DJ tube wall stone detection system and detection method

Abstract

The present invention belongs to the field of medical image-assisted diagnosis, and in particular, relates to a DJ tube wall stone detection system and detection method. The DJ tube wall stone detection system includes a data input module for receiving the patient's biochemical data; a stone prediction module for predicting the probability of the patient having DJ tube wall stones according to the biochemical data received by the data input module, and recommending a DJ tube removal plan to the patient according to the probability of DJ tube wall stones. The detection system and detection method can determine the probability of the patient's DJ tube wall stones according to the patient's biochemical data, so that the patient can know the probability of the stone without taking a CT scan, avoiding the patient's blind medical treatment, and reducing the patient's burden and pain.

Description

Technical Field

[0001] The present invention belongs to the field of medical image-assisted diagnosis, and in particular relates to a DJ tube wall stone detection system and a detection method. Background Art

[0002] The human urinary system refers to the kidneys, ureters, and bladder. After the patient has undergone lithotripsy, a urinary tract stent (DJ tube stent) will be placed in the human body and removed after a period of time. Due to the patient's physical condition and the placement time of the DJ tube, stones may be attached to the DJ tube. If the DJ tube is attached with stones, it cannot be directly removed in the outpatient clinic and requires hospitalization for surgery to remove it. However, the patient cannot be sure whether there are stones attached to the DJ tube before the tube is removed. It is often removed directly in the outpatient clinic. Only when it cannot be removed will the patient be hospitalized for surgery. After hospitalization, a CT scan will be taken, and the doctor needs to carefully observe the condition of the DJ tube wall stones before performing surgery. The patient endures great pain during this process.

[0003] Patent number CN112263269A discloses an intelligent detection system and method for urinary system X-ray plain film stones. The system mainly analyzes X-ray images, detects the presence of stones in the urinary system, and determines other misdiagnosis factors, such as surgical repair / support, etc. The system intelligently detects whether urinary system stones exist, but cannot determine whether there are stones on the DJ tube stent, nor can it determine the presence of stones on the DJ tube stent. It only replaces manual reading.

[0004] Patent number CN111340130A discloses a urinary stone detection and classification method based on deep learning and radiomics. The patent extracts the region of interest of urinary stones through a deep learning model, and then extracts the image group features of the stones to determine the stone category. Similarly, the patent does not analyze and judge the situation of DJ tube stent wall stones.

[0005] In view of this, the present invention is proposed. Summary of the invention

[0006] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a DJ tube wall stone detection system and detection method. The detection system and detection method can judge the probability of DJ tube wall stones in patients according to the biochemical data of the patients, so that patients can know the probability of the existence of stones without taking CT scans, avoiding blind medical treatment for patients and reducing the burden and pain of patients.

[0007] In order to solve the above technical problems, the present invention provides a DJ tube wall stone detection system, comprising:

[0008] A data input module, for receiving biochemical data of the patient;

[0009] The stone prediction module is used to predict the probability of DJ tube mural stones in the patient based on the biochemical data received by the data input module, and recommend a DJ tube removal plan to the patient based on the probability of DJ tube mural stones.

[0010] Further optionally, in the stone prediction module,

[0011] A character recognition algorithm is used to identify and save highly correlated indicators, which are then input into a logistic regression algorithm model. Based on the trained model, a prediction is made as to whether the patient has DJ tube mural stones. The probability of the presence of DJ tube mural stones is output, and a DJ tube removal plan is recommended to the user based on the probability of the presence of DJ tube mural stones.

[0012] Further optionally, the data input module also receives CT image data, and the DJ tube wall stone detection system also includes

[0013] The stone detection module is used to analyze the severity of DJ tube mural stones in patients based on the CT image data received by the data input module.

[0014] Further optionally, in the stone detection module,

[0015] The DJ tube and stone areas in the CT image are segmented, and the segmentation results are filled and corrected to ensure that the DJ tube and wall stone areas can be segmented in each CT image; the segmented area coordinates are mapped to the original image to obtain the DJ tube ROI area in the original image; the DJ tube ROI area in the original image is then threshold segmented to obtain an image containing only the DJ tube and wall stones; the grayscale co-occurrence matrix and Fourier descriptor features in the DJ tube and wall stone images are extracted, and the stones are classified based on the extracted features to determine whether there are wall stones in the segmentation results of each CT. If so, the number, area and relative position of the wall stones are calculated, and the severity of the DJ tube wall stones is distinguished according to the number and area of ​​all stones.

[0016] Further optionally, a neural network is used to segment the DJ tube and stone areas in the CT image.

[0017] Further optionally, SVM or random forest algorithm is used to classify stones based on the extracted features.

[0018] Further optionally, the DJ tube wall stone detection system also includes

[0019] The structured report output module is used to display all the patient's disease diagnosis results and output them in the form of a structured report.

[0020] The present invention also proposes a method for detecting DJ tube wall stones, which uses the DJ tube wall stone detection system according to any one of claims 1 to 7. The detection method includes:

[0021] Obtain the patient's biochemical data;

[0022] The probability of DJ tube mural stones in the patient is predicted based on biochemical data, and a DJ tube removal plan is recommended to the patient based on the probability of DJ tube mural stones.

[0023] Further optionally, when the probability of DJ tube mural stones in the patient is greater than the set probability, surgical treatment is recommended for the patient; when the probability of DJ tube mural stones in the patient is less than or equal to the set probability, outpatient treatment is recommended for the patient.

[0024] Further optionally, when the probability of DJ tube mural stones in the patient is greater than a set probability, the method further includes:

[0025] Acquire CT image data of the patient;

[0026] The severity of DJ tube mural stones in patients was analyzed based on CT image data.

[0027] Further optionally, after acquiring the CT image data of the patient, it is also determined whether the patient has a single DJ tube or a double DJ tube.

[0028] When it is determined that the patient has a single DJ tube, the severity of the DJ tube mural stones in the patient is analyzed directly based on the CT image data;

[0029] When it is determined that the patient has double DJ tubes, the CT image is vertically cropped from the center point, and the two cropped images are used as the single DJ tube case. The severity of the DJ tube mural stones in the patient is analyzed for the two cropped images.

[0030] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0031] The DJ tube wall stone detection system proposed in the present invention has a complete DJ tube wall stone judgment and detection process. First, by analyzing the patient's biochemical data, the patient's high correlation indicators are extracted, and then a machine learning classification algorithm such as logistic regression is used to classify the high correlation indicators to determine whether there are wall stones in the DJ tube in the body, and the probability of stone existence is output. Doctors can recommend patients for outpatient tube removal or hospitalization based on the probability of stones, which largely avoids the pain of patients who cannot remove the DJ tube and more treatment costs.

[0032] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings are part of the present invention and are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the drawings described below are only some embodiments. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the accompanying drawings:

[0034] Figure 1 : is an overall flow chart of the DJ tube wall stone detection system according to an embodiment of the present invention.

[0035] Figure 2 : Workflow diagram of the stone prediction module.

[0036] Figure 3 : Workflow diagram of the stone detection module.

[0037] Figure 4 : The process of acquiring the image for DJ tube.

[0038] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0039] In the description of the present invention, it should be noted that the terms "inside", "outside", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0040] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "contacted", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0041] The current hospital consultation process is that if a patient comes to remove the tube, he or she will first go to the outpatient clinic for tube removal. If there are stones on the DJ tube, it will not be removed and the patient will be in great pain. Then, hospitalization is required. First, surgery is performed to break up the stones on the DJ tube, and then the tube is removed. Before the DJ tube is removed, the patient cannot determine whether there are stones attached to the wall of the DJ tube. Often, the tube is removed directly in the outpatient clinic. Only when it cannot be removed will the patient be hospitalized for surgery. After hospitalization, a CT scan will be taken, and the doctor needs to carefully observe the condition of the DJ tube stones before surgery. During this process, the patient suffers great pain. In view of this situation, this embodiment proposes a DJ tube stone detection system, including

[0042] A data input module, for receiving biochemical data of the patient;

[0043] The stone prediction module predicts the probability of the patient developing DJ tube mural stones based on the biochemical data received by the data input module, and recommends a DJ tube removal plan to the patient based on the probability of DJ tube mural stones.

[0044] like Figure 1 The workflow diagram of the DJ tube wall stone detection system shown in the figure predicts the probability of stone formation in patients through the stone prediction module, and determines whether the probability of stone formation in patients is greater than the set probability. The set probability in this embodiment is 0.5, and the size of the set probability can be adjusted by the doctor according to the actual situation. When the probability of stone formation is greater than the set probability, it is recommended that the patient be hospitalized for surgery.

[0045] For the data input module, the input data includes biochemical data, which is used in the stone prediction module. The specific process is that when a patient comes to the hospital to remove the DJ tube, the doctor enters the patient's biochemical data before the lithotripsy operation into the system, and then the system uses the biochemical data to determine whether the patient has DJ tube mural stones.

[0046] like Figure 2The prediction module workflow diagram shows that for the stone prediction module, the patient's biochemical data is used to predict stone growth and the probability of the patient having DJ tube mural stones. This avoids the pain of the patient being unable to remove the DJ tube, and saves CT medical expenses for patients who do not have mural stones. The biochemical data here refers to the patient's blood test report, urine test report, coagulation function, renal function, etc. before lithotripsy. In order to facilitate the operation of doctors, doctors can directly input biochemical data into the system in the form of images, text, etc. The system has a built-in text recognition function, which can identify various biochemical indicators from the image, and then select highly correlated indicators based on the correlation. Highly correlated indicators can be set as needed. In this embodiment, 15 highly correlated indicators are selected, and the logistic regression algorithm in machine learning is used to use 15 highly correlated biochemical indicators to judge whether the patient has DJ tube wall stones, and calculate the probability of DJ tube wall stones. If the probability is >0.5, it is considered that the patient has a high probability of DJ tube wall stones. If the probability is <=0.5, it is considered that the patient has a high probability of not having DJ tube wall stones, and based on the probability of DJ tube wall stones, a DJ tube removal plan is recommended to the user. For example, when the probability is >0.5, it is recommended that the patient be removed surgically, and when the probability is <=0.5, it is recommended that the patient be removed in the outpatient clinic.

[0047] When patients come to remove the tube, this system will first use blood tests, urine tests and other biochemical data to determine the probability of stones on the DJ tube. If the probability is low, then go directly to the outpatient clinic for tube removal, saving time and money for the patient. If the probability is high, then it is recommended to go to the hospital for treatment, which can avoid the pain of tube removal.

[0048] At present, it is impossible to directly determine whether the DJ tube placed in the patient's body has attached stones to the wall. It can only be determined after taking a CT scan and when it cannot be removed. The former is expensive to take a CT scan, and the occurrence of DJ tube attached stones to the wall is rare. If every patient does a CT scan, it will increase the financial pressure of the patient; the latter will cause the patient to suffer great pain, affecting the patient's physical and mental health. The DJ tube attached stone detection system proposed in this embodiment can determine the probability of DJ tube attached stones according to the patient's biochemical data when the patient seeks medical treatment, and then recommend the patient to seek outpatient treatment or hospitalization, and the medical experience is good.

[0049] Further optionally, the data input module also receives CT image data, and the DJ tube wall stone detection system also includes

[0050] The stone detection module is used to analyze the severity of DJ tube mural stones in patients based on the CT image data received by the data input module.

[0051] like Figure 1The work flow chart of the DJ tube wall stone detection system shown in the figure, the data received by the data input module also includes CT image data, and the CT image data is used for the stone detection module. When it is determined that there is a stone attached to the DJ tube wall, the patient's CT image needs to be input, and then the system detects the location and severity of the stone in the input CT image.

[0052] In order to further understand the severity of DJ tube mural stones in patients, when the probability of stones is judged to be greater than the set probability, the stone detection module is also used to analyze the severity of DJ tube mural stones in patients to provide a reference for subsequent surgical treatment. Figure 3 The stone detection module workflow diagram shown in the figure is for patients whose DJ tube wall stones have been determined by the stone prediction module. When it is determined that the patient's DJ tube has stones, the patient's CT image needs to be input into the stone detection module. The stone detection module first segments the DJ tube and stone area in the CT image, and fills and corrects the segmentation results to ensure that the DJ tube and the wall stone area can be segmented in each CT image; then the segmentation area coordinates are mapped to the original image to obtain the DJ tube ROI area in the original image. Specifically, Figure 4 After the patient's CT image is input into the stone detection module, two images are obtained, namely the segmentation image and the original image. In the segmentation image, only the segmented area is white, and the rest of the area is black. Then, the circumscribed rectangle of the segmented area can be obtained, and then the area with the same coordinates as the circumscribed rectangle is taken out on the original image, which is the ROI area of ​​the DJ tube in the original image.

[0053] Then, the DJ tube ROI area in the original image is segmented by threshold to obtain an image containing only the DJ tube and the stone attached to the wall; specifically, Figure 4 The reason why we do threshold segmentation here is that the segmentation above cannot accurately segment the DJ tube area. Now we have the DJ tube ROI area, which only contains the DJ tube. Then we can directly do a simple pixel value threshold segmentation to accurately obtain the DJ tube image. Mural calculi grow on the DJ tube, and the pixel values ​​of the two are basically the same, and both are displayed as white in the image, so we can use threshold segmentation to segment both at the same time.

[0054] Finally, the grayscale co-occurrence matrix and Fourier descriptor features are extracted from the DJ tube and wall stone images, and the stones are detected based on the extracted features to determine whether there are wall stones in the segmentation results of each CT. Specifically, after the segmented DJ tube and wall stone segmentation images have been obtained, feature extraction can be performed on the image to extract grayscale co-occurrence matrix features and Fourier descriptor features, and then the extracted features can be used for stone detection, that is, classification, to determine whether there are wall stones in the image. There is a certain difference between images with and without wall stones, which can be judged. Because each patient has about two to three hundred CT slice images, each slice image needs to be judged to determine whether there are DJ tube wall stones in each slice. If so, the number, area and relative position of the wall stones to the human body are calculated.

[0055] If present, the number, area and relative position of the mural stones will be calculated. If not present, no record will be made. After all images are judged, the system will screen out all CT images of DJ tubes with mural stones, mark the location and area of ​​the stones, and finally distinguish the severity of DJ tube mural stones based on the number and area of ​​all stones. Doctors can make a secondary judgment on the stone detection results. If the system makes a false detection, the doctor is allowed to modify the current stone detection results to obtain a more accurate diagnosis.

[0056] The DJ tube wall stone detection system proposed in this embodiment has the functions of automatic segmentation, classification and marking of DJ tube wall stones. The doctor only needs to make a judgment based on the stone situation detected by the system. That is, when the CT image is uploaded to this system, the system will automatically segment the CT image to segment the DJ tube and the wall stone area, and then detect or classify the segmented DJ tube and wall stone area to determine whether there are wall stones on each CT slice. If so, the wall stones are marked on the slice. In this way, the automatic film reading is completed, and the doctor only needs to observe the results of the automatic film reading, which greatly improves the film reading efficiency.

[0057] Further optionally, in the stone detection module,

[0058] The DJ tube and stone areas in the CT image are segmented, and the segmentation results are filled and corrected to ensure that the DJ tube and wall stone areas can be segmented in each CT image; the segmented area coordinates are mapped to the original image to obtain the DJ tube ROI area in the original image; the DJ tube ROI area in the original image is then threshold segmented to obtain an image containing only the DJ tube and wall stones; the grayscale co-occurrence matrix and Fourier descriptor features in the DJ tube and wall stone images are extracted, and the stones are classified based on the extracted features to determine whether there are wall stones in the segmentation results of each CT. If so, the number, area and relative position of the wall stones are calculated, and the severity of the DJ tube wall stones is distinguished according to the number and area of ​​all stones.

[0059] Further optionally, a neural network is used to segment the DJ tube and stone areas in the CT image.

[0060] Further optionally, SVM or random forest algorithm is used to classify stones based on the extracted features.

[0061] Further optionally, the DJ tube wall stone detection system also includes

[0062] The structured report output module is used to display all the patient's disease diagnosis results and output them in the form of a structured report.

[0063] like Figure 1 The workflow diagram of the DJ tube mural stone detection system is shown. After diagnosing the patient's condition, the system will display all the patient's diagnosis results, including biochemical data analysis results, the probability of DJ tube mural stones, the presence of stones in CT images (the number and location of stones, etc.) and the severity, and finally output them in the form of a structured report.

[0064] This embodiment also proposes a method for detecting DJ tube wall stones, which uses the above-mentioned DJ tube wall stone detection system. The detection method includes:

[0065] Obtain the patient's biochemical data;

[0066] The probability of DJ tube mural stones in the patient is predicted based on biochemical data, and a DJ tube removal plan is recommended to the patient based on the probability of DJ tube mural stones.

[0067] Further optionally, when the probability of DJ tube mural stones in the patient is greater than the set probability, surgical treatment is recommended for the patient; when the probability of DJ tube mural stones in the patient is less than or equal to the set probability, outpatient treatment is recommended for the patient.

[0068] Further optionally, when the probability of DJ tube mural stones in the patient is greater than a set probability, the method further includes:

[0069] Acquire CT image data of the patient;

[0070] The severity of DJ tube mural stones in patients was analyzed based on CT image data.

[0071] like Figure 2 The detection flow chart includes the following steps:

[0072] Further optionally, after acquiring the CT image data of the patient, it is also determined whether the patient has a single DJ tube or a double DJ tube.

[0073] When it is determined that the patient has a single DJ tube, the severity of the DJ tube mural stones in the patient is analyzed directly based on the CT image data;

[0074] When it is determined that the patient has double DJ tubes, the CT image is vertically cropped from the center point, and the two cropped images are used as the single DJ tube case. The severity of the DJ tube mural stones in the patient is analyzed for the two cropped images.

[0075] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment, it is not used to limit the present invention. Any technician familiar with this patent can make some changes or modify the technical contents suggested above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the solution of the present invention.

Claims

1. A DJ tube wall stone detection system, characterized in that: include Data input module, used to receive the patient's biochemical data and CT image data A stone prediction module is used to predict the probability of the patient having DJ tube wall stones according to the biochemical data received by the data input module, and recommend a DJ tube removal plan to the patient according to the probability of DJ tube wall stones; in the stone prediction module, a character recognition algorithm is used to identify and save high-correlation indicators, and the high-correlation indicators are input into a logistic regression algorithm model, and based on the trained model, a prediction is made as to whether the patient has DJ tube wall stones, and the probability of the existence of DJ tube wall stones is output, and a DJ tube removal plan is recommended to the user based on the probability of the existence of DJ tube wall stones; A stone detection module, used for analyzing the severity of DJ tube mural stones in patients according to the CT image data received by the data input module; When it is determined that the probability of stones is greater than the set probability, it is determined that there are stones attached to the DJ tube wall, and the patient's CT image is input into the stone detection module to analyze the severity of the DJ tube wall stone situation in the patient.

2. A DJ tube wall stone detection system according to claim 1, characterized in that: In the stone detection module, The DJ tube and stone areas in the CT image are segmented, and the segmentation results are filled and corrected to ensure that the DJ tube and wall stone areas can be segmented in each CT image; the segmented area coordinates are mapped to the original image to obtain the DJ tube ROI area in the original image; the DJ tube ROI area in the original image is then threshold segmented to obtain an image containing only the DJ tube and wall stones; the grayscale co-occurrence matrix and Fourier descriptor features in the DJ tube and wall stone images are extracted, and the stones are classified based on the extracted features to determine whether there are wall stones in the segmentation results of each CT. If so, the number, area and relative position of the wall stones are calculated, and the severity of the DJ tube wall stones is distinguished according to the number and area of ​​all stones.

3. A DJ tube wall stone detection system according to claim 1, characterized in that: A neural network was used to segment the DJ tube and stone areas in CT images.

4. A DJ tube wall stone detection system according to claim 1, characterized in that: SVM or random forest algorithm was used to classify stones based on the extracted features.

5. A DJ tube wall stone detection system according to any one of claims 1 to 4, characterized in that: The DJ tube wall stone detection system also includes The structured report output module is used to display all the patient's disease diagnosis results and output them in the form of a structured report.

6. A method for detecting DJ tube wall stones, characterized in that: It adopts the DJ tube wall stone detection system according to any one of claims 1 to 5, and the detection method includes Obtain the patient's biochemical data; The probability of DJ tube mural stones in the patient is predicted based on biochemical data, and a DJ tube removal plan is recommended to the patient based on the probability of DJ tube mural stones.

7. A method for detecting DJ tube wall stones according to claim 6, characterized in that: When the probability of DJ tube mural stones in patients is greater than the set probability, surgical treatment is recommended for the patients; when the probability of DJ tube mural stones in patients is less than or equal to the set probability, outpatient treatment is recommended for the patients.

8. A method for detecting DJ tube wall stones according to claim 7, characterized in that: When the probability of DJ tube mural stones in patients is greater than the set probability, it also includes: Acquire CT image data of the patient; The severity of DJ tube mural stones in patients was analyzed based on CT image data.

9. A method for detecting DJ tube wall stones according to claim 8, characterized in that: After obtaining the patient's CT image data, it is also determined whether the patient has a single DJ tube or a double DJ tube. When it is determined that the patient has a single DJ tube, the severity of the DJ tube mural stones in the patient is analyzed directly based on the CT image data; When it is determined that the patient has double DJ tubes, the CT image is vertically cropped from the center point, and the two cropped images are used as the single DJ tube case. The severity of the DJ tube mural stones in the patient is analyzed for the two cropped images.

Images