Recommendation method and system of focused ultrasound operation plan and readable storage medium

Abstract

The invention provides a recommendation method and system for a focused ultrasound operation plan and a readable storage medium, and relates to the technical field of focused ultrasound, and the recommendation method comprises the steps: determining a plurality of candidate operation plans; the plurality of candidate operation schemes at least comprise historical operation schemes of a plurality of historical patients; constructing an initial training model based on the plurality of candidate operation schemes; training the initial training model based on clinical information and operation effects of a plurality of historical patients to obtain a probability prediction model; inputting the clinical information of the current patient and the expected operation effect into a probability prediction model, and predicting the probability of achieving the expected operation effect after the current patient is treated by the plurality of candidate operation schemes; and determining a recommended operation plan of the current patient based on the prediction result.

Description

[0001] Cross-references

[0002] This application claims priority to Chinese application No. 2024105116652 filed on April 26, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present application relates to the field of focused ultrasound technology, and in particular to a method, system and readable storage medium for recommending a focused ultrasound surgery plan. Background Art

[0004] For a long time, the rectum, prostate and other parts of the human body have been the hardest hit areas for malignant tumors. In recent years, with the increasing incidence of diseases such as rectal cancer and prostate cancer, focused ultrasound surgery has been developing as a means of clinical treatment of rectal cancer, prostate cancer and other diseases. Focused ultrasound surgery uses the heat generated by focused ultrasound to ablate the lesion areas in the patient's rectum, prostate and other parts, which can achieve the purpose of treating diseases such as rectal cancer and prostate cancer. At present, the formulation of focused ultrasound surgery plans mainly depends on the personal experience and technical level of the doctor, which leads to a high learning curve and risk for some doctors when performing surgery. Due to differences in the technical level of doctors, the accuracy, reliability and corresponding treatment effects of the surgical plan will also fluctuate, especially when they perform focused ultrasound surgery in the early stages. Doctors often find it difficult to tailor the best surgical plan for each patient, which may affect the postoperative treatment effect.

[0005] In view of this, it is hoped to provide a method for recommending focused ultrasound surgical plans, which aims to recommend surgical plans suitable for patients based on the actual conditions of patients with malignant tumors (e.g., patients with prostate cancer, patients with rectal cancer, etc.) for doctors' reference. This method can significantly reduce the time doctors spend on formulating surgical plans, improve doctors' work efficiency, and reduce the risk of poor patient prognosis caused by inaccurate surgical plans or unskilled operation. It can also ensure that the treatments received by different patients are more standardized and personalized, thereby improving the overall medical level and efficiency. Summary of the invention

[0006] One of the embodiments of the present application provides a method for recommending a focused ultrasound surgery plan, comprising: determining a plurality of candidate surgery plans; the plurality of candidate surgery plans at least include historical surgery plans of a plurality of historical patients; constructing an initial training model based on the plurality of candidate surgery plans; training the initial training model based on the clinical information and surgical effects of the plurality of historical patients to obtain a probability prediction model; inputting the clinical information and expected surgical effect of the current patient into the probability prediction model, and predicting the probability that the current patient will achieve the expected surgical effect after being treated with the plurality of candidate surgery plans respectively; and determining a recommended surgery plan for the current patient based on the prediction results.

[0007] In some embodiments, determining the plurality of candidate surgical plans comprises: determining the plurality of candidate surgical plans based on historical surgical plans of the plurality of historical patients.

[0008] In some embodiments, determining the plurality of candidate surgical plans based on the historical surgical plans of the plurality of historical patients includes: determining the historical surgical plan of each historical patient as the candidate surgical plan.

[0009] In some embodiments, determining the multiple candidate surgical plans based on the historical surgical plans of the multiple historical patients includes: establishing multiple historical surgical parameter data sets based on multiple historical surgical parameters in the multiple historical surgical plans, and the historical surgical parameter data sets include historical surgical parameters of the same type; wherein the multiple historical surgical parameters include at least: a working path of the ultrasound probe, a working point on the working path, a dwell time of the ultrasound probe at each working point, and a focused ultrasound intensity of the ultrasound probe at each working point; and determining the multiple candidate surgical plans based on the multiple historical surgical parameter data sets.

[0010] In some embodiments, determining multiple candidate surgical plans includes: acquiring a configurable data set of multiple surgical parameters of a focused ultrasound device; wherein the multiple surgical parameters include at least: a working path of the ultrasound probe, working points on the working path, a dwell time of the ultrasound probe at each working point, and a focused ultrasound intensity of the ultrasound probe at each working point; and determining the multiple candidate surgical plans based on the configurable data set of the multiple surgical parameters.

[0011] In some embodiments, the expected surgical effect includes that no diseased tissue is found in the medical images of the patient after surgery and the concentration of tumor markers is within the normal index range.

[0012] In some embodiments, the desired surgical effect also includes the patient experiencing no adverse events after surgery.

[0013] One of the embodiments of the present application provides a recommendation system for a focused ultrasound surgery plan, comprising: a determination module, the determination module is used to determine multiple candidate surgery plans; the multiple candidate surgery plans include at least historical surgery plans of multiple historical patients; a construction module, the construction module is used to construct an initial training model based on the multiple candidate surgery plans; a training module, the training module is used to train the initial training model based on the clinical information and surgical effects of the multiple historical patients to obtain a probability prediction model; a prediction module, the prediction module inputs the clinical information and expected surgical effect of the current patient into the probability prediction model, and predicts the probability that the current patient will achieve the expected surgical effect after being treated with the multiple candidate surgery plans respectively; a recommended surgery plan determination module, the recommended surgery plan determination module is used to determine the recommended surgery plan for the current patient based on the prediction result.

[0014] One of the embodiments of the present application provides a readable storage medium, in which a computer program is stored. When the computer program is executed, it is used to implement the following method: determine a plurality of candidate surgical plans; the plurality of candidate surgical plans include at least a plurality of historical surgical plans of historical patients; construct an initial training model based on the plurality of candidate surgical plans; train the initial training model based on the clinical information and surgical effects of the plurality of historical patients to obtain a probability prediction model; input the clinical information and expected surgical effect of the current patient into the probability prediction model, and predict the probability that the current patient will achieve the expected surgical effect after being treated with the plurality of candidate surgical plans; and determine the recommended surgical plan for the current patient based on the prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0016] Figure 1 is an exemplary flow chart of a recommended method for a focused ultrasound surgery plan according to some embodiments of the present application;

[0017] Figure 2 It is a block diagram of a recommendation system for focused ultrasound surgery plans according to some embodiments of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. On the contrary, the present application covers any substitutions, modifications, equivalent methods and schemes made on the essence and scope of the present application as defined by the claims. Further, in order to enable the public to have a better understanding of the present application, some specific details are described in detail in the detailed description of the present application below. For those skilled in the art, the present application can be fully understood without the description of these details.

[0019] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0020] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0021] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0022] During focused ultrasound surgery, the doctor will use an ultrasound probe to approach the lesion area in the rectum, prostate and other parts through a natural cavity (for example, the rectum). The ultrasound transducer on the ultrasound probe has ultrasound-imaging and treatment functions, so that the ultrasound probe can treat the tissue in the lesion area in the rectum, prostate and other parts under the guidance of ultrasound images.

[0023] When performing focused ultrasound surgery, it is generally performed according to a pre-established surgical plan. However, most current focused ultrasound surgery plans are generally formulated by doctors based on their own experience, judgment and personal level. First, doctors are limited by their experience, and the surgical plans they formulate may be inaccurate and unreliable, which in turn leads to poor prognosis for patients. At the same time, due to individual differences, doctors need to spend a lot of time formulating individualized and precise surgical plans, which is inefficient and may result in surgical plans that do not fully match the patient, resulting in poor postoperative treatment effects.

[0024] The embodiment of the present application provides a method for recommending a focused ultrasound surgery plan, the method comprising: determining multiple candidate surgery plans; the multiple candidate surgery plans at least include multiple historical surgery plans of historical patients; constructing an initial training model based on multiple candidate surgery plans; training the initial training model based on the clinical information and surgical effects of multiple historical patients to obtain a probability prediction model; inputting the clinical information and expected surgical effect of the current patient into the probability prediction model, predicting the probability that the current patient can achieve the expected surgical effect after being treated with multiple candidate surgery plans; and determining the recommended surgery plan for the current patient based on the prediction results. The method for recommending a focused ultrasound surgery plan provided in the embodiment of the present application can scientifically and accurately recommend a suitable surgery plan to a doctor based on the actual situation of patients with different prostate diseases. The surgery plan is accurate and reliable, which can not only shorten the time for doctors to formulate focused ultrasound surgery plans and reduce the burden on doctors, but also reduce the risk of poor prognosis for patients and improve the medical level and efficiency of focused ultrasound surgery. It should be noted that the applicable scenarios of the recommended method of the focused ultrasound surgery scheme provided in the embodiment of the present application can be but not limited to the focused ultrasound surgery for the treatment of rectal cancer and prostate cancer, and can also be the focused ultrasound surgery for the treatment of other diseases (for example, prostate hyperplasia, liver tumors, pancreatic cancer, soft tissue tumors, uterine fibroids, adenomyoma, thoracic and abdominal metastatic cancer, benign and malignant kidney tumors, primary or metastatic bone tumors, retroperitoneal or abdominal, pelvic solid tumors with good ultrasound channels). This application will mainly describe in detail the recommended method of the focused ultrasound surgery scheme for the treatment of rectal cancer and prostate cancer.

[0025] The recommended method of the focused ultrasound surgery solution provided in the embodiment of the present application will be described in detail below with reference to the accompanying drawings.

[0026] Figure 1 is an exemplary flow chart of a recommended method for a focused ultrasound surgery program according to some embodiments of the present application. Figure 1 The proposed method 100 of the focused ultrasound surgery protocol shown may be performed by Figure 2 The recommendation system 200 shown in FIG.

[0027] like Figure 1 As shown, the method 100 for recommending a focused ultrasound surgery solution provided in an embodiment of the present application may include the following steps:

[0028] Step S110, determining multiple candidate surgical plans. Specifically, step S110 may be performed by Figure 2 The determination module 210 in the recommendation system 200 shown in FIG.

[0029] In some embodiments, the candidate surgical plan may be a surgical plan that can perform focused ultrasound surgery. The plurality of candidate surgical plans may include at least a plurality of historical surgical plans of historical patients. In some embodiments, the historical patient refers to a rectal cancer patient or a prostate cancer patient who has undergone focused ultrasound surgery, and the historical surgical plan may refer to a surgical plan used by the historical patient for focused ultrasound surgery.

[0030] In some embodiments, multiple candidate surgical plans can be determined based on the historical surgical plans of multiple historical patients. In some embodiments, the historical surgical plan of each historical patient can be determined as a candidate surgical plan, that is, multiple candidate surgical plans can be composed of the historical surgical plans of multiple historical patients.

[0031] In some embodiments, multiple historical surgical parameter data sets may be established based on multiple historical surgical parameters in multiple historical surgical plans, and then multiple candidate surgical plans may be determined based on the multiple historical surgical parameter data sets. The historical surgical parameter data sets may include historical surgical parameters of the same type, and the historical surgical parameters may be surgical parameters used in the historical surgical plans.

[0032] In some embodiments, the multiple historical surgical parameters of the historical surgical plan may at least include the working path of the ultrasound probe, the working points on the working path, the dwell time of the ultrasound probe at each working point, and the focused ultrasound intensity of the ultrasound probe at each working point. Further, the multiple historical surgical parameter data sets may at least include the historical surgical parameter data set composed of the working path of the ultrasound probe in each historical surgical plan, the historical surgical parameter data set composed of the working points on the working path in each historical surgical plan, the historical surgical parameter data set composed of the dwell time of the ultrasound probe at each working point in each historical surgical plan, and the historical surgical parameter data set composed of the focused ultrasound intensity of the ultrasound probe at each working point in each historical surgical plan.

[0033] In some embodiments, multiple candidate surgical plans can be obtained by selecting a historical surgical parameter from each historical surgical parameter data set each time to combine them into a candidate surgical plan. As an exemplary illustration, there are M historical surgical plans, and there are N historical surgical parameters in the historical surgical plan, so there will be N historical surgical parameter data sets, and then (M×N) candidate surgical plans can be obtained. In some embodiments, considering that multiple historical surgical plans may have the same historical surgical parameters, that is, the same historical surgical parameter data set will have the same historical surgical parameter data, which will cause repeated candidate surgical plans to appear in multiple candidate surgical plans. Therefore, repeated candidate surgical plans need to be eliminated from multiple candidate surgical plans. It can be understood that since the historical surgical parameter data in the historical surgical parameter data set comes from the historical surgical plan, a combination of historical surgical parameters selected from each historical surgical parameter data set will obtain a candidate surgical plan that is the same as a certain historical surgical plan. Further, multiple candidate surgical plans will include candidate surgical plans that are the same as multiple historical surgical plans, which can also be understood as multiple candidate surgical plans including multiple historical surgical plans.

[0034] In some embodiments, a settable data set of multiple surgical parameters of the focused ultrasound device can be obtained, and then multiple candidate surgical plans can be determined based on the settable data sets of multiple surgical parameters. In some embodiments, the multiple surgical parameters may at least include a working path of the ultrasound probe, a working point on the working path, a dwell time of the ultrasound probe at each working point, and a focused ultrasound intensity of the ultrasound probe at each working point. Further, the settable data set of multiple surgical parameters of the focused ultrasound device may at least include a settable data set consisting of a working path of the ultrasound probe that can be set in the focused ultrasound device, a settable data set consisting of working points on the working path that can be set in the focused ultrasound device, a settable data set consisting of a dwell time of the ultrasound probe that can be set in the focused ultrasound device at each working point, and a settable data set consisting of a focused ultrasound intensity of the ultrasound probe that can be set in the focused ultrasound device at each working point.

[0035] In some embodiments, multiple candidate surgical plans can be obtained by selecting a surgical parameter from the settable data sets of various surgical parameters of the focused ultrasound device each time to combine them into a candidate surgical plan. As an exemplary illustration, the focused ultrasound device can set X types of surgical parameters, then there will be X settable data sets, and the X settable data sets respectively include n1, n2, ..., nX surgical parameter settable data, then (n1×nn2×...×nnX) candidate surgical plans can be obtained. It can be understood that since the historical surgical parameters in the historical surgical plan are set in the focused ultrasound device, the settable data will include the same surgical parameter settable data as the historical surgical parameters in the historical surgical plan, and combining a surgical parameter selected from the settable data sets of various surgical parameters of the focused ultrasound device each time will obtain a candidate surgical plan that is the same as a certain historical surgical plan. Further, multiple candidate surgical plans will include candidate surgical plans that are the same as multiple historical surgical plans, and it can also be understood that multiple candidate surgical plans include multiple historical surgical plans.

[0036] Step S120, constructing an initial training model based on multiple candidate surgical plans. Specifically, step S120 can be performed by Figure 2 The construction module 220 in the recommendation system 200 shown in FIG.

[0037] In some embodiments, the initial training model may be a neural network model. In some embodiments, the number of neuron nodes in the output layer of the initial training model may be set based on multiple candidate surgical plans, that is, each neuron node in the output layer of the initial training model may correspond to a candidate surgical plan, that is, each neuron node in the output layer may be used to output the probability that the patient can achieve the desired surgical effect after being treated with the candidate surgical plan corresponding to the neuron node.

[0038] Step S130, training the initial training model based on the clinical information and surgical results of multiple historical patients to obtain a probability prediction model. Specifically, step S130 can be performed by Figure 2 The training module 230 in the recommendation system 200 shown in FIG.

[0039] In some embodiments, clinical information may include personal information and preoperative examination information of the patient. In some embodiments, the patient's personal information may include age, height, weight, medical history, etc. or a combination thereof.

[0040] In some embodiments, the patient's preoperative examination information may include the results of blood routine examination, urine routine examination, liver and kidney function examination, coagulation function examination, puncture biopsy results of the lesion area, tumor marker measurement results, and preoperative medical imaging. In some embodiments, for patients with rectal cancer, tumor markers may include peripheral blood carcinoembryonic antigen (CEA), carbohydrate antigen (CA) n 19-9, etc. In some embodiments, for prostate patients, tumor markers may include prostate specific antigen (PSA). In some embodiments, for patients with rectal cancer, the preoperative examination information may also include the results of routine stool examinations, the results of fecal occult blood tests, the results of anorectal manometry, the results of defecography, the results of pelvic floor MRI, the results of endoscopic ultrasound, the results of urodynamic examinations (including urine flow rate diagrams, urethral pressure diagrams, volume pressure diagrams of bladder filling and emptying, electromyography, etc.), the results of nocturnal penile swelling experiments, and cavernous electromyography (suitable for male sexual function assessment), etc.

[0041] In some embodiments, the patient's preoperative examination information may also include imaging results obtained from preoperative medical images. In some embodiments, the preoperative medical images may include at least one of transrectal ultrasound images, magnetic resonance images, and computer tomography images. In some embodiments, for patients with rectal cancer, the imaging results obtained from preoperative medical images may include the size and location of the lesion area, the degree of external invasion of rectal cancer, lymph node metastasis, tumor infiltration depth, etc. In some embodiments, for patients with prostate cancer, the imaging results obtained from preoperative medical images may include residual urine volume in the bladder, prostate volume, lesion area size, cysts / calcifications in the prostate cavity, etc.

[0042] In some embodiments, preoperative information may also include information assessing the severity of the patient's cancer or assessing physiological functions related to the cancer site (e.g., the rectum or prostate) based on the patient's preoperative examination information, which may be specifically expressed as a grade, stage or score determined by a corresponding scale.

[0043] In some embodiments, for patients with rectal cancer, preoperative information may also include information for evaluating the pelvic organ function of patients with rectal cancer. In some embodiments, pelvic organ function may include defecation function, urination function and sexual function. In some embodiments, the scale for evaluating defecation function may include Parks scale, Wexner score, MSK-BFI score, Womack grading scale, Kirwan functional grading, Williams standard and low anterior resection syndrome (LARS) score, etc. In some embodiments, the scale for evaluating urination function may use the international prostate symptom score (IPSS), which is applicable to both men and women. In some embodiments, the scales for assessing female urinary function may also include the International Incontinence Modular Questionnaire-Female Lower Urinary Tract Symptoms (ICIQ-FLUTS) and the Bristol Female Lower Urinary Tract Symptoms Score (Bristol Female Lower Urinary Tract Symptom Questionnaire, BFLUTS), etc. In some embodiments, the scale for assessing male sexual function may use the International Index of Erectile Function (IIEF), and the scale for assessing female sexual function may use the Female Sexual Function Index (FSFI).

[0044] In some embodiments, for patients with rectal cancer, preoperative information may include information for assessing the severity of cancer in patients with rectal cancer. In some embodiments, commonly used cancer grading can be used to assess the severity of cancer in patients with rectal cancer (e.g., the malignancy of the tumor). Specifically, the abnormality of the histological characteristics of rectal tumor cells and tissues of patients with rectal cancer observed under a microscope compared to normal cells and tissues can be used as the basis for rectal cancer grading. As an exemplary illustration, rectal tumors can be divided into grades 1-4, represented by G1, G2, G3 and G4, respectively, wherein G1n represents a well-differentiated tumor, which is considered to be a low grade with a low degree of malignancy; G2 represents a moderately differentiated tumor, which is considered to be a medium grade with a relatively high degree of malignancy; G3 represents a poorly differentiated tumor, which is considered to be a high grade with a higher degree of malignancy; G4 represents an undifferentiated tumor, which is considered to be the highest grade with the highest degree of malignancy.

[0045] In some embodiments, the TNM staging system can be used to stage rectal cancer to assess the severity of cancer in patients with rectal cancer. Specifically, in the TNM staging system:

[0046] T stands for the primary tumor, which describes the extent of cancer invasion into the five layers of the intestinal wall and may include: Tx, the primary tumor cannot be evaluated; T0, no evidence of a primary tumor; Tis, carcinoma in situ, intramucosal carcinoma (the tumor invades the lamina propria but does not break through the muscularis mucosa); T1, the tumor invades the submucosal layer (the tumor invades the submucosal layer but does not involve the muscularis propria); T2, the tumor invades the muscularis propria; T3, the tumor penetrates the muscularis propria to reach the para-colorectal tissue; T4, the tumor invades the visceral peritoneum or invades or adheres to adjacent organs or structures; T4a, the tumor penetrates the visceral peritoneum (including visible intestinal perforation at the tumor site and continuous infiltration of the tumor through the inflammatory area to the surface of the visceral peritoneum); T4b, the tumor directly invades or adheres to adjacent organs or structures; nN stands for regional lymph nodes.

[0047] N stands for regional lymph nodes, which may include: Nx, regional lymph nodes cannot be evaluated; N0, no regional lymph node metastasis; N1, 1-3 regional lymph node metastasis (tumor diameter in the lymph node ≥0.2 mm), or no regional lymph node metastasis but any number of tumor nodules (tumor deposit, TD); N1a, 1 regional lymph node metastasis; N1b, 2-3 regional lymph node metastasis; N1c, no regional lymph node metastasis, but tumor nodules in the subserosal, mesenteric, or perirectal tissue without peritoneal coverage; N2, 4 or more regional lymph node metastasis; N2a, 4-6 regional lymph node metastasis; N2b, ≥7 regional lymph node metastasis.

[0048] M stands for distant metastasis, which means that the tumor has spread to other sites, and may include: Mx, distant metastasis cannot be evaluated; M0, no distant metastasis on imaging examination, that is, no evidence of metastatic tumor in distant sites and organs (this classification should not be determined by pathologists); M1, the presence of metastasis to one or more distant sites, organs or peritoneum; M1a, distant metastasis is limited to a single distant site or organ, but there is no peritoneal metastasis; M1b, distant metastasis is distributed in two or more distant sites or organs, without peritoneal metastasis M1c, peritoneal metastasis, with or without metastasis to other sites or organs; M1C, peritoneal metastasis with or without metastasis to other organs.

[0049] In some embodiments, T, N, and M in the TNM staging system can be determined based on the biopsy results of the tumor. After T, N, and M are determined, this information can be combined and represented by the number 0 and Roman numerals I to IV to represent a stage, which is called a stage combination. TNM staging combinations can include:

[0050] Stage 0: The cancer is found only in the lining of the rectum. This is also called carcinoma in situ.

[0051] Stage I (T1 / T2, N0, M0): The tumor has grown into the inner wall of the rectum but has not grown beyond the intestinal wall.

[0052] Stage II (T3 / T4a / T4b, N0, M0): The tumor has grown deeper into or has penetrated the rectal wall and may have invaded nearby tissues, but the cancer cells have not spread to the lymph nodes.

[0053] Stage III: Cancer has spread to nearby lymph nodes but not to other parts of the body. Stage III is further divided into IIIA, IIIB, and IIIC. IIIA: (T1 / T2, N1 / N1c, M0) or (T1, N2a, M0); IIIB: (T3 / T4a, N1 / N1c, M0), (T2 / T3, N2a, M0) or (T1 / T2, N2b, M0); IIIC: (T4a, N2a, M0), (T3 / T4a, N2b, M0) or (T4b, N1 / N2, M0).

[0054] Stage IV: The cancer has spread to other parts of the body, such as the liver or lungs. Stage IV can be further divided into IVA, IVB, and IVC. IVA: (any T, any N, M1a); IVB: (any T, any N, M1b); IVC: (any T, any N, M1c).

[0055] In some embodiments, for patients with prostate cancer, the score used to evaluate the severity of prostate cancer can be the Gleason score, which can be obtained based on the results of the puncture biopsy of the lesion area. Further, the Gleason score is calculated by adding the major morphological grade and the minor morphological grade, wherein the major morphological grade refers to the Gleason grade at which more than 50% of the tumor cells in the puncture biopsy sample of the lesion area are located under a microscope, and the minor morphological grade refers to the Gleason grade at which 5% to 49% of the tumor cells in the puncture biopsy sample of the lesion area are located under a microscope, wherein Gleason grade 1 is the lowest score, representing that the tumor cells are well differentiated and have a low degree of malignancy, and Gleason grade 5 is the highest score, representing that the differentiation is very poor and the malignancy is high.

[0056] In some embodiments, since focused ultrasound surgery can also be used to treat prostate hyperplasia, the method provided in the embodiment of the present application can also be applied to the recommendation of focused ultrasound surgery schemes for treating prostate hyperplasia. Further, when the patient is a patient with prostate hyperplasia, the score used to evaluate the severity of prostate hyperplasia symptoms can be the International Prostate Symptom Score (IPSS). The International Prostate Symptom Score of 0 to 35 points represents symptoms ranging from asymptomatic to severe, divided into three levels: mild, moderate, and severe. Among them, 1 to 7 points are mild, 8 to 19 points are moderate, and 20 to 35 points are severe. In some embodiments, the International Prostate Symptom Score can be obtained by a doctor asking the patient. For example, the doctor can ask the patient whether he often has a sense of incomplete urination in the past month, whether the time between two urinations is often less than two hours, whether there has been intermittent urination, whether there is a phenomenon of being unable to wait for urination, whether there is often a phenomenon of thinning urine lines, whether it is often necessary to use force and force to start urination, and how many times you need to get up to urinate from falling asleep to getting up early. Symptoms such as the International Prostate Symptom Score are comprehensively obtained.

[0057] In some embodiments, the surgical effect may include whether diseased tissue appears in the patient's postoperative medical images. As an example, for patients with rectal cancer, the surgical effect may include whether the rectal cancer patient undergoes colonoscopy within 1 year after surgery and rechecks within 3 years to see if polyps appear. In some embodiments, the surgical effect may include whether the patient's postoperative tumor marker concentration is within a normal range. Among them, for patients with rectal cancer, the surgical effect may include whether the patient's postoperative blood carcinoembryonic antigen (CEA) and carbohydrate antigen (CA19-9) concentrations are within a normal range. For patients with prostate cancer, the surgical effect may include whether the patient's postoperative prostate-specific antigen (PSA) concentration is within a normal range. In some embodiments, the surgical effect may also include whether the patient has adverse events after surgery, wherein the adverse event may refer to complications or organ dysfunction after focused ultrasound surgery. In some embodiments, for patients with rectal cancer, adverse events after surgery may include organ dysfunction that occurs after surgery, and organ dysfunction may include defecation dysfunction, urination dysfunction, and sexual dysfunction. In some embodiments, the manifestation of defecation dysfunction may include one or more of urgency of defecation, increased frequency of defecation, aggregation of defecation, incontinence of defecation, constipation, difficulty in defecation, reduced frequency of defecation, alternation of incontinence of defecation and constipation, etc. In some embodiments, the manifestation of urinary dysfunction may include one or more of urinary retention, urinary incontinence and urinary difficulty. In some embodiments, the manifestation of sexual dysfunction may be divided into male and female, wherein the manifestation of male sexual dysfunction may include one or more of decreased libido, decreased sexual activity, erectile dysfunction, premature ejaculation and delayed ejaculation, etc.; the manifestation of female sexual dysfunction may include one or more of sexual desire and sexual arousal disorders, orgasm disorders and sexual intercourse pain, etc. In some embodiments, for patients with prostate cancer, postoperative adverse events may include urinary retention, proteinuria, hematospermia, hematuria, retrograde ejaculation, urinary tract infection, intestinal wall damage (enteric heat), sexual dysfunction such as impotence, urinary difficulty, pain and discomfort, urinary incontinence, osteomyelitis, bladder neck contracture, etc. or a combination thereof. It is understandable that, since the historical patients have undergone focused ultrasound surgery, the historical surgical plans and corresponding surgical results of the historical patients are known and available.

[0058] In step S130, the neural network model may be trained for T rounds based on the clinical information and surgical effects of multiple historical patients to obtain a probability prediction model.

[0059] In each round of training (the current round of training), the clinical information and surgical outcomes of each historical patient are input into the training model obtained from the previous round of training to obtain the output results corresponding to each historical patient in the current round of training. Then, the output results corresponding to each historical patient in the current round of training are compared with the actual results to obtain the comparison results corresponding to each historical patient in the current round of training. Then, based on the comparison results corresponding to each historical patient, the loss values corresponding to each historical patient in the current round of training are calculated through a preset loss function, and the minimum loss value in the current round of training is determined. Then, based on the minimum loss value in the current round of training, the training model obtained from the previous round of training is updated, and the updated training model obtained from the previous round of training can be used as the training model obtained in the current round of training. Among them, the training model obtained in each round of training will be saved. When the initial training model completes T rounds of training, if the minimum loss value in the Kth (K < T) round of training is the minimum among the minimum loss values in all rounds of training, then the training model obtained in the Kth round of training can be determined as the probability prediction model. It can be understood that the number of rounds of training T for the training model can be determined according to the actual situation. For example, T can be 100, 200, 300, etc. In some embodiments, the more complex the training model and the more training samples, the more rounds of training T for the training model.

[0060] In some embodiments, the output results corresponding to the historical patient may include the prediction results of the probabilities that the historical patient reaches the surgical outcome after being treated with each candidate surgical plan. In some embodiments, the prediction results of the probabilities that the historical patient reaches the surgical outcome after being treated with each candidate surgical plan can be represented by a feature vector. Among them, each eigenvalue in the vector can respectively represent the probability that the historical patient reaches the surgical outcome after being treated with each candidate surgical plan. As an example, the prediction results of the probabilities that the historical patient reaches the surgical outcome after being treated with each candidate surgical plan can be expressed as (S1, S2, S3,..., Sn), where n is the number of candidate surgical plans, and S1, S2, S3,..., Sn respectively represent the probabilities that the historical patient reaches the surgical outcome after being treated with the first candidate surgical plan, the second candidate surgical plan, the third candidate surgical plan,..., the nth candidate surgical plan.

[0061] In some embodiments, the actual result corresponding to the historical patient may refer to the probability that the historical patient actually achieves the surgical effect of the historical patient after being treated with each candidate surgical plan. In reality, the historical patient has only been treated with his historical surgical plan and achieved his surgical effect, and has not been treated with other candidate surgical plans other than his historical surgical plan. Therefore, in the actual result corresponding to the historical patient, the probability that the historical patient achieves his surgical effect through the candidate surgical plan that is the same as his historical surgical plan is 1, and the probability of achieving his surgical effect through other candidate surgical plans is 0. As an example, if the historical surgical plan of the historical patient is the second candidate surgical plan among multiple candidate surgical plans, the actual result corresponding to the historical patient can be expressed as (0, 1, 0,..., 0).

[0062] In some embodiments, when updating the training model, the various parameters within the training model can be updated by back-propagating the minimum loss value into the training model to reduce the loss between the output result of the training model and the actual result, so that the output result of the training model can be closer to the actual result, thereby achieving the purpose of training.

[0063] Step S140, input the clinical information and expected surgical effect of the current patient into the probability prediction model, and predict the probability that the current patient can achieve the expected surgical effect after being treated with multiple candidate surgical plans. Figure 2 The prediction module 240 in the recommendation system 200 shown in FIG.

[0064] In some embodiments, the expected surgical effect may refer to the surgical effect that the patient expects to obtain after the focused ultrasound surgery. In some embodiments, the expected surgical effect may include that no diseased tissue appears in the patient's postoperative medical images. In some embodiments, the expected surgical effect may include that the patient's postoperative tumor markers are within the normal range. In some embodiments, the expected surgical effect may also include that the patient has no adverse events after surgery.

[0065] In step S140, the clinical information and expected surgical results of the current patient may be input into the probability prediction model, and the probability prediction model may output the prediction results of the probability of the current patient achieving the expected surgical effect after being treated with multiple candidate surgical plans.

[0066] Step S150, determining the recommended surgical plan for the current patient based on the prediction result. Specifically, step S130 may be performed by Figure 2 The recommended surgical plan determination module 250 in the recommendation system 200 shown in FIG.

[0067] In some embodiments, in step S150, the candidate surgical plan corresponding to the maximum probability in the prediction result can be determined as the recommended surgical plan for the current patient. For example, in the prediction result, if the probability of the current patient achieving the desired surgical effect after treatment with the third candidate surgical plan is the highest, the third candidate surgical plan can be determined as the recommended surgical plan for the current patient. After the recommended surgical plan is determined, the recommended surgical plan can be recommended to the doctor, so that the doctor can formulate or optimize a suitable, accurate and reliable focused ultrasound surgical plan for the current patient based on or with reference to the recommended surgical plan. This can not only shorten the time for the doctor to formulate the focused ultrasound surgical plan and reduce the burden on the doctor, but also reduce the risk of poor prognosis for the patient and improve the medical level and efficiency of focused ultrasound surgery.

[0068] The embodiment of the present application also provides a readable storage medium, which stores a computer program. When the computer program is executed, it can be used to implement the recommended method of the focused ultrasound surgery plan provided in the embodiment of the present application.

[0069] Figure 2 It is a block diagram of a recommendation system for focused ultrasound surgery plans according to some embodiments of the present application.

[0070] like Figure 2 As shown, the embodiment of the present application also provides a recommendation system 200 for a focused ultrasound surgery plan, and the recommendation system 200 can be used to execute the computer program in the readable storage medium provided by the present application. Further. The recommendation system 200 may include: a determination module 210, a construction module 220, a training module 230, a prediction module 240, and a recommended surgery plan determination module 250. Among them, the determination module 210 is used to determine multiple candidate surgery plans; the multiple candidate surgery plans at least include historical surgery plans of multiple historical patients; the construction module 220 is used to construct an initial training model based on multiple candidate surgery plans; the training module 230 is used to train the initial training model based on the clinical information and surgical effects of multiple historical patients to obtain a probability prediction model; the prediction module 240 inputs the clinical information and expected surgical effect of the current patient into the probability prediction model, and predicts the probability of the current patient achieving the expected surgical effect after being treated with multiple candidate surgery plans; the recommended surgery plan determination module 250 is used to determine the recommended surgery plan for the current patient based on the prediction results.

[0071] In some embodiments, the determination module 210 can determine multiple candidate surgical plans based on the historical surgical plans of multiple historical patients. In some embodiments, the determination module 210 can determine the historical surgical plan of each historical patient as a candidate surgical plan. In some embodiments, the determination module 210 can establish multiple historical surgical parameter data sets based on multiple historical surgical parameters in multiple historical surgical plans, and the historical surgical parameter data sets include historical surgical parameters of the same type; wherein the multiple historical surgical parameters at least include: the working path of the ultrasound probe, the working point on the working path, the residence time of the ultrasound probe at each working point, and the focused ultrasound intensity of the ultrasound probe at each working point; then determine multiple candidate surgical plans based on multiple historical surgical parameter data sets.

[0072] In some embodiments, the determination module 210 can obtain a configurable data set of multiple surgical parameters of the focused ultrasound device, wherein the multiple surgical parameters include at least: the working path of the ultrasound probe, the working point on the working path, the residence time of the ultrasound probe at each working point, and the focused ultrasound intensity of the ultrasound probe at each working point, and then determine multiple candidate surgical plans based on the configurable data set of multiple surgical parameters.

[0073] It should be noted that the above description of the determination system and its modules is only for the convenience of description and does not limit the present specification to the scope of the embodiments. It is understandable that, after understanding the principle of the system, those skilled in the art may arbitrarily combine the modules or form a subsystem to connect with other modules without deviating from the principle. In some embodiments, Figure 2 The construction module 220 and the training module 230 disclosed in the specification can be different modules in a system, or one module can realize the functions of two or more modules. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the scope of protection of this specification.

[0074] The beneficial effects that may be brought about by the embodiments of the present application include but are not limited to: (1) The recommended surgical plan determined by the method for recommending a focused ultrasound surgical plan provided by the embodiments of the present application has high accuracy and reliability, and can reduce the risk of poor prognosis for patients; (2) The recommended surgical plan is determined by predicting the probability of a patient achieving the desired surgical effect after being treated with each candidate surgical plan through a probability prediction model, which can shorten the time for doctors to formulate surgical plans, reduce the burden on doctors, and is conducive to improving the medical level and efficiency of focused ultrasound surgery.

[0075] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other beneficial effects that may be obtained.

[0076] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this application. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this application.

[0077] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the embodiments of the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the embodiments of the present invention may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0078] The embodiments of the present application are described with reference to flowcharts and / or block diagrams of methods, terminal devices (systems) or computer program products provided according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0079] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0080] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0081] In addition, when terms such as "first", "second", and "third" are used in the specification of this application to describe various features, these terms are only used to distinguish these features and cannot be understood as indicating or implying the relationship between the features, the relative importance, or implicitly indicating the number of features indicated.

[0082] At the same time, the present application uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in the present application does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0083] Similarly, it should be noted that in order to simplify the description of the disclosure of this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of this application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0084] Finally, it should be understood that the embodiments described in this application are only used to illustrate the principles of the embodiments of the present application. Other variations may also fall within the scope of the present application. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of the present application may be considered to be consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to the embodiments explicitly introduced and described in the present application.

Claims

1. A method for recommending a focused ultrasound surgery plan, characterized in that: include: Determine a plurality of candidate surgical plans; the plurality of candidate surgical plans at least include historical surgical plans of a plurality of historical patients; Building an initial training model based on the multiple candidate surgical plans; Training the initial training model based on the clinical information and surgical results of the multiple historical patients to obtain a probability prediction model; Inputting the clinical information and expected surgical effect of the current patient into the probability prediction model, and predicting the probability of the current patient achieving the expected surgical effect after being treated with the multiple candidate surgical plans; A recommended surgical plan for the current patient is determined based on the prediction results.

2. The method according to claim 1, characterized in that The step of determining a plurality of candidate surgical options comprises: The plurality of candidate surgical plans are determined based on the historical surgical plans of the plurality of historical patients.

3. The method according to claim 2, characterized in that The determining the plurality of candidate surgical plans based on the historical surgical plans of the plurality of historical patients comprises: A historical surgical plan for each historical patient is determined as the candidate surgical plan.

4. The method according to claim 2, characterized in that: The determining the plurality of candidate surgical plans based on the historical surgical plans of the plurality of historical patients comprises: Establishing multiple historical surgical parameter data sets based on multiple historical surgical parameters in the multiple historical surgical plans, wherein the historical surgical parameter data sets include historical surgical parameters of the same type; wherein the multiple historical surgical parameters at least include: a working path of the ultrasound probe, working points on the working path, a dwell time of the ultrasound probe at each working point, and a focused ultrasound intensity of the ultrasound probe at each working point; The plurality of candidate surgical plans are determined based on the plurality of historical surgical parameter data sets.

5. The method according to claim 1, characterized in that The step of determining a plurality of candidate surgical options comprises: Acquire a configurable data set of multiple surgical parameters of a focused ultrasound device; wherein the multiple surgical parameters at least include: a working path of the ultrasound probe, working points on the working path, a dwell time of the ultrasound probe at each working point, and a focused ultrasound intensity of the ultrasound probe at each working point; The plurality of candidate surgical plans are determined based on the configurable data set of the plurality of surgical parameters.

6. The method according to claim 1, characterized in that The expected surgical effect includes that no diseased tissue is found in the medical images of the patient after the operation and the tumor marker concentration is within the normal range.

7. The method according to claim 1, characterized in that The expected surgical effect also includes that the patient experiences no adverse events after the operation.

8. A focused ultrasound surgery plan recommendation system, characterized in that: include: A determination module, the determination module is used to determine a plurality of candidate surgical plans; the plurality of candidate surgical plans at least include a plurality of historical surgical plans of historical patients; A construction module, wherein the construction module is used to construct an initial training model based on the multiple candidate surgical plans; A training module, wherein the training module is used to train the initial training model based on the clinical information and surgical effects of the multiple historical patients to obtain a probability prediction model; A prediction module, wherein the prediction module inputs the clinical information and expected surgical effect of the current patient into the probability prediction model, and predicts the probability of the current patient achieving the expected surgical effect after being treated with the multiple candidate surgical plans; A recommended surgical plan determination module is used to determine a recommended surgical plan for the current patient based on the prediction results.

9. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which, when executed, is used to implement the following method: Determine a plurality of candidate surgical plans; the plurality of candidate surgical plans at least include historical surgical plans of a plurality of historical patients; Building an initial training model based on the multiple candidate surgical plans; Training the initial training model based on the clinical information and surgical results of the multiple historical patients to obtain a probability prediction model; Inputting the clinical information and expected surgical effect of the current patient into the probability prediction model, and predicting the probability of the current patient achieving the expected surgical effect after being treated with the multiple candidate surgical plans; A recommended surgical plan for the current patient is determined based on the prediction results.

Images