Method and device for predicting operation duration and operation progress duration, and electronic equipment

Abstract

The invention provides an operation duration and operation progress duration prediction method and device and electronic equipment, and the method comprises the steps: extracting a target influence factor from operation application information, so as to determine the predicted operation duration of a target operation; and obtaining a first prediction duration, a second prediction duration, a third prediction duration, a fourth prediction duration and a fifth prediction duration, and adding the first prediction duration, the second prediction duration, the third prediction duration, the fourth prediction duration and the prediction operation duration to obtain the operation process prediction duration. According to the mode, the predicted operation duration can be determined based on the target influence factors extracted from the operation application information, and then the predicted operation process duration is determined by combining the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration; the accuracy of predicting the operation duration and the operation progress duration is improved, so that the reasonability of operation arrangement can be effectively ensured, the waste of operation resources is reduced, and the operation safety is also facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method, device and electronic device for predicting the duration of an operation and the duration of the surgical process. Background Art

[0002] Predicting the duration of an operation is of great significance for reasonably arranging and flexibly scheduling surgical resources and scientifically evaluating the capabilities of the operating surgeon. Since the duration of an operation is affected by various factors, it is difficult to predict and the prediction accuracy is poor. At present, most hospitals require the operating surgeon to provide the estimated operation duration when submitting a surgical application. However, in order to reserve sufficient surgical time for themselves, the operating surgeon often fills in a duration much longer than the actual duration. Therefore, the waste of surgical resources is still inevitable during surgical scheduling.

[0003] In addition, there is also a situation where the doctor fails to consider the influencing factors comprehensively, resulting in an underestimated estimated operation duration, while the actual operation duration far exceeds the estimated duration. Clinically, operations lasting more than a certain duration are prone to risks such as pressure sores, infections, thrombosis, and complications. If no pre-plan is made for longer-duration operations in advance, measures cannot be taken in a timely manner when risks occur, which is not conducive to surgical safety.

[0004] Generally, the duration of past similar operations (surgical procedures) is used as a simple mathematical average as the predicted duration. This method cannot reflect the personalized and accurate prediction of the operation and cannot identify potential factors affecting the operation duration. Therefore, the prediction deviation is relatively large and it is difficult to be directly applied. Therefore, the prior art cannot achieve accurate prediction of the operation duration, and thus it is difficult to achieve accurate prediction of the surgical process, which is difficult to ensure the rationality of surgical arrangements, wastes surgical resources relatively, and is not conducive to surgical safety. Summary of the Invention

[0005] The purpose of the present invention is to provide a method, device and electronic device for predicting the duration of an operation and the duration of the surgical process, so as to effectively ensure the rationality of surgical arrangements, reduce the waste of surgical resources, and be conducive to surgical safety.

[0006] A method for predicting the duration of a surgical operation and the duration of the surgical process provided by the present invention is applied to a server. The method includes: obtaining the surgical application information corresponding to the target operation; extracting the target influencing factors from the surgical application information; wherein, the target influencing factors include: the name of the operation, the age of the patient, the gender of the patient, the weight of the patient, the data of the severity of the patient's disease, the information of the surgeon in charge of the operation, and the data of the patient's underlying diseases; based on the target influencing factors, determining the predicted surgical duration corresponding to the target operation; wherein, the predicted surgical duration represents the predicted value of the duration required from the start to the end of the operation; obtaining the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration; wherein, the first predicted duration is used to indicate the predicted value of the duration required from the patient entering the operating room to the start of anesthesia; the second predicted duration is used to indicate the predicted value of the duration required from the start of anesthesia to the start of the operation; the third predicted duration is used to indicate the predicted value of the duration required from the end of the operation to the patient leaving the operating room; the fourth predicted duration is used to indicate the predicted value of the cleaning duration of the operating room after the operation; the fifth predicted duration is used to indicate the predicted value of the duration required from the completion of the cleaning of the operating room to the next patient entering the operating room; adding up the predicted surgical duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration to obtain the predicted surgical process duration.

[0007] Further, the step of determining the predicted surgical duration corresponding to the target operation based on the target influencing factors includes: according to the name of the operation, obtaining a first parameter that matches the name of the operation; wherein, the first parameter is the first historical average surgical duration or the first median surgical duration determined according to the preset historical data; according to the name of the operation and the information of the surgeon in charge of the operation, obtaining a second parameter that matches the name of the operation and the information of the surgeon in charge of the operation; wherein, the second parameter is the second historical average surgical duration or the second median surgical duration of the type of operation performed by the first surgeon in charge of the operation determined according to the historical data; the first surgeon in charge of the operation is the surgeon in charge of the operation corresponding to the information of the surgeon in charge of the operation; obtaining the weight coefficient corresponding to each of the other target influencing factors except the name of the operation and the information of the surgeon in charge of the operation, and for each of the other target influencing factors, calculating the product of the basic adjustment value of the target influencing factor and the weight coefficient corresponding to the target influencing factor to obtain the product result; adding up the target parameter and each product result to obtain the predicted surgical duration corresponding to the target operation; wherein, the target parameter is the first parameter or the second parameter.

[0008] Further, a plurality of preset influencing factors are pre-stored in the server, each preset influencing factor corresponds to a plurality of classification categories, and each classification category has a corresponding adjustment coefficient; the steps of obtaining the weight coefficient corresponding to each target influencing factor other than the operation name and the information of the surgeon in charge of the operation include: for each target influencing factor other than the operation name and the information of the surgeon in charge of the operation, obtaining the factor value of the target influencing factor; according to the factor value, determining the target classification category corresponding to the target influencing factor; obtaining the adjustment coefficient corresponding to the target classification category, and determining the adjustment coefficient corresponding to the target classification category as the weight coefficient corresponding to the target influencing factor.

[0009] Further, the adjustment coefficient corresponding to each classification category is determined in advance by the following method: obtaining the operation application sample information corresponding to the sample operation; extracting the sample influencing factors from the operation application sample information; wherein, the sample influencing factors include: sample operation name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, sample surgeon in charge of the operation information, sample patient's underlying disease data; according to the sample operation name, obtaining a first sample parameter matching the sample operation name; wherein, the first sample parameter is the first sample historical average operation duration or the first sample operation duration median determined according to the preset historical data; according to the sample operation name and the information of the surgeon in charge of the sample operation, obtaining a second sample parameter; wherein, the second sample parameter is the second sample historical average operation duration or the second sample operation duration median determined according to the historical data; obtaining the sample factor value corresponding to each sample influencing factor other than the sample operation name and the information of the surgeon in charge of the operation, and the initial sample coefficient corresponding to each sample factor value; for each other sample influencing factor, calculating the product of the sample basic adjustment value of the sample influencing factor and the initial sample coefficient corresponding to the sample influencing factor to obtain a sample product result, and adding the target sample parameter and each sample product result to obtain the predicted sample operation duration corresponding to the sample operation; wherein, the target sample parameter is the first sample parameter or the second sample parameter;

[0010] Obtain the actual sample operation duration corresponding to the sample operation; calculate the difference between the actual sample operation duration and the predicted sample operation duration to obtain a difference result; if the difference result does not meet the preset first error range, for each other sample influencing factor, compare the actual sample operation duration with the historical duration data corresponding to the classification category to which the sample factor value of this sample influencing factor belongs to obtain a comparison result; wherein, the historical duration data is: the sample historical average operation duration or the median of the sample operation duration corresponding to the classification category to which the sample factor value belongs; if the comparison result does not meet the preset second error range, obtain a coefficient adjustment instruction for this classification category, and adjust the adjustment coefficient of this classification category according to the coefficient adjustment instruction, and repeat the step of obtaining the operation application sample information corresponding to the sample operation until the comparison result meets the preset second error range.

[0011] Further, the step of determining the predicted operation duration corresponding to the target operation based on the target influencing factor includes: inputting the factor value of each target influencing factor into a pre-trained operation duration prediction model, so as to output the predicted operation duration corresponding to the target operation through the operation duration prediction model.

[0012] Further, the operation duration prediction model is trained in the following way: obtain the operation application sample information corresponding to the sample operation and the actual operation duration corresponding to the sample operation; extract the sample factor values of the sample influencing factors from the operation application sample information; wherein, the sample influencing factors include: sample operation name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, sample operation surgeon information, sample patient underlying disease data; input each sample factor value into the initial model to output the predicted operation duration corresponding to the sample operation through the initial model; based on the predicted operation duration and the actual operation duration, calculate the loss value, update the parameters of the initial model based on the loss value, and repeat the steps of obtaining the operation application sample information corresponding to the sample operation and the actual operation duration corresponding to the sample operation until the loss value converges to obtain the trained operation duration prediction model.

[0013] Further, the target operation includes multiple sub-operations; the step of obtaining the operation application information corresponding to the target operation includes: for each sub-operation in the target operation, obtain the operation application information corresponding to this sub-operation; the step of determining the predicted operation duration corresponding to the target operation based on the target influencing factor includes: for each sub-operation, determine the predicted sub-operation duration corresponding to this sub-operation based on the target influencing factor; obtain the weight coefficient corresponding to each sub-operation; based on each weight coefficient, perform weighted calculation on each predicted sub-operation duration to obtain the predicted operation duration corresponding to the target operation.

[0014] Further, the specified prediction duration is: the first prediction duration, the third prediction duration, or the fifth prediction duration; the specified prediction duration is determined in advance in the following manner: obtaining multiple time values corresponding to the specified prediction duration within a preset historical time period; extracting the maximum value and the minimum value from the multiple time values; if the difference between the maximum value and the minimum value is less than a preset difference threshold, calculating the average value of the multiple time values, and taking the average value as the specified prediction duration.

[0015] Further, the fourth prediction duration is determined in advance in the following manner: determining the infection classification to which the target surgery belongs; obtaining the historical cleaning durations corresponding to multiple historical surgeries belonging to the infection classification from historical data; calculating a first calculation result of the multiple historical cleaning durations; wherein, the first calculation result is the median or the mean; taking the first calculation result as the fourth prediction duration.

[0016] Further, the second prediction duration is determined in advance in the following manner: extracting a first influencing factor from the surgery application information; wherein, the first influencing factor includes: patient age, anesthesia method, preoperative risk assessment information, patient's underlying diseases; determining the second prediction duration based on the first influencing factor.

[0017] Further, the method further includes: displaying the predicted surgery duration, the first prediction duration, the second prediction duration, the third prediction duration, the fourth prediction duration, and the fifth prediction duration; if an update instruction for the predicted surgery duration is received, updating the predicted surgery duration according to the update instruction; if an adjustment instruction for the target prediction duration is received, adjusting the target prediction duration according to the adjustment instruction; wherein, the target prediction duration includes at least one of the following: the first prediction duration, the second prediction duration, the third prediction duration, the fourth prediction duration, and the fifth prediction duration.

[0018] Further, the method further includes: displaying the confidence level for the target duration information; wherein, the target duration information includes: the predicted surgery duration, the predicted surgery progress duration.

[0019] Further, the confidence level is calculated by the following formula:

[0020] α = 1 - |(target duration information - reference value) / reference value| × 100%;

[0021] Wherein, when the number of historical surgery samples corresponding to the operating surgeon in the operating surgeon information meets a preset quantity threshold, determining the reference value based on the average value or the median of the durations of multiple historical surgery samples of the operating surgeon; when the number of historical surgery samples corresponding to the operating surgeon in the operating surgeon information does not meet the preset quantity threshold, determining the reference value based on the average value or the median of the durations of multiple historical surgery samples of other operating surgeons with the same professional title as the operating surgeon.

[0022] Further, the method further includes: if a first instruction indicating that an abnormal event has occurred is received, extracting first surgical information associated with the surgical name from preset historical data; extracting duration adjustment information associated with the abnormal event from the first surgical information; and adjusting the target duration information according to the duration adjustment information.

[0023] A prediction device for surgical duration and surgical process duration provided by the present invention is disposed on a server. The device includes: a first acquisition module for acquiring surgical application information corresponding to a target surgery; an extraction module for extracting target influencing factors from the surgical application information, where the target influencing factors include: surgical name, patient age, patient gender, patient weight, patient disease severity data, information of the surgeon in charge of the surgery, and patient underlying disease data; a determination module for determining a predicted surgical duration corresponding to the target surgery based on the target influencing factors, where the predicted surgical duration represents an estimated value of the duration required from the start to the end of the surgery; a second acquisition module for acquiring a first predicted duration, a second predicted duration, a third predicted duration, a fourth predicted duration, and a fifth predicted duration, where the first predicted duration is used to indicate an estimated value of the duration required from the patient entering the operating room to the start of anesthesia; the second predicted duration is used to indicate an estimated value of the duration required from the start of anesthesia to the start of the surgery; the third predicted duration is used to indicate an estimated value of the duration required from the end of the surgery to the patient leaving the operating room; the fourth predicted duration is used to indicate an estimated value of the cleaning duration of the operating room after the surgery; the fifth predicted duration is used to indicate an estimated value of the duration required from the completion of the cleaning of the operating room to the next surgical patient entering the operating room; and a processing module for performing an addition process on the predicted surgical duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to obtain a predicted surgical process duration.

[0024] An electronic device provided by the present invention includes a processor and a memory. The memory stores machine-executable instructions that can be executed by the processor, and the processor executes the machine-executable instructions to implement the prediction method for surgical duration and surgical process duration in any one of the above.

[0025] A machine-readable storage medium provided by the present invention stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions cause the processor to implement the prediction method for surgical duration and surgical process duration in any one of the above.

[0026] The prediction method, device and electronic device for the duration of a surgery and the duration of the surgical process provided by the present invention obtain the surgical application information corresponding to the target surgery; extract the target influencing factors from the surgical application information; wherein, the target influencing factors include: the name of the surgery, the age of the patient, the gender of the patient, the weight of the patient, the data of the severity of the patient's disease, the information of the surgeon in charge of the surgery, and the data of the patient's underlying diseases; based on the target influencing factors, determine the predicted duration of the target surgery; wherein, the predicted duration of the surgery represents the predicted value of the duration required from the start to the end of the surgery; obtain the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration; wherein, the first predicted duration is used to indicate the predicted value of the duration required from the patient entering the operating room to the start of anesthesia; the second predicted duration is used to indicate the predicted value of the duration required from the start of anesthesia to the start of the surgery; the third predicted duration is used to indicate the predicted value of the duration required from the end of the surgery to the patient leaving the operating room; the fourth predicted duration is used to indicate the predicted value of the cleaning duration of the operating room after the surgery; the fifth predicted duration is used to indicate the predicted value of the duration required from the completion of the cleaning of the operating room to the next surgical patient entering the operating room; perform an addition process on the predicted duration of the surgery, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration to obtain the predicted duration of the surgical process. This method can determine the predicted duration of the surgery based on the target influencing factors extracted from the surgical application information, and then combine the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration to determine the predicted duration of the surgical process; improving the accuracy of predicting the duration of the surgery and the duration of the surgical process, and thus can effectively ensure the rationality of surgical arrangements, reduce the waste of surgical resources, and is also beneficial to surgical safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0028] Figure 1 It is a flowchart of a prediction method for the duration of a surgery and the duration of the surgical process provided by an embodiment of the present invention;

[0029] Figure 2 It is a schematic diagram of the basic process of iterative improvement of a surgical duration prediction model provided by an embodiment of the present invention;

[0030] Figure 3 It is a flowchart of a surgical duration prediction algorithm provided by an embodiment of the present invention;

[0031] Figure 4 A structural schematic diagram of a prediction device for the duration of an operation and the duration of the operation process provided by an embodiment of the present invention;

[0032] Figure 5 A structural schematic diagram of an electronic device provided by an embodiment of the present invention. Specific embodiments

[0033] Next, the technical solutions of the present invention will be clearly and completely described in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0034] Currently, the duration of an operation is generally defined as: the end time of the operation - the start time of the operation. Predicting the duration of an operation is of great significance for reasonably arranging and flexibly scheduling operation resources and scientifically evaluating the capabilities of the operating surgeon. Since the duration of an operation is affected by various factors, including the name of the operation (different operation names correspond to different specialties and different complexities), the health status of the patient, the experience of the operating surgeon, and various unforeseeable factors during the operation process, it is difficult to predict and the prediction accuracy is relatively poor. Currently, there are few domestic cases of operation duration prediction research applied to engineering practice. Therefore, most hospitals require the operating surgeon to provide the estimated operation duration when submitting an operation application. However, in order to reserve sufficient operation time for themselves, the operating surgeon often fills in a duration much longer than the actual duration. Therefore, the phenomenon of waste of operation resources is still inevitable during operation scheduling.

[0035] There are also operating surgeons who, due to insufficient recognition of the influencing factors of the operation duration and underestimation of the influence degree, result in an underestimated operation duration and an incomplete risk plan. Because a long-duration operation will bring risks such as pressure sores, infections, thrombosis, and complications, accurately predicting the operation duration is also an important measure to ensure the safety and quality of the operation.

[0036] In addition, by evaluating the difference between the actual operation duration and the predicted operation duration, the operation treatment ability and operation quality of the operating surgeon can also be evaluated. When the credibility of the predicted value is high, if the difference between the operation duration and the predicted duration is too large, special attention can be paid to this operation to evaluate whether there is an abnormal operation duration caused by the ability level of the operation personnel or the operation quality.

[0037] Based on this, the embodiments of the present invention provide a method, device, and electronic device for predicting the duration of an operation and the duration of the operation process. This technology can be applied to applications that require accurate prediction of the duration of an operation and the duration of the operation process.

[0038] To facilitate the understanding of this embodiment, first, a method for predicting the duration of a surgical operation and the duration of the surgical process disclosed in the embodiments of the present invention will be introduced. This method is applied to a server, such as Figure 1 As shown, the method includes the following steps:

[0039] Step S102, obtain the surgical application information corresponding to the target surgery;

[0040] The above-mentioned target surgery can be a surgery to be performed. For example, the target surgery is the surgical information applied by the surgery department to the operating room, which can be surgeries in different surgical departments such as ophthalmic surgery and orthopedic surgery; the above-mentioned surgical application information can be the relevant information of the target surgery submitted by the surgeon in charge when applying for the target surgery. The surgical application information usually includes: surgical name, patient age, patient gender, patient weight, diagnosis, requirements for surgical resources, the estimated surgical duration by the surgeon in charge, and other information. In actual implementation, when predicting the duration of the target surgery, it is usually necessary to first obtain the surgical application information of the target surgery.

[0041] Step S104, extract the target influencing factors from the surgical application information; among them, the target influencing factors include: surgical name, patient age, patient gender, patient weight, patient disease severity data, surgeon in charge information, and patient underlying disease data;

[0042] The above-mentioned surgical name can be understood as the specific surgical procedure of the target surgery, which can be a single surgical name or a combination of multiple surgical names, and can be represented by a surgical code. The surgical code and surgical name in the "National Clinical Version 3.0 Surgical Operation Coding" or the "Grouping Scheme for Paying by Disease Group and Disease Type 2.0 Version" can be referred to; the above-mentioned patient disease severity data can be words, grades, etc. that can represent the disease severity; for example, it can be severe, mild, or clinical scores, evaluations, classifications, etc.; the above-mentioned surgeon in charge information usually includes the name, qualifications, experience level, etc. of the surgeon in charge; the above-mentioned patient underlying disease data can be understood as one or more chronic diseases that the patient already has and may affect the health status; in actual implementation, when the surgical application information is obtained, the target influencing factors that affect the surgical duration, such as surgical name, patient age, patient gender, patient weight, patient disease severity data, surgeon in charge information, and patient underlying disease data, can be extracted from it.

[0043] Step S106, based on the target influencing factors, determine the predicted surgical duration corresponding to the target surgery; among them, the predicted surgical duration represents the estimated value of the duration required from the start to the end of the surgery;

[0044] The above predicted operation duration can also be understood as the predicted value of the difference between the operation end time and the operation start time, and the predicted value of the duration of the operation procedure; the predicted operation duration can be expressed by the following expression: Predicted operation duration = Operation end time - Operation start time.

[0045] Step S108, obtain the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration; wherein, the first predicted duration is used to indicate the predicted value of the duration required from the patient entering the operating room to the start of anesthesia; the second predicted duration is used to indicate the predicted value of the duration required from the start of anesthesia to the start of the operation; the third predicted duration is used to indicate the predicted value of the duration required from the end of the operation to the patient leaving the operating room; the fourth predicted duration is used to indicate the predicted value of the cleaning duration of the operating room after the operation; the fifth predicted duration is used to indicate the predicted value of the duration required from the completion of the cleaning of the operating room to the next surgical patient entering the operating room.

[0046] In actual implementation, in order to accurately predict the process duration of the target operation, it is necessary to obtain the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration corresponding to the above target operation; wherein, the first predicted duration can be understood as the predicted value of the duration corresponding to the period from the start of timing after the patient enters the operating room to the end of timing when anesthesia starts; the second predicted duration can be understood as the predicted value of the duration corresponding to the period from the start of timing when anesthesia starts to the end of timing when the operation starts; the third predicted duration can be understood as the predicted value of the duration corresponding to the period from the end of timing when the operation ends to the end of timing when the patient leaves the operating room; the fourth predicted duration can be understood as the predicted value of the duration corresponding to the period from the start of timing when the patient leaves the operating room to the end of timing when the cleaning of the operating room ends; the fifth predicted duration can be understood as the predicted value of the duration corresponding to the period from the start of timing when the cleaning of the operating room ends to the end of timing when the next surgical patient enters the operating room.

[0047] Step S110, perform an addition process on the predicted operation duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to obtain the predicted operation process duration.

[0048] The above predicted operation process duration can be expressed by the following expression: Predicted operation process duration = Predicted operation duration + First predicted duration + Second predicted duration + Third predicted duration + Fourth predicted duration + Fifth predicted duration.

[0049] The above prediction method for the operation duration and the operation process duration obtains the operation application information corresponding to the target operation; extracts the target influencing factors from the operation application information; wherein, the target influencing factors include: operation name, patient age, patient gender, patient weight, patient disease severity data, operation surgeon information, patient underlying disease data; determines the predicted operation duration corresponding to the target operation based on the target influencing factors; wherein, the predicted operation duration represents the predicted value of the duration required from the start to the end of the operation; obtains the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration; wherein, the first predicted duration is used to indicate the predicted value of the duration required from the patient entering the operating room to the start of anesthesia; the second predicted duration is used to indicate the predicted value of the duration required from the start of anesthesia to the start of the operation; the third predicted duration is used to indicate the predicted value of the duration required from the end of the operation to the patient leaving the operating room; the fourth predicted duration is used to indicate the predicted value of the cleaning duration of the operating room after the operation; the fifth predicted duration is used to indicate the predicted value of the duration required from the completion of the cleaning of the operating room to the next operation patient entering the operating room; performs an addition process on the predicted operation duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration to obtain the predicted operation process duration. This method can determine the predicted operation duration based on the target influencing factors extracted from the operation application information, and then combine the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration to determine the predicted operation process duration; improves the accuracy of predicting the operation duration and the operation process duration, and thus can effectively ensure the rationality of the operation arrangement, reduce the waste of operation resources, and is also beneficial to operation safety.

[0050] The embodiment of the present invention also provides another prediction method for the duration of an operation and the duration of the operation process. This method is implemented on the basis of the method of the above embodiment, and this method includes the following steps:

[0051] Step 1, obtain the operation application information corresponding to the target operation;

[0052] Step 2, extract the target influencing factors from the operation application information; wherein, the target influencing factors include: operation name, patient age, patient gender, patient weight, patient disease severity data, operation surgeon information, patient underlying disease data;

[0053] The prediction of the predicted operation duration needs to be realized based on real, accurate and complete historical operation information. For this purpose, it is necessary to first collect historical operation data and perform data governance to ensure the quality of the original data.

[0054] In view of the normalized intersection of surgical basic data, surgical information based on electronic medical records and surgical information based on a surgical business system (such as a surgical anesthesia information system) are respectively selected as the basic data. The two surgical basic data are respectively subjected to prediction calculations under a unified method and model.

[0055] There are many factors affecting the accuracy of surgical duration prediction. These influencing factors need to be identified and converted into adjustment coefficients for surgical duration prediction. Through the analysis of historical surgical data, the target influencing factors affecting surgical duration identified in this application are mainly: surgical name, patient age, patient gender, patient weight, patient disease severity data, information of the surgeon in charge of the operation, and patient's underlying disease data.

[0056] Step 3: Obtain a first parameter that matches the surgical name according to the surgical name; wherein, the first parameter is the first historical average surgical duration or the first median surgical duration determined according to preset historical data;

[0057] The above historical data may be historical surgical data corresponding to a preset historical time period; the above first historical average surgical duration can be understood as the average value of multiple historical surgical durations related to the surgical name obtained from the historical data; the above first median surgical duration can be understood as the median of multiple historical surgical durations related to the surgical name obtained from the historical data. In actual implementation, a first parameter T can be obtained according to the above surgical name, M which includes: the first historical average surgical duration or the first median surgical duration.

[0058] Step 4: Obtain a second parameter that matches the surgical name and the information of the surgeon in charge of the operation according to the surgical name and the information of the surgeon in charge of the operation; wherein, the second parameter is the second historical average surgical duration or the second median surgical duration of the type of operation performed by the first surgeon in charge of the operation determined according to historical data; the first surgeon in charge of the operation is the surgeon in charge of the operation corresponding to the information of the surgeon in charge of the operation;

[0059] The above second historical average surgical duration can be understood as the average value of multiple historical surgical durations related to the surgical name and the information of the surgeon in charge of the operation obtained from the historical data; the above second median surgical duration can be understood as the median of multiple historical surgical durations related to the surgical name and the information of the surgeon in charge of the operation obtained from the historical data. In actual implementation, a second parameter T' can be obtained according to the above surgical name and the information of the surgeon in charge of the operation, M which includes: the second historical average surgical duration or the second median surgical duration.

[0060] Step 5: Obtain the weight coefficient corresponding to each other target influencing factor except the operation name and the information of the surgeon in charge of the operation. For each other target influencing factor, calculate the product of the basic adjustment value of this target influencing factor and the weight coefficient corresponding to this target influencing factor to obtain a product result; add the target parameter and each product result to obtain the predicted operation duration corresponding to the target operation; where the target parameter is the first parameter or the second parameter.

[0061] The above basic adjustment value is usually a pre-determined duration value. The basic adjustment values corresponding to different target influencing factors are usually different. For example, the basic adjustment value corresponding to the patient's age is 10 minutes, and the basic adjustment value corresponding to the patient's weight is 5 minutes, etc. It can be specifically set according to actual needs and is not limited here; in actual implementation, the target parameter is usually adjusted according to the patient's age, patient's gender, patient's weight, patient's disease severity, and patient's underlying disease data to obtain the predicted operation duration corresponding to the target operation.

[0062] When the target parameter is the first parameter T M the predicted operation duration T 0 corresponding to the operation is calculated by the formula:

[0063]

[0064] where the definitions of each variable are as follows:

[0065] (1) i represents the number of other target influencing factors except the operation name and the information of the surgeon in charge of the operation. That is, in this embodiment, the value of n is 5. It can be defined that 1 is influencing factor 1: patient's age, 2 is influencing factor 2: patient's gender, 3 is influencing factor 3: patient's weight, 4 is influencing factor 4: patient's disease severity data, and 5 is influencing factor 5: patient's underlying disease data; for the same target influencing factor, the weight coefficients corresponding to different patients are usually different. For example, the weight coefficient corresponding to a patient with 1 underlying disease is usually different from the weight coefficient corresponding to a patient with 2 or more underlying diseases.

[0066] (2) T i is the basic adjustment value of the i-th target influencing factor on the predicted operation duration T 0 This value can be the average value or median value determined based on historical data, etc.; for example, taking the patient's age as an example, the pre-determined basic adjustment value is 10 minutes, etc.

[0067] (3) represents the weight coefficient corresponding to the i-th target influencing factor. This weight coefficient is related to the patient. According to the complexity of this target influencing factor, it can be defined as different values. For example, T 1is the basic adjustment value corresponding to the patient's age. When the patient's age is 19 - 45 years old, the corresponding weight coefficient is 0.1. When the patient's age is 46 - 60 years old, the corresponding weight coefficient is 0.5, etc.

[0068] The target parameter is the second parameter T'. M At this time, the predicted operation duration T' corresponding to the operation 0 The calculation formula is:

[0069]

[0070] Where, T i is the basic adjustment value of the i-th target influencing factor on the predicted operation duration T'. 0 K f (T i ) represents the weight coefficient corresponding to the i-th target influencing factor. In actual implementation, after obtaining the above first parameter and second parameter, the weight coefficient corresponding to each other target influencing factor except the operation name and the information of the surgeon in charge of the operation can be continuously obtained. For each other target influencing factor, calculate the product of the basic adjustment value of this target influencing factor and the weight coefficient corresponding to this target influencing factor to obtain the product result; add the first parameter and each product result to obtain the predicted operation duration corresponding to the target operation, that is: It is also possible to add the second parameter and each product result to obtain the predicted operation duration corresponding to the target operation, that is Where each weight coefficient can take positive values, negative values, or 0. Therefore, the value of T 0 or T 0 's value will be able to increase or decrease compared to the first historical average operation duration or the second historical average operation duration.

[0071] In actual implementation, multiple preset influencing factors are pre-stored in the server. Each preset influencing factor corresponds to multiple classification categories, and each classification category has a corresponding adjustment coefficient; the above preset influencing factors usually include at least the above target influencing factors, and the target influencing factors corresponding to different operations may be the same or different; the number of the above classification categories can be set according to actual needs, and the number of classification categories corresponding to each preset influencing factor may be the same or different; for example, taking the preset influencing factor as the patient's age, the age can be divided into: <3 years old, 4 - 6 years old, 7 - 12 years old, 13 - 18 years old, 19 - 45 years old, 46 years old - 60 years old, 61 - 70 years old, 71 - 90 years old, >90 years old, 9 classification categories, and each classification category has a corresponding adjustment coefficient, which can be represented by K 11 -K 19 means.

[0072] The steps of obtaining the weight coefficient corresponding to each other target influencing factor except the surgical name and the information of the surgeon in charge of the operation can be implemented through the following steps 50 to 52:

[0073] Step 50, for each other target influencing factor except the surgical name and the information of the surgeon in charge of the operation, obtain the factor value of this target influencing factor;

[0074] Step 51, according to the factor value, determine the target classification category corresponding to this target influencing factor;

[0075] Step 52, obtain the adjustment coefficient corresponding to the target classification category, and determine the adjustment coefficient corresponding to the target classification category as the weight coefficient corresponding to this target influencing factor.

[0076] For example, taking the target influencing factor as the patient's age, if the patient's age is 50 years old (corresponding to the above factor value), then according to this factor value, the determined target classification category can be the classification category of "46 years old - 60 years old", and the adjustment coefficient corresponding to this classification category is K 16 then K 16 can be determined as the weight coefficient corresponding to this target influencing factor.

[0077] The adjustment coefficient corresponding to each classification category is determined in advance through the following steps 520 to 529:

[0078] Step 520, obtain the surgical application sample information corresponding to the sample operation;

[0079] Step 521, extract the sample influencing factors from the surgical application sample information; among them, the sample influencing factors include: sample surgical name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, sample surgeon in charge of the operation information, sample patient's underlying disease data;

[0080] Step 522, according to the sample surgical name, obtain the first sample parameter that matches the sample surgical name; among them, the first sample parameter is the first sample historical average operation duration or the first sample operation duration median determined according to the preset historical data;

[0081] Step 523, according to the sample surgical name and the information of the sample surgeon in charge of the operation, obtain the second sample parameter; among them, the second sample parameter is the second sample historical average operation duration or the second sample operation duration median determined according to the historical data;

[0082] Step 524, obtain the sample factor value corresponding to each other sample influencing factor except the sample surgical name and the information of the sample surgeon in charge of the operation, and the initial sample coefficient corresponding to each sample factor value;

[0083] Step 525: For each of the other sample influencing factors, calculate the product of the sample-based adjustment value of the sample influencing factor and the corresponding weight coefficient to obtain a sample product result. Add the target sample parameter and each sample product result to obtain the predicted sample operation duration corresponding to the sample operation; where the target sample parameter is the first sample parameter or the second sample parameter.

[0084] In actual implementation, after extracting sample influencing factors such as the sample operation name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, sample operation surgeon information, and sample patient underlying disease data from the operation application sample information corresponding to the sample operation, the first sample parameter L corresponding to the sample operation name can be calculated according to historical data and the sample operation name, such as the surgical procedure or operation code, etc. M , which includes: the first sample historical average operation duration or the first sample operation duration median; further add the sample operation surgeon information as a screening condition, and calculate the second sample parameter L corresponding to the sample operation name and the sample operation surgeon information according to historical data, the sample operation name, and the sample operation surgeon information. M ’, which includes the second sample historical average operation duration or the second sample operation duration median.

[0085] Furthermore, on the basis of the sample operation name and the sample operation surgeon information, add the sample patient age as a screening condition, and statistically calculate the historical average operation duration or the operation duration median of each surgical procedure or operation code according to different age groups. For example, if the age groups are divided into: <3 years old, 4 - 6 years old, 7 - 12 years old, 13 - 18 years old, 19 - 45 years old, 46 years old - 60 years old, 61 - 70 years old, 71 - 90 years old, >90 years old, a total of 9 categories, then statistically calculate the corresponding sample historical average operation duration L 11 -L 19 and the sample operation duration median L 11 ’-L 19 ’. In actual application, the sample historical average operation duration and the sample operation duration median are two commonly used reference values, and the appropriate reference value can be selected as needed.

[0086] In this embodiment, taking the selection and use of L 11 -L 19 as an example, compare the differences between L 11 -L 19 and L M to calculate the influence coefficient of the age factor on the operation duration, and initially determine K 11 -K 19As the age factor adjustment coefficient. For example: For the same surgical procedure performed by the same doctor, the deviation rate of the operation duration or median corresponding to patients aged > 90 years old, between 19 - 45 years old, etc., and the average duration or median of all operations can be used to determine the weight coefficient. The greater the deviation value, the greater the weight value of age.

[0087] Based on the sample surgical name and the information of the sample surgeon in charge, respectively add and compare in sequence the data of the gender of different sample patients, the weight of sample patients, the disease severity data of sample patients, and the underlying disease data of sample patients to determine the adjustment coefficient K of different factors ij ; where i is the i-th sample influencing factor, and j corresponds to the j-th classification category included in this sample influencing factor.

[0088] The following formula can be used to preliminarily estimate the predicted sample operation duration:

[0089] The predicted sample operation duration can be saved in the prediction database.

[0090] When performing surgical scheduling and operation dispatching, L M ’, L 0 ’ is more practical, but when analyzing the current situation and development trend of the overall surgical service ability of medical institutions and the personal ability of the surgeons in charge, L M , L 0 is more valuable.

[0091] Step 526, obtain the actual sample operation duration corresponding to the sample operation;

[0092] Step 527, calculate the difference between the actual sample operation duration and the predicted sample operation duration to obtain a difference result;

[0093] Step 528, if the difference result does not meet the preset first error range, for each other sample influencing factor, compare the actual sample operation duration with the historical duration data corresponding to the classification category to which the sample factor value of this sample influencing factor belongs to obtain a comparison result; among them, the historical duration data is: the sample historical average operation duration or the median of the sample operation duration corresponding to the classification category to which this sample factor value belongs.

[0094] The above first error range can be set according to actual requirements; in actual implementation, the actual sample operation duration corresponding to the sample operation can be obtained, and the actual sample operation duration is compared with the above predicted sample operation duration to obtain a difference result. If the difference result is within the preset first error range, the adjustment coefficient of each sample influencing factor does not need to be adjusted. If the difference result is outside the preset first error range, the adjustment coefficients of each sample influencing factor need to be analyzed in turn. For example, taking the sample influencing factor as the age of the sample patient, the analysis method is: extract the age value of the sample patient's age, and compare the actual sample operation duration with the sample historical average operation duration or the median of the sample operation duration corresponding to the classification category to which the age value belongs to obtain a comparison result.

[0095] Step 529, if the comparison result does not meet the preset second error range, obtain a coefficient adjustment instruction for this classification category, and adjust the adjustment coefficient of this classification category according to the coefficient adjustment instruction, and repeat the step of obtaining the surgical application sample information corresponding to the sample operation until the comparison result meets the preset second error range.

[0096] The above second error range can be set according to actual requirements; in actual implementation, if the comparison result indicates that the difference between the two is outside the second error range, the age coefficient needs to be determined as the coefficient to be adjusted. According to this method, further analyze and determine whether the adjustment coefficients of other influencing factors need to be adjusted to achieve iterative K ij weight coefficient. In addition, during iterative update, more or more accurate factors affecting the operation duration will be identified with the participation of humans, and then the prediction calculation will be performed according to the above steps; for example, the system calculates K 16 to be 1.2, but it is manually judged that it should be adjusted to 1.15, etc.

[0097] Step six, obtain the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration; among them, the first predicted duration is used to indicate the predicted value of the duration required from the patient entering the operating room to the start of anesthesia; the second predicted duration is used to indicate the predicted value of the duration required from the start of anesthesia to the start of the operation; the third predicted duration is used to indicate the predicted value of the duration required from the end of the operation to the patient leaving the operating room; the fourth predicted duration is used to indicate the predicted value of the cleaning duration of the operating room after the operation; the fifth predicted duration is used to indicate the predicted value of the duration required from the completion of the cleaning of the operating room to the entry of the next surgical patient into the operating room.

[0098] Step seven, perform an addition process on the predicted operation duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to obtain the predicted operation process duration.

[0099] The above prediction method for the operation duration and the operation process duration can determine the predicted operation duration based on the target influencing factors extracted from the operation application information, and then combine the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to determine the predicted operation process duration; improving the accuracy of predicting the operation duration and the operation process duration, thereby effectively ensuring the rationality of operation arrangements, reducing the waste of operation resources, and being beneficial to operation safety.

[0100] The embodiment of the present invention also provides another prediction method for the operation duration and the operation process duration. This method is implemented on the basis of the method in the above embodiment. In this method, the target operation includes multiple sub-operations; the method includes the following steps:

[0101] Step Eleven, for each sub-operation in the target operation, obtain the operation application information corresponding to the sub-operation;

[0102] The operation duration is closely related to the surgical procedure. In actual operations, one operation is often a combination of several single operations. In some combined operations, the single operations are fixed, while in others they are not, that is: Operation A = a1 + a2 + a3.., where a1, a2, a3 may or may not be fixed combinations. However, the duration of a combined operation containing multiple sub-operations is not simply the sum of the durations of each sub-operation. Therefore, prediction needs to be carried out through computational simulation.

[0103] Step Twelve, extract the target influencing factors from the operation application information; among them, the target influencing factors include: operation name, patient age, patient gender, patient weight, patient disease severity data, information of the surgeon in charge of the operation, and patient underlying disease data;

[0104] Step Thirteen, for each sub-operation, based on the target influencing factors, determine the predicted sub-operation duration corresponding to the sub-operation;

[0105] In actual implementation, since the target operation includes multiple sub-operations and belongs to a combined operation, when predicting the duration of such a combined operation, it is necessary to refer to the method in the foregoing embodiment to predict the predicted sub-operation duration corresponding to each sub-operation respectively.

[0106] Step Fourteen, obtain the weight coefficient corresponding to each sub-operation;

[0107] Step Fifteen, based on each weight coefficient, perform a weighted calculation on each predicted sub-operation duration to obtain the predicted operation duration corresponding to the target operation.

[0108] The predicted duration of the combined operation is:

[0109]

[0110] In the above formula,

[0111] ① i is the number of sub - surgeries.

[0112] ② P i is the weight coefficient corresponding to the i - th sub - surgery.

[0113] ③ T 0i and T 0i ′ both represent the predicted operation duration of the i - th sub - surgery; among them, T 0i and T′ 0i can be calculated with reference to the calculation method in the foregoing embodiments.

[0114] For example, taking the target surgery including: sub - surgery a1, sub - surgery a2, and sub - surgery a3 as an example, analyze the relationship between the sum of the predicted sub - surgery durations of sub - surgery a1, sub - surgery a2, and sub - surgery a3 as single surgeries and the predicted operation duration of the target surgery A, find out the weight coefficients corresponding to sub - surgery a1, sub - surgery a2, and sub - surgery a3 respectively, and perform weighted calculation on each predicted sub - surgery duration to obtain the predicted operation duration corresponding to the target surgery.

[0115] In addition, also for three sub - surgeries, there is usually a main surgery among the three sub - surgeries. When the main surgery is different, the overall operation duration may be different, that is, the weight coefficients of each sub - surgery will change. According to this information, analyze the weight coefficients. For example, when a1 is the main surgery, its coefficient is 1.2, but when it is an auxiliary surgery, it will become 0.7, etc.

[0116] Step Sixteen, obtain the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration; among them, the first predicted duration is used to indicate the predicted value of the duration required from the patient's entry into the operating room to the start of anesthesia; the second predicted duration is used to indicate the predicted value of the duration required from the start of anesthesia to the start of the operation; the third predicted duration is used to indicate the predicted value of the duration required from the end of the operation to the patient's leaving the operating room; the fourth predicted duration is used to indicate the predicted value of the cleaning duration of the operating room after the operation; the fifth predicted duration is used to indicate the predicted value of the duration required from the completion of the cleaning of the operating room to the entry of the next surgical patient into the operating room;

[0117] Step Seventeen, perform an addition process on the predicted operation duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to obtain the predicted operation process duration.

[0118] The embodiment of the present invention also provides another method for predicting the operation duration and the operation process duration. This method is implemented on the basis of the method in the foregoing embodiment, and this method includes the following steps:

[0119] Step Twenty, obtain the operation application information corresponding to the target surgery;

[0120] Step 21, extract target influencing factors from the surgical application information; among them, the target influencing factors include: surgical name, patient age, patient gender, patient weight, patient disease severity data, information of the surgeon in charge of the surgery, and patient's underlying disease data;

[0121] Step 22, input the factor values of each target influencing factor into the pre-trained surgical duration prediction model, so as to output the predicted surgical duration corresponding to the target surgery through the surgical duration prediction model.

[0122] In actual implementation, a machine learning algorithm based on a neural network can also be used to establish a more accurate surgical duration prediction model. Inputting the factor values of each target influencing factor into the trained surgical duration prediction model can directly output the predicted surgical duration corresponding to the target surgery.

[0123] The surgical duration prediction model is trained through the following steps 170 to 173:

[0124] Step 170, obtain the surgical application sample information corresponding to the sample surgery and the actual surgical duration corresponding to the sample surgery;

[0125] Step 171, extract the sample factor values of the sample influencing factors from the surgical application sample information; among them, the sample influencing factors include: sample surgical name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, information of the sample surgeon in charge of the surgery, and sample patient's underlying disease data;

[0126] Step 172, input each sample factor value into the initial model, so as to output the predicted surgical duration corresponding to the sample surgery through the initial model;

[0127] Step 173, calculate the loss value based on the predicted surgical duration and the actual surgical duration, update the parameters of the initial model based on the loss value, and repeat the step of obtaining the surgical application sample information corresponding to the sample surgery and the actual surgical duration corresponding to the sample surgery until the loss value converges, and obtain the trained surgical duration prediction model.

[0128] First, a simulation neural network can be built, that is, the above initial model: using the sample surgical name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity data, information of the sample surgeon in charge of the surgery, and sample patient's underlying disease data as neuron nodes. These neuron nodes can be connected to form a complex network structure due to mutual influence. Each neuron node receives inputs from other neuron nodes and generates outputs based on these inputs.

[0129] Then perform initialization calculations: Send the historical surgical information to the input ends of each neuron node in the neural network. The historical surgical information includes: sample surgical name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity data, sample information of the surgeon in charge of the operation, sample patient's underlying disease data, etc.; Then, through the weights and activation functions in the neural network, each neuron node will output the result after processing the weighted sum of its input through the activation function, and this result is the predicted surgical duration. In this solution, in order to enable the neural network to learn and simulate complex relationships, Sigmoid is used as the activation function. The initial value of the activation function can be determined with reference to the weight coefficient calculation method in the foregoing embodiments.

[0130] Next, perform error calculations: Further, at the output layer of the neural network, calculate the loss value by comparing the difference between the predicted surgical duration of the network and the actual surgical duration. The mean square error method can be used to complete this.

[0131] Finally, perform iterative updates: According to the calculated loss value, adjust the weights of each neuron in the neural network through the backpropagation algorithm. This process is achieved by calculating the gradient of the loss value with respect to the weights and updating the weights in the negative direction of the gradient. By continuously iterating this process, the prediction performance of the neural network will gradually improve.

[0132] Similarly, during iterative updates, when the difference between the predicted surgical duration and the actual surgical duration is relatively large, for example, when the difference / actual surgical duration is greater than 70%, before repeating the above steps, more or more accurate factors affecting the surgical duration will be identified with the participation of humans, and then the neuron nodes will be supplemented or deleted.

[0133] Step 23, obtain the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration; among them, the first predicted duration is used to indicate the predicted value of the duration required from the patient's entry into the operating room to the start of anesthesia; the second predicted duration is used to indicate the predicted value of the duration required from the start of anesthesia to the start of the operation; the third predicted duration is used to indicate the predicted value of the duration required from the end of the operation to the patient's departure from the operating room; the fourth predicted duration is used to indicate the predicted value of the cleaning duration of the operating room after the operation; the fifth predicted duration is used to indicate the predicted value of the duration required from the completion of the cleaning of the operating room to the entry of the next surgical patient into the operating room;

[0134] Designate the predicted duration as: the first predicted duration, the third predicted duration, or the fifth predicted duration; the designated predicted duration is determined in advance through the following steps 180 to 182:

[0135] Step 180, obtain multiple time values corresponding to the designated predicted duration within a preset historical time period;

[0136] Step 181: Extract the maximum and minimum values from multiple time values;

[0137] Step 182: If the difference between the maximum value and the minimum value is less than a preset difference threshold, calculate the average value of the multiple time values and use the average value as the specified prediction duration.

[0138] The above-mentioned preset historical time period can be set according to actual needs. For example, it can be the most recent year or the most recent half year in history, etc.; the above-mentioned preset difference threshold can be set according to actual needs. For example, it can be 5 minutes, etc.; taking the first prediction duration as an example, in actual implementation, the first prediction duration is used to indicate the difference between the anesthesia start time and the patient's entry time into the operating room. Then, from the preset historical time period, multiple time values representing the difference between the anesthesia start time and the patient's entry time into the operating room can be obtained, and the maximum and minimum values among the multiple time values can be found. If the difference between the maximum value and the minimum value is less than the preset difference threshold, the average value of the corresponding multiple time values can be used as the first prediction duration.

[0139] The fourth prediction duration is determined in advance through the following steps 183 to 186:

[0140] Step 183: Determine the infection classification to which the target surgery belongs;

[0141] Step 184: Obtain the historical cleaning durations corresponding to multiple historical surgeries belonging to the infection classification from historical data;

[0142] Step 185: Calculate the first calculation result of the multiple historical cleaning durations; where the first calculation result is the median or the mean;

[0143] Step 186: Use the first calculation result as the fourth prediction duration.

[0144] The above-mentioned infection classification can include: routine, infection, etc.; since surgeries with different infection classifications usually require different cleaning durations, when determining the fourth prediction duration, the infection classification to which the target surgery belongs can be determined first, and the historical cleaning durations corresponding to multiple historical surgeries belonging to this infection classification can be obtained from historical data. Calculate the median or the mean of the multiple historical cleaning durations, and the calculated median or mean can be used as the fourth prediction duration.

[0145] The second prediction duration is determined in advance through the following steps 187 and 188:

[0146] Step 187: Extract the first influencing factor from the surgery application information; where the first influencing factor includes: patient age, anesthesia method, preoperative risk assessment information, patient's underlying diseases;

[0147] Step 188: Determine the second prediction duration based on the first influencing factor.

[0148] In actual implementation, the second predicted duration is usually related to information such as the patient's age, anesthesia method, preoperative risk assessment information, and the patient's underlying diseases. When determining the second predicted duration, the same method as for predicting the surgical duration can be adopted. First, identify the first influencing factor and quantify the weight coefficient of the first influencing factor on the second predicted duration, or use a machine learning method to determine the second predicted duration.

[0149] Step twenty-four: Add up the predicted surgical duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to obtain the predicted surgical process duration.

[0150] To improve the accuracy of surgical prediction, various types of surgeries can be grouped, and different groups can be predicted using the methods in the foregoing embodiments respectively. The purpose of grouping is to combine similar surgeries together so that there will not be too much difference during measurement. For example, transplantation surgeries generally take a long time, possibly exceeding 8 hours, while most ophthalmic surgeries only take 30 minutes. If the surgical durations are measured together, there will be a large deviation.

[0151] The bases for grouping usually include: grouping according to the degree of difference in the duration of a certain type of surgery in historical surgical data, grouping according to specialties (such as ophthalmology, orthopedic surgery, neurosurgery, etc.), grouping according to the level of the surgeon in charge (such as attending physician, deputy chief physician, chief physician, etc.), and grouping according to outpatient surgeries, day surgeries, and non-day surgeries (day surgery: refers to surgeries that have been included in the day surgery catalog of medical institutions), etc.

[0152] Step twenty-five: Display the predicted surgical duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration;

[0153] Step twenty-six: If an update instruction for the predicted surgical duration is received, update the predicted surgical duration according to the update instruction;

[0154] Step twenty-seven: If an adjustment instruction for the target predicted duration is received, adjust the target predicted duration according to the adjustment instruction; where the target predicted duration includes at least one of the following: the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration.

[0155] To accelerate the iteration efficiency of the surgical duration prediction model, an artificial participation node is added to this solution.

[0156] Adjustment by the operating surgeon: When the operating surgeon fills in the operation application, the predicted operation duration can be automatically provided to the operating surgeon. The operating surgeon can initially determine whether the predicted operation duration is reasonable. If it is considered unreasonable, it can be adjusted. In order to guide the operation application duration information filled in by the operating surgeon to be as close to the actual situation as possible, when providing the predicted operation duration to the operating surgeon, the personal past operation duration of this surgical procedure, as well as the median, maximum, and minimum operation durations of the whole hospital, will also be provided for reference.

[0157] Adjustment by the operation scheduling staff: Based on past work experience and other information they have, the operation scheduling staff analyzes the target predicted duration and the predicted operation duration adjusted by the operating surgeon, and adjusts and confirms the target predicted duration and the predicted operation duration adjusted by the operating surgeon as appropriate, as the data finally participating in the operation scheduling. For example, if the predicted operation duration adjusted by the operating surgeon predicts that the operation will take 1 hour, and the operation scheduling staff knows this operating surgeon well and the rhythm is relatively slow, it may be adjusted to 1 hour and 15 minutes, etc.

[0158] On the Gantt chart for operation scheduling prediction drawn by the system, the operation scheduling staff adjusts the duration of the Gantt chart according to experience, and manually adjusts data such as the connection duration between two operations and the duration from the end of the operation to the patient leaving the operating room, so as to generate the final operation process.

[0159] The predicted operation duration adjusted by the operating surgeon, the target predicted duration and the predicted operation duration adjusted by the operation scheduling staff, etc. will all be fed back to the algorithm model to participate in subsequent iterations.

[0160] Step 28: Display the confidence level for the target duration information; among them, the target duration information includes: predicted operation duration, predicted operation process duration.

[0161] In order to provide the surgical relevant personnel with the accuracy information of the operation duration and operation process prediction, this method can provide the confidence level α. When the confidence level is in different numerical ranges, different colors can be used to mark on the operation process chart to facilitate the surgical personnel to understand the accuracy of this operation duration and operation process prediction. When this method outputs the predicted operation duration and the predicted operation process duration, it can also output the corresponding confidence level to remind the operation scheduling staff of the accuracy of this predicted value.

[0162] The confidence level is calculated by the following formula:

[0163] α = 1 - |(target duration information - reference value) / reference value| × 100%;

[0164] Among them, when the number of historical surgical samples corresponding to the surgeon in the information of the operating surgeon meets the preset quantity threshold, the reference value is determined based on the average or median duration of multiple historical surgical samples of the surgeon; when the number of historical surgical samples corresponding to the surgeon in the information of the operating surgeon does not meet the preset quantity threshold, the reference value is determined based on the average or median duration of multiple historical surgical samples of other operating surgeons with the same professional title as the operating surgeon.

[0165] The value range of α is 0 - 100%, which can be divided into three interval dimensions: 0 - 30% with the lowest confidence level; 31 - 70% with a medium confidence level; >70% with the highest confidence level, and the division threshold of the confidence interval dimension can be adjusted.

[0166] The calculation basis of the above reference value is as follows:

[0167] When the number of samples in the historical surgical data of the operating surgeon is sufficient (for example, >50 cases), the average or median duration of multiple historical surgical samples of the operating surgeon is used as the reference value;

[0168] When the number of samples in the historical surgical data of the operating surgeon is insufficient, the average or median duration of multiple historical surgical samples of other operating surgeons with the same professional title as the surgical procedure is used as the reference value;

[0169] For the surgical procedures newly carried out by the hospital and the surgeries with the total number of historical surgical cases in the whole hospital not exceeding 20 cases, no prediction is made and the operating surgeon fills it in by himself / herself. After the cumulative number of cases of this surgical procedure exceeds 20 cases, the prediction is carried out in the manner of the foregoing embodiments and confidence information is provided.

[0170] Step twenty-nine, if a first instruction indicating an abnormal event is received, extract the first surgical information associated with the surgical name from the preset historical data;

[0171] Step thirty, extract the duration adjustment information associated with the abnormal event from the first surgical information;

[0172] Step thirty-one, adjust the target duration information according to the duration adjustment information.

[0173] In this embodiment, during the surgical operation, according to the occurrence of special intraoperative events, the predicted surgical duration and the predicted duration of the surgical process can be adjusted in real time and dynamically.

[0174] A type of special event is an event that occurs abnormally during the operation, including rescue, timeout for the intraoperative specimen result to be transmitted back, temporary change of the intraoperative surgical plan, damage to surgical instruments, the patient not arriving at the operating room on schedule, timeout for the operating room cleaning time, etc., which will all affect the surgical duration and the duration of the surgical process.

[0175] Therefore, this embodiment can provide several ways for adjustment. One way is manual intervention. When an abnormal event occurs, the operating nurse manually fills in the abnormal event and the possible affected duration. After the system obtains this information, it will adjust the predicted operation duration and the predicted operation progress duration of this operation. For example, abnormal information can be filled in manually, the predicted progress bar can be dragged manually, the predicted operation duration can be lengthened, etc. Based on the relationship between the pre-stored abnormal events and the duration changes, such as when this abnormal event occurs, the operation duration increases by 5 minutes, etc., the target influencing factors and weight coefficients of the duration prediction model can be enriched; the system marks the operations with abnormal events that occurred during the operation and classifies them separately, which is used as the basis for the intraoperative dynamic prediction of the operation duration when similar abnormal events occur in the future. For example, if amniotic fluid embolism or other rescue events occur during a cesarean section operation, then this operation will be marked. After the marked operation is confirmed and approved by several other physicians at the same level or the superior doctor, it can be used as a normal medical record, and the relevant data of this operation will be set as the influencing factors for similar rescue operations. In this way, when an abnormal event occurs again, the information of similar events that occurred in similar operations can be extracted from the previous operation information, and the operation duration or the change of the operation progress can be compared and adjusted automatically.

[0176] Another type of special event is a normal event during the operation, but the operation plan will be determined according to the progress of the operation during the operation, resulting in possible different operation durations. For example, the pathological tissue removed during the operation needs to be biopsied. When this event occurs, the system's processing plan for the predicted operation duration of this operation is: on the one hand, according to the routine pathological specimen examination, the waiting duration will be automatically added when predicting the operation duration; on the other hand, the system further calculates based on the obtained pathological specimen examination report results and the information of the subsequent operation plan that the doctor intends to adopt, such as directly removing the lesion, or expanding the resection range, or suspending the operation, etc., and automatically increases or shortens the expected operation duration on the basis of the predicted operation duration data before the operation, and at the same time makes a mark, including the results of benign or malignant of the pathological specimen examination report, the selection of the operation plan, etc., as a reference for the subsequent operation duration prediction.

[0177] There is another way. When no abnormal event occurs, but the operation status information, operation progress, and patient's vital sign information are obtained in real time during the operation implementation process, the difference between the actual elapsed duration of the operation and the predicted data of this stage is compared in stages. This difference indicates that the operation duration has been extended by XX minutes or shortened by XX minutes. The system immediately starts to correct the predicted data, supplements or reduces the influencing factors and adjusts the weight coefficients, and replaces the duration data predicted before the operation with the real-time calculation result. This process will continue throughout the operation implementation process, and the predicted difference of each stage will be used as the basic data for starting the correction in the next stage. After continuous correction, the operation end time becomes more and more accurate.

[0178] In practical applications, the above-mentioned prediction methods for the surgical duration and the surgical process duration can be implemented through a duration prediction system, which can be composed of a front-end application and a back-end service.

[0179] The front-end application can include a surgical application initiation module, a surgical application reception module, a surgical application arrangement module, a surgical scheduling module, a data dashboard module, etc.

[0180] Among them, the surgical application initiation module provides an input interface for the surgeon-in-charge to initiate a surgical application; the surgical application reception module is used to receive and display surgical application information; the surgical application arrangement module can arrange each surgery to a suitable operating room according to the surgical application; the surgical scheduling module can arrange newly added surgical applications to a suitable operating room; the data dashboard module displays in real time a surgical process prediction chart drawn based on the surgical duration, an actual surgical operation process chart, a surgical duration anomaly reminder, and the surgery ending before the predicted time, etc.

[0181] The core modules of the back-end service are: a surgical duration database and an algorithm module. The surgical duration database includes a surgical duration prediction database, a combined surgical duration prediction database, and a dynamic surgical duration prediction database. After the surgery ends, the data in the dynamic surgical duration prediction database will be transferred to the first two databases. The algorithm module is a surgical duration prediction model. The algorithm module can be docked with the surgical application reception module at the front end, input as model parameters, and the output results will be docked with the surgical application arrangement module, the surgical scheduling module, and the data dashboard module.

[0182] For easy understanding, another embodiment is provided below to illustrate the construction of the surgical duration prediction model and its application in the surgical arrangement and scheduling process. In practical applications, surgical duration prediction is of great significance for the reasonable arrangement and flexible scheduling of surgical resources and the scientific evaluation of the surgical capabilities of surgeons-in-charge. However, surgical duration prediction is a complex problem because it is affected by various factors, including the surgical name, the patient's health status, the experience of the surgeon-in-charge, and various unforeseeable factors during the surgery. At present, some medical institutions and research teams are using big data and artificial intelligence technologies to develop more accurate surgical duration prediction models, but there are few cases of surgical duration prediction used in engineering practice in China. These models predict the possible duration of the surgery by analyzing a large amount of surgical data, including information such as the surgical name, patient age, gender, weight, disease severity, and the experience level of the surgeon-in-charge. This embodiment intends to predict the duration of the surgery applied by doctors mainly based on the surgeon-in-charge and the surgical procedure on the basis of analyzing the existing surgical applications and implementation data in the hospital; subsequently, it intends to further comprehensively apply technologies such as big data and artificial intelligence, and further optimize the model in combination with data such as the basic situation of surgical patients.

[0183] The basic process of iterative improvement of the operation duration prediction model is briefly described as follows:

[0184] 1. Obtain the basic dictionary and operation data related to surgeries in the hospital;

[0185] 2. Based on the basic data, establish 3 sets of operation duration prediction libraries, and continuously update the corresponding operation duration prediction data during the operation of the system;

[0186] 3. In each link such as operation application, operation arrangement, operation process browsing, operation scheduling, etc., predict the operation duration according to the data in the link;

[0187] 4. Support scientific research and model improvement, continuously monitor and analyze the operation duration prediction and actual result data, and iteratively improve the operation duration prediction model.

[0188] As Figure 2 shown in the schematic diagram of the basic process of iterative improvement of an operation duration prediction model, in the operation application stage, the doctor makes an operation application, and can input the surgeon, the planned operation, the basic situation of the patient, etc. into the operation duration prediction model. Through the operation duration prediction model, the predicted operation duration t1 is obtained and fed back to the doctor. The doctor corrects it to the predicted operation duration t2. In the operation arrangement stage, the operation safety personnel can correct part of the predicted operation duration t3 according to experience. In the operation process stage, through the terminal device, the actual time t4 of the implemented operation and the predicted operation durations t1, t2, t3 of the operations not yet carried out can be displayed; based on the implemented operation, the planned operation, the standard name and the relevant durations t1, t2, t3, t4, monitoring and analysis can be carried out. Specifically, the following methods can be referred to for analyzing and iterating the prediction algorithm and the operation duration prediction library generation algorithm:

[0189] (1) Analysis of the compliance between the planned operation and the implemented operation;

[0190] (2) Analysis of the compliance between the predicted operation durations t1, t2, t3 and the actual duration t4;

[0191] (3) Add the duration of combined operations (operations including multiple sub-operations) to the model;

[0192] (4) Improve the algorithm by adding factors such as patient age, gender, weight, and disease severity.

[0193] Combined with Figure 2 , the initialization and update of the operation duration prediction library are described as follows:

[0194] 1. Single-operation duration library

[0195] Based on the surgical dictionary, combined with the main surgery in the surgical record of the historical medical record front page and the data of the surgical duration in the anesthesia record, count the number of cases of each main surgery and the median surgical duration, and conduct statistical analysis regularly as the basis for predicting the surgical duration within this cycle.

[0196] 2. Surgical combination duration library

[0197] Combine the main surgery and assistant surgery in the surgical record of the historical medical record front page, and the surgical duration and surgeon data in the anesthesia record, count the number of cases of each surgical combination for each surgeon and the median surgical duration; ignore the surgeon and count the number of cases of each surgical combination and the median surgical duration.

[0198] Statistical update is performed once after 24:00 every day.

[0199] 3. Surgical combination duration library (based on anesthesia record)

[0200] Split the surgical name in the anesthesia record, count the number of cases of each surgical combination for each surgeon and the median surgical duration; ignore the surgeon and count the number of cases of each surgical combination and the median surgical duration.

[0201] Statistical update is performed once after 24:00 every day.

[0202] Historical surgical application and arrangement data, historical anesthesia records, historical surgical records on the front page of medical records, and surgical dictionary data can be obtained as basic data, and the database table names and fields associated with these basic data, as well as the cycle for obtaining these basic data, are specified. Of course, according to actual needs, appropriate data content can be selected from other relevant table names, and the fields and cycles can also be adjusted as needed, which are not limited here.

[0203] The following explains the new application surgical duration prediction algorithm in this embodiment. As Figure 3 shown in the flowchart of a surgical duration prediction algorithm, the prediction in other stages refers to this process. In the surgical application stage, the surgical duration prediction algorithm is called. The general process is as follows: First, complete the filling or modification of the main surgical information, call the surgical duration prediction service to obtain the predicted surgical duration t1. According to the surgeon, surgical technique, and other application information, match the surgical combination duration library. If the match fails, match the surgical combination duration library again. If the match fails, loop through each surgical name to match the single surgical duration library. If the match fails, a default value defined by the operating room is given; if the match is successful in any of the above matching processes, the predicted surgical duration t1 can be used as the surgical prediction duration t1 (corresponding to the above predicted surgical duration). Based on t1, the doctor predicts and fills in the surgical prediction duration t2 according to his own experience. This surgical prediction duration t2 can be the same as or different from t1, and can be specifically confirmed according to the doctor's own experience.

[0204] The following explains the analysis and iteration process of the surgical duration prediction model:

[0205] 1. Analysis Iteration before System Implementation

[0206] (1) Based on the data of the previous year (or other appropriate data can be selected as needed), a single surgical duration library, a combined surgical duration library, and a combined surgical duration library can be experimentally established (;

[0207] (2) The data of the most recent year (or other appropriate data can be selected as needed) can be used as an application to verify the accuracy of the predicted surgical duration, and the algorithm can be optimized accordingly.

[0208] Based on this embodiment, corresponding expansions can also be carried out. For example, the surgical ability can be evaluated based on the surgical duration prediction, etc.

[0209] The prediction method for surgical duration and surgical process duration disclosed in this solution includes duration prediction and process prediction. Among them, the duration prediction is to predict the duration based on the surgical application information on the day before the operation, and this is used as the basis for surgical scheduling; the dynamic adjustment of the intraoperative duration is adjusted according to intraoperative abnormal events, and the event information and adjustment values used for adjustment are fed back into the prediction model. When predicting the surgical process, the start time of the next operation cannot be accurately predicted only based on the surgical duration, because the interval between two operations also includes cleaning events and events of receiving the next patient. The duration of the cleaning event is also related to characteristics such as whether the operation is an infectious operation, and the duration of receiving the patient to the operating room is also related to the handover arrangement and the patient's status. These factors are comprehensively considered in the surgical process prediction of this solution.

[0210] When this solution predicts the surgical duration, it goes through two steps: single surgery and combined surgery; before sample collection, instead of simply using all the data, it is first preliminarily classified according to conditions such as specialty, doctor, and medical team, and then the relevant parameters are input into the model, which is a supervised prediction method. In addition, during the prediction process, an artificial intervention adjustment method is also used to ensure that the predicted data is more accurate.

[0211] The prediction method adopted by this solution does not have a function for solving the surgical duration. Instead, through the analysis of different samples, the weight coefficients of various influencing factors on the surgical duration are calculated. The functional relationship between each influencing factor and the surgical duration is unknown and can only be determined after comparison and analysis, and further determine the coefficient value.

[0212] The prediction method for surgical duration and surgical process duration provided by this solution realizes the accurate prediction of surgical duration, improves the utilization rate of surgical resources, reduces the length of stay of patients in the hospital, and improves the overall medical service level of the hospital.

[0213] Based on the prediction of the surgical duration, further combining data such as the anesthesia duration from the start of patient anesthesia to the start of surgery and the cleaning duration of the operating room, the prediction of the surgical process can be achieved, thereby providing accurate information for surgical scheduling, preparation for consecutive surgeries, etc., and providing support for surgical arrangement and dispatching, analysis of medical staff capabilities, analysis of surgical operation efficiency, etc.

[0214] The present invention realizes the accurate prediction of the duration consumed in each stage of each surgical link, excavates and analyzes the factors affecting the surgical duration and the surgical process duration in each link, further classifies the surgical categories in more detail, and extracts the personalized characteristics of patients. On this basis, it analyzes the influence degree of these factors and personalized characteristics affecting the surgical duration and the surgical process duration on different categories of surgeries, establishes a prediction model for the surgical duration and the surgical process duration, and further makes the model more accurate through actual data fitting and regression calculation.

[0215] The above method for predicting the surgical duration and the surgical process duration can determine the predicted surgical duration based on the target influencing factors extracted from the surgical application information, and then combine the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to determine the predicted surgical process duration; it can achieve accurate prediction and dynamic update of the surgical duration and the surgical process duration, and this method has the following characteristics:

[0216] 1. Obtain the average surgical duration of this category according to the surgical category and historical data, as the most basic factor.

[0217] 2. Accurately analyze the contribution (influence) of the durations of different links to the overall surgical duration and the surgical process duration.

[0218] 3. For complex surgeries, identify combined surgeries and sub - surgeries, and further analyze the influence of factors in sub - surgeries and combined surgeries.

[0219] 4. Identify, through big data, the characteristics in the personalized characteristics of patients that can affect the surgical duration, and analyze, through data analysis, the weights of these characteristics on the surgical duration (for example, in cases with different values, it may affect an increase or a decrease in the surgical duration).

[0220] 5. During the implementation process of this method, use the actual data of implementation and application to perform regression calculation on the model for dynamic update;

[0221] 6. The above methods and processes are integrated into an integrated and comprehensive application system, achieving accurate prediction of the surgical duration and the surgical process duration.

[0222] This solution realizes predictable, calibratable, and iterative prediction of the surgical duration and the surgical process duration, and has achieved the following remarkable benefits.

[0223] (1) Accurate prediction of the surgical duration and surgical process data can make surgical arrangements more reasonable, fully utilize limited surgical resources, improve the efficiency of surgical connection, and facilitate the reasonable arrangement of work for surgical staff.

[0224] (2) Based on the confidence index, remind the surgical scheduling staff to manually adjust the duration or process duration based on the confidence, and actively pay attention to the details of the implementation process of this surgery.

[0225] (3) When the actual surgical duration exceeds the predicted data by a certain range during the surgery, the prompt information can remind medical staff to take corresponding measures (such as measures to prevent pressure sores, antibiotic administration measures, anesthesia measures, etc.), providing guarantee for surgical safety.

[0226] (4) Timely correction of the predicted data during the operation provides more accurate information for the next surgery, making surgical management more orderly.

[0227] (5) Actively guide the surgeon to actively participate in the prediction of the surgical duration, which provides greater help for the usability and accuracy of the system.

[0228] (6) The system data provides a basis for evaluating the capabilities of surgeons, the capabilities of surgeons, and the overall service capabilities of medical institutions.

[0229] An embodiment of the present invention provides a prediction device for surgical duration and surgical process duration. The device is set in a server; as Figure 4 shown, the device includes:

[0230] A first acquisition module 40, configured to acquire surgical application information corresponding to a target surgery;

[0231] An extraction module 41, configured to extract target influencing factors from the surgical application information; wherein, the target influencing factors include: surgical name, patient age, patient gender, patient weight, patient disease severity data, surgeon information, patient underlying disease data;

[0232] A determination module 42, configured to determine a predicted surgical duration corresponding to the target surgery based on the target influencing factors; wherein, the predicted surgical duration represents an estimated value of the duration required from the start to the end of the surgery;

[0233] The second acquisition module 43 is configured to acquire a first predicted duration, a second predicted duration, a third predicted duration, a fourth predicted duration, and a fifth predicted duration; wherein, the first predicted duration is used to indicate an estimated value of the duration required from the patient's entry into the operating room to the start of anesthesia; the second predicted duration is used to indicate an estimated value of the duration required from the start of anesthesia to the start of the operation; the third predicted duration is used to indicate an estimated value of the duration required from the end of the operation to the patient's departure from the operating room; the fourth predicted duration is used to indicate an estimated value of the cleaning duration of the operating room after the operation; the fifth predicted duration is used to indicate an estimated value of the duration required from the completion of the cleaning of the operating room to the entry of the next surgical patient into the operating room;

[0234] The processing module 44 is configured to perform an addition process on the predicted operation duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to obtain a predicted operation process duration.

[0235] The above-mentioned operation duration prediction device can determine the predicted operation duration based on the target influencing factors extracted from the operation application information, and then combine the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration to determine the predicted operation process duration; improving the accuracy of predicting the operation duration and the operation process duration, and thus can effectively ensure the rationality of the operation arrangement, reduce the waste of operation resources, and is also beneficial to operation safety.

[0236] Further, the determination module is further configured to: obtain a first parameter that matches the operation name according to the operation name; wherein, the first parameter is the first historical average operation duration or the first operation duration median determined according to the preset historical data; obtain a second parameter that matches the operation name and the information of the primary surgeon of the operation according to the operation name and the information of the primary surgeon of the operation; wherein, the second parameter is the second historical average operation duration or the second operation duration median of the type of operation performed by the first primary surgeon of the operation; the first primary surgeon of the operation is the primary surgeon corresponding to the information of the primary surgeon of the operation; obtain the weight coefficient corresponding to each other target influencing factor except the operation name and the information of the primary surgeon of the operation, and for each other target influencing factor, calculate the product of the basic adjustment value of the target influencing factor and the weight coefficient corresponding to the target influencing factor to obtain a product result; add the target parameter and each product result to obtain the predicted operation duration corresponding to the target operation; wherein, the target parameter is the first parameter or the second parameter.

[0237] Further, a plurality of preset influencing factors are pre-stored in the server, each preset influencing factor corresponds to a plurality of classification categories respectively, and each classification category has a corresponding adjustment coefficient; the determination module is further configured to: for each target influencing factor other than the operation name and the information of the surgeon in charge of the operation, obtain the factor value of the target influencing factor; determine the target classification category corresponding to the target influencing factor according to the factor value; obtain the adjustment coefficient corresponding to the target classification category, and determine the adjustment coefficient corresponding to the target classification category as the weight coefficient corresponding to the target influencing factor.

[0238] Further, the adjustment coefficient corresponding to each classification category is determined in advance by the following method: obtain the operation application sample information corresponding to the sample operation; extract the sample influencing factors from the operation application sample information; wherein, the sample influencing factors include: sample operation name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, sample surgeon in charge of the operation information, sample patient's underlying disease data; according to the sample operation name, obtain the first sample parameter that matches the sample operation name; wherein, the first sample parameter is the first sample historical average operation duration or the first sample operation duration median determined according to the preset historical data; according to the sample operation name and the information of the surgeon in charge of the sample operation, obtain the second sample parameter; wherein, the second sample parameter is the second sample historical average operation duration or the second sample operation duration median determined according to the historical data; obtain the sample factor value corresponding to each sample influencing factor other than the sample operation name and the information of the surgeon in charge of the sample operation, and the initial sample coefficient corresponding to each sample factor value; for each other sample influencing factor, calculate the product of the sample basic adjustment value of the sample influencing factor and the initial sample coefficient corresponding to the sample influencing factor to obtain a sample product result, and add the target sample parameter and each sample product result to obtain the predicted sample operation duration corresponding to the sample operation; wherein, the target sample parameter is the first sample parameter or the second sample parameter; obtain the actual sample operation duration corresponding to the sample operation; calculate the difference between the actual sample operation duration and the predicted sample operation duration to obtain a difference result; if the difference result does not meet the preset first error range, for each sample influencing factor, compare the actual sample operation duration with the historical duration data corresponding to the classification category to which the sample factor value of the sample influencing factor belongs to obtain a comparison result; wherein, the historical duration data is: the sample historical average operation duration or the sample operation duration median corresponding to the classification category to which the sample factor value belongs; if the comparison result does not meet the preset second error range, obtain the coefficient adjustment instruction for this classification category, so as to adjust the adjustment coefficient of this classification category according to the coefficient adjustment instruction, and repeat the step of obtaining the operation application sample information corresponding to the sample operation until the comparison result meets the preset second error range.

[0239] Further, the determination module is further configured to: input the factor values of each target influencing factor into a pre-trained surgical duration prediction model, so as to output the predicted surgical duration corresponding to the target surgery through the surgical duration prediction model.

[0240] Further, the surgical duration prediction model is trained in the following manner: obtaining the surgical application sample information corresponding to the sample surgery and the actual surgical duration of the sample surgery; extracting the sample factor values of the sample influencing factors from the surgical application sample information; wherein, the sample influencing factors include: sample surgery name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, sample surgeon information, sample patient's underlying disease data; inputting each sample factor value into the initial model to output the predicted surgical duration corresponding to the sample surgery through the initial model; calculating a loss value based on the predicted surgical duration and the actual surgical duration, updating the parameters of the initial model based on the loss value, and repeating the steps of obtaining the surgical application sample information corresponding to the sample surgery and the actual surgical duration of the sample surgery until the loss value converges, thereby obtaining the trained surgical duration prediction model.

[0241] Further, the target surgery includes multiple sub-surgeries; the first acquisition module is further configured to: for each sub-surgery in the target surgery, obtain the surgical application information corresponding to the sub-surgery; the step of determining the predicted surgical duration corresponding to the target surgery based on the target influencing factors includes: for each sub-surgery, determining the predicted sub-surgical duration corresponding to the sub-surgery based on the target influencing factors; obtaining the weight coefficient corresponding to each sub-surgery; and performing weighted calculation on each predicted sub-surgical duration based on each weight coefficient to obtain the predicted surgical duration corresponding to the target surgery.

[0242] Further, the specified prediction duration is: the first prediction duration, the third prediction duration, or the fifth prediction duration; the specified prediction duration is determined in the following manner in advance: obtaining multiple time values corresponding to the specified prediction duration within a preset historical time period; extracting the maximum value and the minimum value from the multiple time values; if the difference between the maximum value and the minimum value is less than the preset difference threshold, calculating the average value of the multiple time values and taking the average value as the specified prediction duration.

[0243] Further, the fourth prediction duration is determined in the following manner in advance: determining the infection classification to which the target surgery belongs; obtaining the historical cleaning durations corresponding to multiple historical surgeries belonging to the infection classification from the historical data; calculating a first calculation result of the multiple historical cleaning durations; wherein, the first calculation result is the median or the mean value; and taking the first calculation result as the fourth prediction duration.

[0244] Further, the second predicted duration is determined in advance in the following manner: extracting a first influencing factor from the surgical application information; wherein, the first influencing factor includes: patient age, anesthesia method, preoperative risk assessment information, and patient's underlying diseases; determining the second predicted duration based on the first influencing factor.

[0245] Further, the device is further configured to: display the predicted surgical duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration; if receiving an update instruction for the predicted surgical duration, updating the predicted surgical duration according to the update instruction; if receiving an adjustment instruction for the target predicted duration, adjusting the target predicted duration according to the adjustment instruction; wherein, the target predicted duration includes at least one of the following: the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration.

[0246] Further, the device is further configured to: display the confidence level for the target duration information; wherein, the target duration information includes: the predicted surgical duration, the predicted surgical progress duration.

[0247] Further, the confidence level is calculated by the following formula:

[0248] α = 1 - |(target duration information - reference value) / reference value| × 100%;

[0249] Wherein, when the number of historical surgical samples corresponding to the operating surgeon in the operating surgeon information meets the preset quantity threshold, determining the reference value based on the average or median duration of multiple historical surgical samples of the operating surgeon; when the number of historical surgical samples corresponding to the operating surgeon in the operating surgeon information does not meet the preset quantity threshold, determining the reference value based on the average or median duration of multiple historical surgical samples of other operating surgeons with the same professional title as the operating surgeon.

[0250] Further, the device is further configured to: if receiving a first instruction indicating that an abnormal event has occurred, extracting first surgical information associated with the surgical name from the preset historical data; extracting duration adjustment information associated with the abnormal event from the first surgical information; adjusting the target duration information according to the duration adjustment information.

[0251] The prediction device for surgical duration and surgical progress duration provided by the embodiments of the present invention has the same implementation principle and the same technical effects as the foregoing embodiments of the prediction method for surgical duration and surgical progress duration. For the sake of brief description, for the parts not mentioned in the embodiments of the prediction device for surgical duration and surgical progress duration, reference may be made to the corresponding content in the foregoing embodiments of the prediction method for surgical duration and surgical progress duration.

[0252] Embodiments of the present invention further provide an electronic device. SeeFigure 5 As shown, the electronic device includes a processor 130 and a memory 131. The memory 131 stores machine-executable instructions that can be executed by the processor 130. The processor 130 executes the machine-executable instructions to implement the above-mentioned prediction method for the surgical duration and the surgical process duration.

[0253] Furthermore, Figure 5 The electronic device shown further includes a bus 132 and a communication interface 133. The processor 130, the communication interface 133, and the memory 131 are connected through the bus 132.

[0254] Among them, the memory 131 may include a high-speed random access memory (RAM, Random Access Memory), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 133 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 132 can be an ISA bus, a PCI bus, an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 5 only a bidirectional arrow is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0255] The processor 130 may be an integrated circuit chip with the ability to process signals. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in the processor 130 or the instructions in the form of software. The above-mentioned processor 130 may be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it may also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field-programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of the hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory 131, and the processor 130 reads the information in the memory 131 and combines its hardware to complete the steps of the method in the foregoing embodiments.

[0256] An embodiment of the present invention also provides a machine-readable storage medium, which stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions cause the processor to implement the above-mentioned prediction method for the surgical duration and the surgical process duration. For the specific implementation, reference can be made to the method embodiment, which will not be elaborated here.

[0257] The computer program product of the prediction method, device, and electronic device for the surgical duration and the surgical process duration provided by the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the foregoing method embodiments. For the specific implementation, reference can be made to the method embodiment, which will not be elaborated here.

[0258] When the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0259] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of various embodiments of the present invention.

Claims

1. A method for predicting the duration of an operation and the duration of an operation process, characterized in that: The method is applied to a server; the method comprises: Obtain the surgical application information corresponding to the target surgery; Extracting target influencing factors from the surgery application information; wherein the target influencing factors include: surgery name, patient age, patient gender, patient weight, patient disease severity data, surgical surgeon information, and patient underlying disease data; Based on the target influencing factors, determine the predicted operation time corresponding to the target operation; wherein the predicted operation time represents an estimated value of the time required from the start of the operation to the end of the operation; Obtain a first predicted duration, a second predicted duration, a third predicted duration, a fourth predicted duration, and a fifth predicted duration; wherein the first predicted duration is used to indicate an estimated value of the time required from the patient entering the room to the start of anesthesia; the second predicted duration is used to indicate an estimated value of the time required from the start of anesthesia to the start of surgery; the third predicted duration is used to indicate an estimated value of the time required from the end of surgery to the patient leaving the room; the fourth predicted duration is used to indicate an estimated value of the time required to clean the operating room after the end of surgery; the fifth predicted duration is used to indicate an estimated value of the time required from the completion of cleaning the operating room to the entry of the next patient into the room; The predicted operation duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration are added together to obtain the predicted operation process duration.

2. The method according to claim 1, characterized in that The step of determining the predicted operation duration corresponding to the target operation based on the target influencing factors includes: According to the operation name, a first parameter matching the operation name is obtained; wherein the first parameter is a first historical average operation time or a first median operation time determined according to preset historical data; According to the operation name and the operating physician information, a second parameter matching the operation name and the operating physician information is obtained; wherein the second parameter is the second historical average operation time or the second median operation time of the first operating physician performing the operation of the type of the operation name determined according to the historical data; the first operating physician is the operating physician corresponding to the operating physician information; Obtain the weight coefficient corresponding to each target influencing factor other than the operation name and the surgeon information, and for each other target influencing factor, calculate the product of the basic adjustment value of the target influencing factor and the weight coefficient corresponding to the target influencing factor to obtain the product result; add the target parameter and each of the multiplication results to obtain the predicted operation duration corresponding to the target operation; wherein the target parameter is the first parameter or the second parameter.

3. The method according to claim 2, characterized in that The server pre-stores a plurality of preset influencing factors, each of the preset influencing factors corresponds to a plurality of classification categories, and each of the classification categories has a corresponding adjustment coefficient; The step of obtaining the weight coefficient corresponding to each target influencing factor other than the operation name and the operation surgeon information includes: For each target influencing factor other than the operation name and the surgical surgeon information, obtain the factor value of the target influencing factor; Determine the target classification category corresponding to the target influencing factor according to the factor value; The adjustment coefficient corresponding to the target classification category is obtained, and the adjustment coefficient corresponding to the target classification category is determined as the weight coefficient corresponding to the target influencing factor.

4. The method according to claim 3, characterized in that The adjustment coefficient corresponding to each classification category is determined in advance in the following way: Obtaining surgical application sample information corresponding to the sample surgery; Extracting sample influencing factors from the surgical application sample information; wherein the sample influencing factors include: sample surgery name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, sample surgery surgeon information, and sample patient basic disease data; According to the sample operation name, a first sample parameter matching the sample operation name is obtained; wherein the first sample parameter is a first sample historical average operation time or a first sample operation time median determined according to preset historical data; According to the sample operation name and the sample operation surgeon information, a second sample parameter is obtained; wherein the second sample parameter is the second sample historical average operation time or the second sample operation median time determined according to the historical data; Obtaining the sample factor value corresponding to each sample influencing factor except the sample surgery name and the sample surgery surgeon information, and the initial sample coefficient corresponding to each sample factor value; For each other sample influencing factor, the product of the sample basic adjustment value of the sample influencing factor and the initial sample coefficient corresponding to the sample influencing factor is calculated to obtain a sample product result, and the target sample parameter and each sample product result are added to obtain the predicted sample surgery duration corresponding to the sample surgery; wherein the target sample parameter is the first sample parameter or the second sample parameter; Obtaining actual sample surgery duration corresponding to the sample surgery; Calculate the difference between the actual sample operation time and the predicted sample operation time to obtain a difference result; If the difference result does not meet the preset first error range, for each of the other sample influencing factors, the actual sample operation time is compared with the historical operation time data corresponding to the classification category to which the sample factor value of the sample influencing factor belongs, to obtain a comparison result; wherein the historical operation time data is: the historical average operation time of the sample corresponding to the classification category to which the sample factor value belongs or the median of the sample operation time; If the comparison result does not meet the preset second error range, obtain the coefficient adjustment instruction for the classification category, adjust the adjustment coefficient of the classification category according to the coefficient adjustment instruction, and repeat the step of obtaining the surgical application sample information corresponding to the sample surgery until the comparison result meets the preset second error range.

5. The method according to claim 1, characterized in that The step of determining the predicted operation duration corresponding to the target operation based on the target influencing factors includes: The factor value of each target influencing factor is input into a pre-trained surgery duration prediction model, so as to output the predicted surgery duration corresponding to the target surgery through the surgery duration prediction model.

6. The method according to claim 5, characterized in that The operation duration prediction model is trained in the following way: Obtaining surgical application sample information corresponding to the sample surgery and the actual duration of the surgery corresponding to the sample surgery; Extracting sample factor values ​​of sample influencing factors from the surgical application sample information; wherein the sample influencing factors include: sample surgery name, sample patient age, sample patient gender, sample patient weight, sample patient disease severity, sample surgery surgeon information, and sample patient basic disease data; Inputting each of the sample factor values ​​into the initial model, so as to output the predicted duration of the operation corresponding to the sample operation through the initial model; Based on the predicted surgery duration and the actual surgery duration, a loss value is calculated, and the parameters of the initial model are updated based on the loss value. The steps of obtaining the surgery application sample information corresponding to the sample surgery and the actual surgery duration corresponding to the sample surgery are repeated until the loss value converges to obtain a trained surgery duration prediction model.

7. The method according to claim 1, characterized in that The target surgery includes multiple sub-surgeries; The steps of obtaining the surgery application information corresponding to the target surgery include: For each of the sub-surgeries in the target surgery, obtaining surgery application information corresponding to the sub-surgery; The step of determining the predicted operation duration corresponding to the target operation based on the target influencing factors includes: For each of the sub-surgeries, based on the target influencing factors, determining a predicted sub-surgery duration corresponding to the sub-surgery; Obtaining a weight coefficient corresponding to each of the sub-surgeries; Based on each of the weight coefficients, a weighted calculation is performed on each of the predicted sub-surgery durations to obtain the predicted surgery duration corresponding to the target surgery.

8. The method according to claim 1, characterized in that The specified prediction duration is: the first prediction duration, the third prediction duration, or the fifth prediction duration; the specified prediction duration is determined in advance by: Obtain multiple time values ​​corresponding to the specified prediction duration within a preset historical time period; Extract the maximum and minimum values ​​from multiple time values; If the difference between the maximum value and the minimum value is less than a preset difference threshold, an average value of multiple time values ​​is calculated, and the average value is used as the designated predicted duration.

9. The method according to claim 2, characterized in that: The fourth predicted duration is determined in advance by: Determining the infection classification to which the target surgery belongs; Acquire historical cleaning durations respectively corresponding to a plurality of historical surgeries belonging to the infection classification from the historical data; Calculate a first calculation result of the plurality of historical cleaning durations; wherein the first calculation result is a median or a mean; The first calculation result is used as the fourth predicted duration.

10. The method according to claim 1, characterized in that The second predicted duration is determined in advance by: Extracting a first influencing factor from the surgery application information; wherein the first influencing factor includes: patient age, anesthesia method, preoperative risk assessment information, and patient underlying diseases; Based on the first influencing factor, the second predicted duration is determined.

11. The method according to claim 1, characterized in that: The method further comprises: Display the predicted operation duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration, and the fifth predicted duration; If an update instruction for the predicted operation duration is received, updating the predicted operation duration according to the update instruction; If an adjustment instruction for a target prediction duration is received, the target prediction duration is adjusted according to the adjustment instruction; wherein the target prediction duration includes at least one of the following: a first prediction duration, a second prediction duration, a third prediction duration, a fourth prediction duration and a fifth prediction duration.

12. The method according to claim 1, characterized in that The method further comprises: Display the confidence level for the target duration information; wherein the target duration information includes: predicted surgery duration, predicted surgery process duration.

13. The method according to claim 12, characterized in that The confidence level is calculated using the following formula: α=1-|(target duration information-reference value) / reference value|×100%; Among them, when the number of historical surgical samples corresponding to the surgeon in the surgeon information meets the preset number threshold, the benchmark value is determined based on the average duration or median of multiple historical surgical samples of the surgeon; when the number of historical surgical samples corresponding to the surgeon in the surgeon information does not meet the preset number threshold, the benchmark value is determined based on the average duration or median of multiple historical surgical samples of other surgeons with the same professional title as the surgeon.

14. The method according to claim 12, characterized in that The method further comprises: If a first instruction indicating that an abnormal event has occurred is received, extracting first operation information associated with the operation name from preset historical data; extracting duration adjustment information associated with the abnormal event from the first surgery information; The target duration information is adjusted according to the duration adjustment information.

15. A device for predicting the duration of an operation and the duration of an operation process, characterized in that: The device is arranged on a server; the device comprises: A first acquisition module is used to acquire surgery application information corresponding to a target surgery; An extraction module is used to extract target influencing factors from the surgery application information; wherein the target influencing factors include: surgery name, patient age, patient gender, patient weight, patient disease severity data, surgical surgeon information, and patient basic disease data; A determination module, configured to determine a predicted operation duration corresponding to the target operation based on the target influencing factors; wherein the predicted operation duration represents an estimated value of the time required from the start of the operation to the end of the operation; The second acquisition module is used to obtain a first predicted duration, a second predicted duration, a third predicted duration, a fourth predicted duration and a fifth predicted duration; wherein the first predicted duration is used to indicate an estimated value of the time required from the patient entering the room to the start of anesthesia; the second predicted duration is used to indicate an estimated value of the time required from the start of anesthesia to the start of surgery; the third predicted duration is used to indicate an estimated value of the time required from the end of surgery to the patient leaving the room; the fourth predicted duration is used to indicate an estimated value of the cleaning time of the operating room after the end of surgery; the fifth predicted duration is used to indicate an estimated value of the time required from the completion of cleaning of the operating room to the entry of the next patient into the room; A processing module is used to add the predicted operation duration, the first predicted duration, the second predicted duration, the third predicted duration, the fourth predicted duration and the fifth predicted duration to obtain a predicted operation process duration.

16. An electronic device, characterized in that: It includes a processor and a memory, the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the method for predicting the operation duration and the operation process duration described in any one of claims 1-14.

17. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores machine-executable instructions. When the machine-executable instructions are called and executed by the processor, the machine-executable instructions prompt the processor to implement the method for predicting the operation duration and the operation process duration described in any one of claims 1-14.

Images