit by any two pathological detection results in the priority sorting queue are the same, a pathologicalDocket No. 701818-WO-2 detection result corresponding to a shorter remaining image interpretation completion duration has a higher priority.

[0097] For example, if the quantities of critical pathologies hit by any two pathological detection results are the same, this represents that urgency levels of conditions of two corresponding subjects are close to each other. In this case, image interpretation for examination of a subject waiting for a longer time may be preferably completed. That is, a first input first output principle may be followed, so that a pathological detection result corresponding to a shorter remaining image interpretation completion duration has a higher priority. For example, if a pathological detection result 3 hits two critical pathologies and a corresponding remaining image interpretation completion duration is 2 hours, and a pathological detection result 4 hits two critical pathologies and a corresponding remaining image interpretation completion duration is 1 hour, the priority of the pathological detection result 4 may be higher than that of the pathological detection result 3.

[0098] It should be noted that the quantities of hit critical pathologies being the same includes a case in which the quantities of hit critical pathologies are greater than 0 and a case in which the quantities of hit critical pathologies are equal to 0.

[0099] In this way, when the quantities of hit critical pathologies are the same, image interpretation of a subject waiting for a longer time is performed earlier, thereby improving the examination experience of a subject to be examined.

[0100] In a possible implementation, in the priority sorting queue, the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is less than the first threshold is higher than the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is greater than or equal to the first threshold.

[0101] For example, in the priority sorting queue, if the remaining image interpretation completion duration of a specific pathological detection result is already less than the first threshold, this represents that the image interpretation waiting time of a corresponding subject has expired. If a remaining image interpretation completion duration of another pathological detection result is greater than or equal to the first threshold, it represents that image interpretation waiting time of a corresponding subject is still short. In this case, imageDocket No. 701818-WO-2 interpretation of the subject whose waiting time has expired may be preferably completed without considering the quantity of critical pathologies hit by the pathological detection result. That is, in the priority sorting queue, the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is less than the first threshold is higher than the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is greater than or equal to the first threshold.

[0102] For example, if a pathological detection result 5 hits one critical pathology and a corresponding remaining image interpretation completion duration is -0.5 hours, and a pathological detection result 6 hits two critical pathologies and a corresponding remaining image interpretation completion duration is 1 hour, the priority of the pathological detection result 5 may be higher than that of the pathological detection result 6.

[0103] In this way, according to the method for controlling image interpretation for medical examination of the present disclosure, a higher priority may be set for a subject with insufficient remaining image interpretation completion duration, thereby reducing the time consumed by image interpretation for the subject with insufficient remaining image interpretation completion duration, preventing a subject to be examined from waiting for an excessively long time, and improving the examination experience of the subject to be examined.

[0104] A person skilled in the art should understand that the manner of setting priority is not limited to the foregoing manner. For example, in another possible implementation, for a plurality of pathological detection results whose remaining image interpretation completion duration is less than the first threshold, priority sorting can still be performed based on the quantities of hit critical pathologies. The larger the quantity of hit critical pathologies, the higher the priority.

[0105] An exemplary method for implementing the above function by using the priority sorting queue of this embodiment of the present disclosure is described below.

[0106] In a possible implementation, the priority sorting queue includes a first sub-queue, a second sub-queue and a third sub-queue, the priority of the first sub-queue is higher than the priority of the second sub-queue, and the priority of the second sub-queue is higher than the priority of the third sub-queue.Docket No. 701818-WO-2

[0107] Said adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject includes:

[0108] Adding the pathological detection result of the current subject to the second subqueue when the quantity of critical pathologies hit by the pathological detection result of the current subject is greater than 0, wherein in the second sub-queue, a pathological detection result hitting more critical pathologies has a higher priority, and when the quantities of critical pathologies hit by pathological detection results are the same, a pathological detection result corresponding to earlier examination completion time has a higher priority;

[0109] Adding the pathological detection result of the current subject to the third sub-queue when the quantity of critical pathologies hit by the pathological detection result of the current subject is equal to 0, wherein in the third sub-queue, a pathological detection result corresponding to earlier examination completion time has a higher priority; and

[0110] Monitoring remaining image interpretation completion durations corresponding to pathological detection results in the second sub-queue and the third sub-queue, and when remaining image interpretation completion duration corresponding to any pathological detection result is less than the first threshold, moving the pathological detection result from a current subqueue to the first sub-queue, wherein in the first sub-queue, a pathological detection result corresponding to a smaller remaining image interpretation completion duration value has a higher priority.[0U1] FIG. 4a-FIG. 5 are each a schematic diagram of a structure of a priority sorting queue according to an embodiment of the present disclosure.

[0112] For example, as shown in FIG. 4a-FIG. 5, the priority sorting queue may include the first sub-queue, the second sub-queue, and the third sub-queue. The priority of the first subqueue is higher than the priority of the second sub-queue, and the priority of the second subqueue is higher than the priority of the third sub-queue. The quantity of critical pathologies hit by a pathological detection result in the second sub-queue is greater than 0, and the quantity of critical pathologies hit by a pathological detection result in the third sub-queue is equal to 0. Remaining image interpretation completion durations corresponding to the pathological detection results in the second sub-queue and the third sub-queue are greater than or equal to theDocket No. 701818-WO-2 first threshold, and remaining image interpretation completion durations corresponding to the pathological detection results in the first sub-queue are less than the first threshold.

[0113] For radiological examination of a single subject, time consumed by pathological detection and determining a hit situation of a pathological detection result for a critical pathology is usually less than the expected image interpretation queue time consumption value. Therefore, a sorting module adds the pathological detection result of the current subject to the priority sorting queue, the remaining image interpretation completion duration of the pathological detection result usually being greater than the first threshold. Therefore, the pathological detection result of the current subject may be added to one of the second sub-queue or the third sub-queue based on the quantity of critical pathologies hit by the pathological detection result of the current subject.

[0114] For example, when the quantity of critical pathologies hit by the pathological detection result of the current subject is greater than 0, the pathological detection result of the current subject may be added to the second sub-queue; or when the quantity of critical pathologies hit by the pathological detection result of the current subject is equal to 0, the pathological detection result of the current subject may be added to the third sub-queue.

[0115] In the second sub-queue, a pathological detection result hitting more critical pathologies has a higher priority, and when the quantities of critical pathologies hit by pathological detection results are the same, a pathological detection result corresponding to earlier examination completion time has a higher priority. For example, as shown in FIG. 4a, it is assumed that a pathological detection result 1 and a pathological detection result 2 already exist in the second sub-queue, and the pathological detection result 1 hits two critical pathologies (2C) and the pathological detection result 2 hits three critical pathologies (3C). Then, the priority of the pathological detection result 1 is second only to that of the pathological detection result 2.

[0116] As shown in FIG. 4b, when a pathological detection result 7 is added to the priority sorting queue, if the pathological detection result 7 hits three critical pathologies (3C), the priority of the pathological detection result 7 may be higher than that of the pathological detection result 1. It is assumed that the examination completion time corresponding to the pathological detection result 1 is 9:00, the examination completion time corresponding to theDocket No. 701818-WO-2 pathological detection result 2 is 10:00, and the examination completion time corresponding to the pathological detection result 7 is 10:30. Then, the examination completion time corresponding to the pathological detection result 2 is earlier than that of the pathological detection result 7. In addition, the quantity of critical pathologies hit by the pathological detection result 2 and the quantity of critical pathologies hit by the pathological detection result 7 are the same. Therefore, the priority of the pathological detection result 2 is higher than that of the pathological detection result 7.

[0117] That is, after the pathological detection result 7 is added to the priority sorting queue, the priorities of the pathological detection result 2 and a pathological detection result with a higher priority remain unchanged, the priority of the pathological detection result 7 is equal to the priority of the pathological detection result 1 before the pathological detection result 7 is added to the priority sorting queue, and the priorities of the pathological detection result 1 and a pathological detection result with a lower priority are lowered by one level.

[0118] The priorities of pathological detection results in the third sub-queue are determined by using a pre-queuing timestamp algorithm. The pre-queuing timestamp algorithm ensures that the pathological detection results in the queue are sorted based on examination completion time sequence. In addition, a pathological detection result corresponding to an earlier examination completion time has a higher priority. Because a pathological detection result of an image that is submitted earlier is added to the third sub-queue earlier, the pathological detection results in the third sub-queue are also sorted based on a first input first output (FIFO) sequence. For example, as shown in FIG. 4b, it is assumed that a pathological detection result 8 and a pathological detection result 9 already exist in the third sub-queue, the quantities of critical pathologies hit by the pathological detection result 8 and the pathological detection result 9 are equal to 0 (0C), the examination completion time corresponding to the pathological detection result 8 is 11 :00, and the examination completion time corresponding to the pathological detection result 9 is 11 :30. Then, the priority of the pathological detection result 9 in the third sub-queue is lowest. When a pathological detection result 10 hitting 0 (0C) critical pathologies is added to the third sub-queue, if the examination completion time corresponding to the pathological detection result 10 is 12:00, the priority of the pathological detection result 10 may be lower than that of the pathological detection result 9. In this case, the priorities of original pathological detection results in the priority sorting queue remain unchanged.Docket No. 701818-WO-2

[0119] The remaining image interpretation completion durations corresponding to the pathological detection results in the second sub-queue and the third sub-queue may be monitored in real time by using an internal queuing timestamp check algorithm. When a remaining image interpretation completion duration corresponding to any pathological detection result is less than the first threshold, the pathological detection result is moved from a current sub-queue to the first sub-queue. In the first sub-queue, a pathological detection result corresponding to an earlier examination completion time has a higher priority.

[0120] For example, after the pathological detection result 10 is added to the priority sorting queue, at a specific moment, if the remaining image interpretation completion duration corresponding to the pathological detection result 1 in the second sub-queue is less than the first threshold, the pathological detection result 1 may be moved from the second sub-queue to the first sub-queue. As shown in FIG. 5, because move-in times of original pathological detection results (pathological detection results 11-13) in the first sub-queue are earlier, values of corresponding remaining image interpretation completion durations are also smaller. Therefore, the priority of the pathological detection result 1 in the first sub-queue may be lowest. That is, the pathological detection results in the first sub-queue are also sorted based on the first input first output (FIFO) sequence. In this case, in the priority sorting queue, the priorities of the original pathological detection results in the first sub-queue remain unchanged. In the second sub-queue, the priority of another pathological detection result whose original priority is higher than that of the pathological detection result 1 is lowered by one level, and another pathological detection result whose original priority is lower than that of the pathological detection result 1 and the priorities of the pathological detection results in the third sub-queue remain unchanged.

[0121] After the pathological detection result in the third sub-queue is moved to the first sub-queue, a manner of changing the priorities of the pathological detection results in each subqueue is similar to moving the pathological detection result in the second sub-queue to the first sub-queue. Details are not described again here.

[0122] A person skilled in the art should understand that priority sorting of the first subqueue and the third sub-queue may alternatively be implemented in more ways. For example, a pathological detection result corresponding to a shorter remaining image interpretation completion duration may have a higher priority. The specific manner of sorting priority in theDocket No. 701818-WO-2 first sub-queue and the third sub-queue is not limited in the embodiments of the present disclosure, provided that the priority sorting queue can meet the conditions: “when remaining image interpretation completion durations corresponding to any two pathological detection results in the priority sorting queue are greater than or equal to the first threshold, a pathological detection result hitting more critical pathologies has a higher priority”, “when the quantities of critical pathologies hit by any two pathological detection results in the priority sorting queue are the same, a pathological detection result corresponding to shorter remaining image interpretation completion duration has a higher priority”, and “in the priority sorting queue, the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is less than the first threshold is higher than the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is greater than or equal to the first threshold”.

[0123] FIG. 6 is a schematic flowchart of a method for controlling image interpretation for medical examination according to an embodiment of the present disclosure;

[0124] As shown in FIG. 6, in a possible implementation, the method further includes:

[0125] Step S61: Determining a hit situation of the pathological detection result of the current subject for an incidental pathology based on a preset incidental pathology list, wherein an urgency level of the incidental pathology is less than an urgency level of the critical pathology;

[0126] Step S62: Generating a second label of the current subject when the quantity of incidental pathologies hit by the pathological detection result of the current subject is greater than 0, wherein the second label records an incidental pathology hit by the pathological detection result of the current subject;

[0127] Step S63: Sending the second label to the picture archiving and communication system, the picture archiving and communication system being further used to display the second label, and the second label being displayed based on the same image interpretation sequence as the first label.

[0128] For example, a pathology whose urgency level is less than that of the critical pathology and that may deteriorate as time goes on may be marked as an incidental pathology. Each type of pathology can only be set as a critical pathology or an incidental pathology. TableDocket No. 701818-WO-23 shows an example of an incidental pathology list. An identifier of an incidental pathology may be U.Table 3

[0129] A person skilled in the art should understand that whether each type of pathology is an incidental pathology may be adjusted at any time based on an application scenario. Which pathologies are incidental pathologies is not limited in the present disclosure.

[0130] It should be noted that a single pathology is only regarded as one of a critical pathology or an incidental pathology, and cannot be regarded as both a critical pathology and an incidental pathology.

[0131] The hit situation of the pathological detection result of the current subject for the incidental pathology may be determined based on the preset incidental pathology list, wherein the urgency level of the incidental pathology is lower than the urgency level of the critical pathology. The hit situation of the pathological detection result of the current subject for the incidental pathology may indicate the quantity of incidental pathologies hit by the pathological detection result of the current subject and a hit incidental pathology. Optionally, the hit situation may further include a confidence level of a pathology. The pathological detection result in FIG. 3 is used as an example. Then, the pathological detection result hits two incidental pathologies.Docket No. 701818-WO-2

[0132] The second label of the current subject may be generated when the quantity of incidental pathologies hit by the pathological detection result of the current subject is greater than 0, wherein the second label records the incidental pathology hit by the pathological detection result of the current subject. The second label may be encapsulated into an HL7 message. An exemplary encapsulation manner is provided below.

[0133] The second label may also be sent to the picture archiving and communication system. The picture archiving and communication system is further used to display the second label, and the second label and the first label of the same subject are displayed based on the same image interpretation sequence. That is, the image, the first label, and the second label of the same subject are displayed based on the same image interpretation sequence.

[0134] In this way, the picture archiving and communication system can further display, to the doctor, the incidental pathology hit by the pathological detection result of the current subject, which is more conducive to accurately analyzing a condition of the current subject by the doctor.

[0135] In a possible implementation, the method further includes:

[0136] Acquiring a submission time of a report of the current subject from the picture archiving and communication system;

[0137] Determining image analysis time consumption of the current subject based on the difference between the submission time of the report of the current subject and the examination completion time of the current subject; and

[0138] Adjusting one or a plurality of a parameter of the pathological detection model, the expected image interpretation queue time consumption value, or the first threshold based on a size relationship between the image analysis time consumption and the expected image interpretation queue time consumption value.

[0139] For example, after the first label is sent to the picture archiving and communication system, it is possible to query the picture archiving and communication system at a specific frequency whether the report of the current subject is submitted. If it is found that the report of the current subject is submitted, the submission time of the report of the current subject may be acquired from the picture archiving and communication system.Docket No. 701818-WO-2

[0140] The time consumed by image analysis for the current subject may be determined based on the difference between the submission time of the report of the current subject and the examination completion time of the current subject. One or a plurality of the parameter of the pathological detection model, the expected image interpretation queue time consumption value, or the first threshold may be adjusted based on a size relationship between the image analysis time consumption and the expected image interpretation queue time consumption value. For example, the first threshold, the parameter of the pathological detection model, and the expected image interpretation queue time consumption value may remain unchanged if the image analysis time consumption is less than the expected image interpretation queue time consumption value. The first threshold may be increased and the parameter of the pathological detection model may be adjusted if the image analysis time consumption is greater than the expected image interpretation queue time consumption value, to speed up detection of the pathological detection model and increase the expected image interpretation queue time consumption value. In this case, when any subject has waited for a long time and the priority of image interpretation of the subject needs to be raised, the priority may be raised earlier, thereby reducing the time consumed by image analysis for such a subject.

[0141] A person skilled in the art should understand that the picture archiving and communication system may also proactively send the submission time of the report to the system for controlling image interpretation for medical examination. The specific manner in which the system for controlling image interpretation for medical examination acquires the submission time of the report is not limited in the embodiments of the present disclosure.

[0142] A person skilled in the art should understand that the system for controlling image interpretation for medical examination may further make more adjustments based on the submission time of the report, for example, adjust the sorting rule for the priority sorting queue. The adjustment manner that can be implemented by the system for controlling image interpretation for medical examination is not limited in the embodiments of the present disclosure.

[0143] The following describes an exemplary method for encapsulating the first label, the second label, the time label and the priority label into the HL7 message.Docket No. 701818-WO-2

[0144] For example, the first label, the second label, the time label and the priority label may be stored by using an order message (ORM) in the HL7 message. A segment in the order message ORM and information filled therein follow an HL7 protocol.

[0145] The order message ORM includes a plurality of segments. For example, a message header segment (MSH) includes an identifier and metadata information of a message. A patient identification segment (PID) is used to convey patient identification information, including other optional identification information such as a patient name, an identification (ID) number, and a birth date. A patient visit segment (PV1) includes a patient visit type, a visit number, a visit date, a visit department, and the like. A common order segment (ORC) is used to transmit a medical order document, including a medical order status, a medical order number, and the like. An observation request segment (OBR) is used to transmit specific information of diagnostic observation, examination, or evaluation, including specific content such as an examination type and an examination part. An observation result (OBX) segment is used to transmit a single observation or observation fragment result, for example, a pathological detection result.

[0146] The ORM message may be filled with the following content based on the first label, the second label, the time label, and the priority label:

[0147] (1) An identifier or examination number of a pathological detection result to be prioritized. The pathological detection result to be prioritized may refer to a pathological detection result in the first sub-queue and the second sub-queue.

[0148] (2) The priority of a pathological detection result, which may be represented in a priority-related entry in the OBR, and may include expected image interpretation completion time. A pathological detection result hitting more critical pathologies has an earlier expected image interpretation completion time.

[0149] (3) An identifier for distinguishing priority, which is used to distinguish whether a pathological detection result belongs to the first sub-queue or the second sub-queue.

[0150] (4) The pathological detection result, including a pathology, a confidence level of the pathology, and an identifier indicating whether the pathology is a critical pathology / an incidental pathology.Docket No. 701818-WO-2

[0151] A person skilled in the art should understand that content filled in the ORM message is not limited to the foregoing examples. Specific content stored in the ORM message is not limited in the present disclosure, provided that the content is content that can be stored in the ORM message.

[0152] In a possible implementation, the present disclosure further provides a picture archiving and communication system. FIG. 7 is a schematic diagram of a workflow of a picture archiving and communication system according to the present disclosure.

[0153] As shown in FIG. 7, the system is used to perform steps S71-S75.

[0154] Step S71 : Receiving an image of a current subject;

[0155] Step S72: Receiving a first label, a time label, and a priority label of the current subject, wherein the first label records a critical pathology hit by a pathological detection result of the current subject, the time label records examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in a priority sorting queue;

[0156] Step S73: Determining, based on the time label, whether a remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to a first threshold;

[0157] Step S74: When the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to the first threshold, determining a first image interpretation sequence corresponding to the current subject based on the priority label, displaying the image and the first label of the current subject based on the first image interpretation sequence, and updating an image interpretation sequence corresponding to another subject whose image interpretation sequence is lower than or equal to the first image interpretation sequence;

[0158] Step S75: When the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is less than the first threshold, determining a second image interpretation sequence corresponding to the current subject based on remaining image interpretation completion duration corresponding to each subject, displaying the image and the first label of the current subject based on the second imageDocket No. 701818-WO-2 interpretation sequence, and updating an image interpretation sequence corresponding to another subject whose image interpretation sequence is lower than or equal to the second image interpretation sequence, wherein the second image interpretation sequence is higher than the first image interpretation sequence.

[0159] For example, the picture archiving and communication system may display the first label of the current subject in a work list after receiving the first label of the current subject. Image interpretation of the current subject may be overdue or not overdue when the picture archiving and communication system receives the first label of the current subject. Therefore, the picture archiving and communication system may first determine, based on the time label, whether image interpretation of the current subject is overdue, and determine a display priority of the image and the first label of the current subject based on a determining result.

[0160] For example, the picture archiving and communication system may determine, based on the time label, whether the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to the first threshold. Image interpretation of the current subject is not overdue when the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to the first threshold, and the priority label is still valid. The first image interpretation sequence corresponding to the current subject may be determined based on the priority label, so as to display the image and the first label of the current subject based on the first image interpretation sequence and update the image interpretation sequence corresponding to the another subject whose image interpretation sequence is lower than or equal to the first image interpretation sequence. A manner of calculating the remaining image interpretation completion duration is already described above. Details are not described again here.

[0161] After the image and the first label of the current subject are displayed, examination of the current subject may not start immediately, but need to be started after waiting for a period of time. In a process of waiting, the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject may be less than the first threshold. In this case, image interpretation of the current subject is overdue, and the priority label and the determined first image interpretation sequence are invalid. In this case, the secondDocket No. 701818-WO-2 image interpretation sequence corresponding to the current subject may be determined based on the remaining image interpretation completion duration, so as to display the image and the first label of the current subject based on the second image interpretation sequence, and update the image interpretation sequence corresponding to the another subject whose image interpretation sequence is lower than or equal to the second image interpretation sequence, wherein the second image interpretation sequence is higher than the first image interpretation sequence.

[0162] FIG. 8, FIG. 9a, and FIG. 9b are each a schematic diagram of a work list of a picture archiving and communication system according to an embodiment of the present disclosure.

[0163] For example, it is assumed that the pathological detection result of the current subject is a pathological detection result 10. The picture archiving and communication system receives the priority label of the current subject at a first moment, and determines that the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to the first threshold. Therefore, for the first image interpretation sequence corresponding to the current subject that is determined based on the received priority label, refer to FIG. 8 (a first label i records a critical pathology hit by a pathological detection result i, and i is a positive integer). In this case, the priority indicated by the priority label is the same as the first image interpretation sequence. The first image interpretation sequence corresponding to the current subject may be ranked ninth, and the remaining image interpretation completion duration corresponding to the pathological detection result 10 (displayed by using a first label 10) is greater than the first threshold. Because image interpretation sequences of other subjects are higher than the first image interpretation sequence, the image interpretation sequences of other subjects do not need to be updated.

[0164] If the picture archiving and communication system receives a new first label or image interpretation of any subject is overdue, the picture archiving and communication system updates an image interpretation sequence corresponding to each subject. It is assumed that at a second moment, image interpretation of pathological detection results 11-13 has already been completed, remaining image interpretation completion durations corresponding to pathological detection results 1, 2, 7, 8, and 9 are already less than the first threshold but image interpretation is not completed. The picture archiving and communication system further receives priority labels corresponding to first labels 14 and 15 after the first moment and before the secondDocket No. 701818-WO-2 moment. At the second moment, remaining image interpretation completion durations corresponding to pathological detection results 10, 14, and 15 are greater than the first threshold. It is assumed that values of the remaining image interpretation completion durations corresponding to the pathological detection results 2, 7, 1, 8, and 9 decrease successively, and the quantities of critical pathologies hit by the pathological detection results 14, 15, and 10 decrease successively. In this case, the image interpretation sequence corresponding to each subject is shown in FIG. 9a.

[0165] It is assumed that at a third moment, the remaining image interpretation completion duration corresponding to the pathological detection result 10 is less than the first threshold, and the remaining image interpretation completion durations corresponding to the pathological detection results 14 and 15 are still greater than the first threshold. Then, it may be determined, based on the remaining image interpretation completion duration of the pathological detection result of each subject, that the image interpretation sequence corresponding to the current subject should be higher than those of subjects corresponding to the pathological detection results 14 and 15, and an image interpretation sequence of a subject corresponding to another pathological detection result is not affected. The determined second image interpretation sequence of the current subject may be ranked sixth. The image interpretation sequences of the subjects corresponding to the pathological detection results 14 and 15 whose image interpretation sequences are lower than or equal to the second image interpretation sequence are separately lowered by one position, and final image interpretation sequences are shown in FIG. 9b.

[0166] The system for controlling image interpretation for medical examination and the picture archiving and communication system re-adjust the image interpretation sequences of the pathological detection results based on a size relationship between the remaining image interpretation completion duration of the pathological detection result of each subject and the first threshold. Therefore, when no pathological detection result of a new subject is added to the priority sorting queue, an image interpretation sequence corresponding to a specific subject on the system for controlling image interpretation for medical examination may be the same as the priority of a first label of the same subject on the picture archiving and communication system.

[0167] In this case, if the priority indicated by a priority label of a new subject A that is received by the picture archiving and communication system is the same as an imageDocket No. 701818-WO-2 interpretation sequence of a specific subject B, it indicates that an image interpretation sequence of a pathological detection result of the new subject A should be higher than the image interpretation sequence of the subject B, and the image interpretation sequence of the subject B and an image interpretation sequence of another subject whose image interpretation sequence is lower than that of the subject B are each lowered by one position.

[0168] In this way, for the picture archiving and communication system, a subject has sufficient remaining image interpretation completion duration when a priority label is received. However, if image interpretation is not started yet when the remaining duration for completing image interpretation is insufficient, the subject may obtain a higher image interpretation sequence, thereby reducing the time consumed by image interpretation for the subject for which the remaining duration for completing image interpretation is insufficient, shortening waiting duration of such a subject, and improving the examination experience of a subject to be examined.

[0169] FIG. 10 is a schematic diagram of a workflow of a picture archiving and communication system according to the present disclosure.

[0170] As shown in FIG. 10, in a possible implementation, the system is further configured to perform steps S101-S102.

[0171] Step S101 : Receiving a second label of the current subject, wherein the second label records an incidental pathology hit by the pathological detection result of the current subject;

[0172] Step S102: Displaying the second label based on the same image interpretation sequence as the first label.

[0173] For example, when the picture archiving and communication system further receives the second label of the current subject, the picture archiving and communication system also displays the second label of the current subject. The second label records the incidental pathology hit by the pathological detection result of the current subject. The second label and the first label of the current subject are displayed based on the same image interpretation sequence. In this case, a doctor may check the image, the first label, and the second label of the subject simultaneously.Docket No. 701818-WO-2

[0174] According to the method for controlling image interpretation for medical examination in the embodiments of the present disclosure, the critical pathology may be intelligently set, and the priority may be determined based on the hit situation of the pathological detection result for the critical pathology, thereby improving the efficiency of a radiological imaging examination and ensuring timely processing of an urgent case, reducing risks caused by delayed clinical intervention, and greatly reducing image interpretation time consumption of a subject with an urgent condition.

[0175] According to the method for controlling image interpretation for medical examination in the embodiments of the present disclosure, the hit situation of the pathological detection result for the incidental pathology is also transmitted to the picture archiving and communication system, to supplementarily remind a doctor to provide timely medical care to a patient and make a medical record of the patient more detailed In addition, the probability of some asymptomatic and thus easily overlooked time-sensitive pathologies being noticed also increases, and the pathologies can be treated earlier before conditions deteriorate over time.

[0176] The present disclosure further provides a system for controlling image interpretation for medical examination. FIG. 11 is a schematic diagram of a structure of a system for controlling image interpretation for medical examination according to an embodiment of the present disclosure.

[0177] In a possible implementation, as shown in FIG. 11, the system includes:

[0178] A pathological detection module 110, used to input an image of a current subject into a pathological detection model, so that the pathological detection model outputs a pathological detection result of the current subject;

[0179] A first determining module 111, used to determine a hit situation of the pathological detection result of the current subject for a critical pathology based on a preset critical pathology list;

[0180] A sorting module 112, used to add the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject, wherein when remaining image interpretation completion duration corresponding to any twoDocket No. 701818-WO-2 pathological detection results in the priority sorting queue is greater than or equal to a first threshold, a pathological detection result hitting more critical pathologies has a higher priority;

[0181] A first generation module 113, used to generate a first label, a time label, and a priority label of the current subject, wherein the first label records a critical pathology hit by the pathological detection result of the current subject, the time label records examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in the priority sorting queue; and

[0182] A first sending module 114, used to send the first label, the time label, and the priority label to a picture archiving and communication system, so that the system determines an image interpretation sequence corresponding to the current subject based on the priority label and the time label, displays the image and the first label of the current subject based on the image interpretation sequence, and updates an image interpretation sequence corresponding to another subject.

[0183] In a possible implementation, the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is equal to the difference between expected image interpretation completion time of the current subject and current time, and the expected image interpretation completion time of the current subject is equal to the sum of the examination completion time of the current subject and an expected image interpretation queue time consumption value.

[0184] In a possible implementation, when the quantities of critical pathologies hit by any two pathological detection results in the priority sorting queue are the same, a pathological detection result corresponding to a shorter remaining image interpretation completion duration has a higher priority.

[0185] In a possible implementation, in the priority sorting queue, the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is less than the first threshold is higher than the priority of a pathological detection result whose corresponding remaining image interpretation completion duration is greater than or equal to the first threshold.Docket No. 701818-WO-2

[0186] In a possible implementation, the priority sorting queue includes a first sub-queue, a second sub-queue and a third sub-queue, the priority of the first sub-queue is higher than the priority of the second sub-queue, and the priority of the second sub-queue is higher than the priority of the third sub-queue.

[0187] Said adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject includes:

[0188] Adding the pathological detection result of the current subject to the second subqueue when the quantity of critical pathologies hit by the pathological detection result of the current subject is greater than 0, wherein in the second sub-queue, a pathological detection result hitting more critical pathologies critical pathologies has a higher priority, and when the quantities of critical pathologies hit by pathological detection results are the same, a pathological detection result corresponding to an earlier examination completion time has a higher priority;

[0189] Adding the pathological detection result of the current subject to the third sub-queue when the quantity of critical pathologies hit by the pathological detection result of the current subject is equal to 0, wherein in the third sub-queue, a pathological detection result corresponding to an earlier examination completion time has a higher priority; and

[0190] Monitoring remaining image interpretation completion durations corresponding to pathological detection results in the second sub-queue and the third sub-queue, and when a remaining image interpretation completion duration corresponding to any pathological detection result is less than the first threshold, moving the pathological detection result from a current subqueue to the first sub-queue, wherein in the first sub-queue, a pathological detection result corresponding to a smaller remaining image interpretation completion duration value has a higher priority.

[0191] In a possible implementation, the system further includes:

[0192] A second determining module, used to determine a hit situation of the pathological detection result of the current subject for an incidental pathology based on a preset incidental pathology list, wherein an urgency level of the incidental pathology is less than an urgency level of the critical pathology;Docket No. 701818-WO-2

[0193] A second generation module, used to generate a second label of the current subject when the quantity of incidental pathologies hit by the pathological detection result of the current subject is greater than 0, wherein the second label records an incidental pathology hit by the pathological detection result of the current subject; and

[0194] A second sending module, used to send the second label to the picture archiving and communication system, the picture archiving and communication system being further used to display the second label, and the second label being displayed based on the same image interpretation sequence as the first label.

[0195] In a possible implementation, the system further includes:

[0196] An acquisition module, used to acquire a submission time of a report of the current subject from the picture archiving and communication system;

[0197] A third determining module, used to determine image analysis time consumption of the current subject based on the difference between the submission time of the report of the current subject and the examination completion time of the current subject; and

[0198] An adjustment module, used to adjust one or a plurality of a parameter of the pathological detection model, the expected image interpretation queue time consumption value, or the first threshold based on a size relationship between the image analysis time consumption and the expected image interpretation queue time consumption value.

[0199] In some embodiments, the functions or modules of the system for controlling image interpretation for medical examination provided in the embodiments of the present disclosure may be configured to perform the method described in the foregoing method embodiments. For specific implementations thereof, refer to the description of the foregoing method embodiments. Details are not described again here, for brevity.

[0200] An embodiment of the present disclosure further provides a computer-readable storage medium, having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the above method. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium.Docket No. 701818-WO-2

[0201] An embodiment of the present disclosure further provides an electronic device, including: a processor; and a memory, used to store instructions executable by the processor, wherein the processor is configured to implement the above method when executing the instructions stored in the memory.

[0202] An embodiment of the present disclosure further provides a computer program product, including computer-readable code or a non-volatile computer-readable storage medium bearing computer-readable code, wherein when the computer-readable code is run in a processor in an electronic device, the processor in the electronic device performs the above method.

[0203] FIG. 12 is a block diagram of an apparatus 1900 according to an embodiment of the present disclosure. For example, the apparatus 1900 may be an apparatus for controlling image interpretation for medical examination, and is provided as a server or a terminal device.Referring to FIG. 12, the apparatus 1900 includes a processing assembly 1922 that further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions executable by the processing assembly 1922, for example, an application. The application stored in the memory 1932 can include one or more modules each corresponding to a set of instructions. In addition, the processing assembly 1922 is configured to execute instructions to perform the above method.

[0204] The apparatus 1900 may further include a power supply assembly 1926 configured to perform power management of the apparatus 1900, a wired or wireless network interface 1950 configured to connect the apparatus 1900 to a network, and an input / output interface (VO interface) 1958. The apparatus 1900 may operate based on an operating system stored in the memory 1932, for example, Windows Server™, Mac OS X™, Unix™, Linux™, or FreeBSD™.

[0205] In an exemplary embodiment, further provided is a non-volatile computer-readable storage medium, for example, the memory 1932 including computer program instructions that are executable by the processing assembly 1922 of the apparatus 1900 to complete the above method.

[0206] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium, having computer-readable program instructions thereon for causing a processor to implement various aspects of the present disclosure.Docket No. 701818-WO-2

[0207] The computer-readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. The computer-readable storage medium may be, for example, but is not limited to an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor memory device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of the computer-readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read only memory (ROM), a erasable programmable read only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, mechanical coding equipment, such as a punch card with instructions stored thereon or a structure of bumps within recessions, and any suitable combination thereof. The computer- readable storage medium used herein is not interpreted as transient signals themselves, such as radio waves or other freely propagated electromagnetic waves, electromagnetic waves propagated through a waveguide or other transmission media (e.g., light pulses passing through a fiber optic cable), or electrical signals transmitted through electric wires.

[0208] The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to various computing / processing devices or downloaded to an external computer or external storage device via a network such as the Internet, a local area network, a wide area network and / or a wireless network. The network may include copper transmission cables, fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or a network interface in each computing / processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions, for storing them in a computer-readable storage medium in each computing / processing device.

[0209] Computer program instructions for performing the operations of the present disclosure can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, the programming language including object oriented programming languages such as Smalltalk, C++ and the like, and conventional procedural programming languages such as the “C” language or similar programming languages. The computer-readable program instructions can beDocket No. 701818-WO-2 executed entirely or partly on a user computer, executed as a stand-alone software package, executed partly on a user computer and partly on a remote computer, or executed entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to a user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or connected to an external computer (for example, through the Internet, using an Internet service provider). In some embodiments, an electronic circuit, for example, a programmable logic circuit, a field-programmable gate array (FPGA), or a programmable logic array (PLA), may execute the computer-readable program instructions by utilizing state information of the computer-readable program instructions to personalize the electronic circuit, in order to implement various aspects of the present disclosure

[0210] The aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, apparatuses (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams and combinations of various blocks in the flowcharts and / or block diagrams can be implemented by computer-readable program instructions.

[0211] These computer-readable program instructions may be provided to a processor of a general -purpose computer, a special-purpose computer, or other programmable data processing apparatuses, to produce a machine, so that these instructions, when executed by the processor of the computer or other programmable data processing apparatuses, produce an apparatus for implementing the functions / actions specified in one or more blocks of the flowcharts and / or block diagrams. Also, these computer-readable program instructions may be stored in a computer-readable storage medium. These instructions cause a computer, a programmable data processing device, and / or other devices to work in a specific manner. Thus, the computer- readable medium storing the instructions includes an artifact, including instructions that implement various aspects of the functions / actions specified in one or more the flowcharts and / or block diagrams.

[0212] The computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatuses, or other devices, such that the computer, other programmable data processing apparatuses or other devices perform a series of operational steps,Docket No. 701818-WO-2 to generate a computer-implemented process, such that the functions / actions specified in one or more of the flowcharts and / or block diagrams are implemented by the instructions executed on the computer, other programmable data processing apparatuses, or other devices.

[0213] The flowcharts and block diagrams in the accompanying drawings illustrate system architectures, functions, and operations of possible implementations of the system, method, and computer program product according to a plurality of embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams may represent a portion of a module, program segment, or instruction that contains one or more executable instructions for implementing the specified logical functions. In some alternative implementations, the functions denoted in the blocks can also occur in a different order than that illustrated in the drawings. For example, two consecutive blocks can actually be performed substantially in parallel, and sometimes can also be performed in a reverse order, depending upon the functions involved. It should also be noted that each block of the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts can be implemented in a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0214] The embodiments of the present disclosure have been described above. The foregoing description is illustrative rather than limiting, and is not limited to the disclosed embodiments. Many modifications and variations are apparent to those of ordinary skill in the art without departing from the scope and spirit of the embodiments illustrated. The selection of terms used herein is intended to best explain the principles, practical applications, or improvements to the technologies in the market of the embodiments, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

Docket No. 701818-WO-2CLAIMS1. A method for controlling image interpretation for medical examination, characterized by comprising: inputting an image of a current subject into a pathological detection model, the pathological detection model outputting a pathological detection result of the current subject; determining a hit situation of the pathological detection result of the current subject for a critical pathology based on a preset critical pathology list; adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and a remaining image interpretation completion duration corresponding to the pathological detection result of the current subject, wherein when remaining image interpretation completion durations corresponding to any two pathological detection results in the priority sorting queue are greater than or equal to a first threshold, a pathological detection result hitting more critical pathologies has a higher priority; generating a first label, a time label, and a priority label of the current subject, wherein the first label records a critical pathology hit by the pathological detection result of the current subject, the time label records examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in the priority sorting queue; and sending the first label, the time label, and the priority label to a picture archiving and communication system, the system determining an image interpretation sequence corresponding to the current subject based on the priority label and the time label, displaying the image and the first label of the current subject based on the image interpretation sequence, and updating an image interpretation sequence corresponding to another subject.

2. The method according to claim 1, wherein the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is equal to a difference between an expected image interpretation completion time of the current subject and a current time, and the expected image interpretation completion time of the current subject is equal to a sum of the examination completion time of the current subject and an expected image interpretation queue time consumption value.Docket No. 701818-WO-23. The method according to claim 1, wherein when quantities of key pathologies hit by any two pathological detection results in the priority sorting queue are the same, a pathological detection result corresponding to shorter remaining image interpretation completion duration has a higher priority.

4. The method according to claim 1, wherein in the priority sorting queue, a priority of a pathological detection result whose corresponding remaining image interpretation completion duration is less than the first threshold is higher than a priority of a pathological detection result whose corresponding remaining image interpretation completion duration is greater than or equal to the first threshold.

5. The method according to claim 1, wherein the priority sorting queue comprises a first sub-queue, a second sub-queue, and a third sub-queue, a priority of the first sub-queue is higher than a priority of the second sub-queue, and the priority of the second sub-queue is higher than a priority of the third sub-queue.

6. The method according to claim 5, wherein said adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject includes: adding the pathological detection result of the current subject to the second sub-queue when the quantity of critical pathologies hit by the pathological detection result of the current subject is greater than 0, wherein in the second sub-queue, a pathological detection result hitting more critical pathologies critical pathologies has a higher priority, and when the quantities of critical pathologies hit by pathological detection results are the same, a pathological detection result corresponding to an earlier examination completion time has a higher priority.

7. The method according to claim 6, wherein said adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject includes:Docket No. 701818-WO-2 adding the pathological detection result of the current subject to the third sub-queue when the quantity of critical pathologies hit by the pathological detection result of the current subject is equal to 0, wherein in the third sub-queue, a pathological detection result corresponding to an earlier examination completion time has a higher priority.

8. The method according to claim 7, wherein said adding the pathological detection result of the current subject to a priority sorting queue based on the hit situation and remaining image interpretation completion duration corresponding to the pathological detection result of the current subject includes: monitoring remaining image interpretation completion durations corresponding to pathological detection results in the second sub-queue and the third sub-queue, and when a remaining image interpretation completion duration corresponding to any pathological detection result is less than the first threshold, moving the pathological detection result from a current subqueue to the first sub-queue, wherein in the first sub-queue, a pathological detection result corresponding to a smaller remaining image interpretation completion duration value has a higher priority.

9. The method according to claim 1, further comprising: determining a hit situation of the pathological detection result of the current subject for an incidental pathology based on a preset incidental pathology list, wherein an urgency level of the incidental pathology is less than an urgency level of the critical pathology; generating a second label of the current subject when the quantity of incidental pathologies hit by the pathological detection result of the current subject is greater than 0, wherein the second label records an incidental pathology hit by the pathological detection result of the current subject; and sending the second label to the picture archiving and communication system, the picture archiving and communication system being further configured to display the second label, and the second label being displayed based on the same image interpretation sequence as the first label.Docket No. 701818-WO-210. The method according to claim 2, further comprising acquiring a submission time of a report of the current subject from the picture archiving and communication system.

11. The method according to claim 10, further comprising determining image analysis time consumption of the current subject based on the difference between the submission time of the report of the current subject and the examination completion time of the current subject; and12. The method according to claim 11, further comprising based on a size relationship between the image analysis time consumption and the expected image interpretation queue time consumption value adjusting one or a plurality of a parameter of the pathological detection model, the expected image interpretation queue time consumption value, or the first threshold.

13. A picture archiving and communication system, characterized in that the system is configured to: receive an image of a current subject; receive a first label, a time label, and a priority label of the current subject, wherein the first label records a critical pathology hit by a pathological detection result of the current subject, the time label records examination completion time of the current subject, and the priority label records the priority of the pathological detection result of the current subject in a priority sorting queue; determine, based on the time label, whether a remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to a first threshold; when the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is greater than or equal to the first threshold, determine a first image interpretation sequence corresponding to the current subject based on the priority label, display the image and the first label of the current subject based on the first image interpretation sequence, and update an image interpretation sequence corresponding to another subject whose image interpretation sequence is lower than or equal to the first image interpretation sequence; andDocket No. 701818-WO-2 when the remaining image interpretation completion duration corresponding to the pathological detection result of the current subject is less than the first threshold, determine a second image interpretation sequence corresponding to the current subject based on remaining image interpretation completion duration corresponding to each subject, display the image and the first label of the current subject based on the second image interpretation sequence, and update an image interpretation sequence corresponding to another subject whose image interpretation sequence is lower than or equal to the second image interpretation sequence, wherein the second image interpretation sequence is higher than the first image interpretation sequence.

14. The system according to claim 13, further configured to receive a second label of the current subject, wherein the second label records an incidental pathology hit by a pathological detection result of the current subject.

15. The system according to claim 14, further configured to display the second label based on the same image interpretation sequence as the first label.

Images