Method, device, and computer program for predicting metabolic syndrome after bariatric surgery

Abstract

The present invention is characterized in that it comprises a method for predicting remission of metabolic syndrome after bariatric surgery, the method comprising: receiving preoperative abdominal and pelvic CT images of one or more obese patients; setting a region of interest centered on the circumference of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images and extracting radiomics features within the region of interest through radiomics analysis; selecting statistically significant features among the extracted radiomics features using logistic regression analysis; and inputting the selected radiomics features into a prediction model to predict remission of metabolic syndrome after bariatric surgery.

Description

Technology Field

[0001] The present invention relates to a method, apparatus, and computer program for predicting remission of metabolic syndrome after bariatric surgery, and in particular to a method, apparatus, and computer program for predicting remission of metabolic syndrome after bariatric surgery using skeletal muscle analysis of preoperative abdominal CT. Background Technology

[0002] The prevalence of obesity is steadily increasing worldwide, emerging as a serious public health issue. Obesity often leads to metabolic abnormalities such as type 2 diabetes, impaired glucose tolerance, dyslipidemia, hypertension, non-alcoholic fatty liver disease, and metabolic syndrome, thereby increasing cardiovascular disease mortality.

[0003] Bariatric surgery has emerged as the most effective and sustainable treatment for weight loss and significant improvement in obesity-related comorbidities. Numerous studies report the positive effects of bariatric surgery on metabolic syndrome. However, not all obese individuals experience alleviation of metabolic syndrome after bariatric surgery. Therefore, predicting who will benefit from metabolic syndrome improvement prior to surgery is crucial for preoperative evaluation.

[0004] Skeletal muscle, the largest insulin-sensitive tissue, regulates systemic glucose utilization in response to insulin stimulation. Metabolic syndrome is closely associated with insulin resistance in skeletal muscle. Reduced skeletal muscle mass and intramuscular fat accumulation can cause contractile disorders and lead to metabolic abnormalities. Considering the relationship between skeletal muscle and insulin metabolism, there is a possibility that preoperative skeletal muscle mass may influence the improvement of metabolic syndrome after bariatric surgery.

[0005] Radiomics refers to the extraction of high-dimensional data obtainable from medical imaging sources such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and ultrasound. It is primarily used in the field of tumor biology within oncology to improve diagnosis and prognosis. The strength of radiomics lies in its ability to enable more personalized targeted therapy based on individual patient characteristics and to advance the field of precision medicine. Unlike studies that have used radiomics to verify tumor biology in oncology, there are a few studies that have applied radiomics to skeletal muscle. However, no research has been proposed to apply radiomics to the skeletal muscle of patients who have undergone bariatric surgery to predict the improvement of metabolic syndrome.

[0006] Therefore, the present invention aims to develop and evaluate the performance of a machine learning prediction model for predicting metabolic syndrome remission after bariatric surgery based on preoperative skeletal muscle characteristics using medical imaging information systems (radiomics). Prior art literature

[65535] Republic of Korea Registered Patent No. 10-2321427 (Registered Nov. 04, 2021) Meng Y, Sun J, Qu N, Zhang G, Yu T, Piao H. Application of Radiomics for Personalized Treatment of Cancer Patients. Cancer Manag Res. 2019;11:10851-8. Epub 20191230. The problem to be solved

[0008] One objective of the present invention is to provide a method, apparatus, and computer program for predicting remission of metabolic syndrome after bariatric surgery based on preoperative skeletal muscle characteristics using medical imaging information systems (radiomics) and a machine learning prediction model. means of solving the problem

[0009] To achieve the above objective, the present invention comprises a method for predicting remission of metabolic syndrome after bariatric surgery, the method comprising: receiving preoperative abdominal and pelvic CT images of one or more obese patients; setting a region of interest centered on the circumference of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images and extracting radiomics features within the region of interest through radiomics analysis; selecting statistically significant features among the extracted radiomics features using logistic regression analysis; and inputting the selected radiomics features into a prediction model to predict remission of metabolic syndrome after bariatric surgery.

[0010] Preferably, the obese patient may be defined as a patient having at least one of the following: blood pressure of 130 / 85 mmHg or higher or using antihypertensive drugs, fasting blood glucose level of 126 mg / dL or higher, hemoglobin A1c (HbA1c) level of 6.5% or higher or using diabetes drugs, and a waist-to-hip ratio (WHR) of 0.95 or higher for women or 1.03 or higher for men.

[0011] Preferably, the step of extracting the above medical imaging information (radiomics) features can extract a region of interest centered on the entire abdominal muscle area including the psoas muscle, paravertebral muscles, quadratus lumborum, rectus abdominis, transverse abdominis, or internal and external oblique muscles in each slice across the entire third lumbar vertebra (L3).

[0012] Preferably, the step of extracting the medical imaging information systems (radiomics) features comprises using linear interpolation to set the spatial resolution of the CT image to 1*1*1mm 3 It can be resampled.

[0013] Preferably, the step of selecting the significant features may include a first step of generating a feature map using univariate logistic regression analysis.

[0014] Preferably, the step of selecting the significant features may further include a second step of verifying the potential association between the selected significant features and metabolic syndrome remission after bariatric surgery.

[0015] Preferably, the step of selecting the significant features may further include a third step of evaluating redundancy or correlation between the significant features verified as potentially associated in the second step using absolute correlation (COR) or variance inflation factor (VIF).

[0016] Preferably, the step of selecting the significant features may further include a fourth step of evaluating the relevance or importance of the significant features verified as having a possible association in the second step.

[0017] Preferably, the first step may reflect the verification or evaluation results performed in the second to fourth steps in the feature map.

[0018] Preferably, the step of selecting the significant features may further include a fifth step of selecting optimal features from the feature map using LASSO (Least Absolute Shrinkage and Selection Operator), EN (Elastic-Net), and RF (Random Forest).

[0019] Preferably, the third step uses absolute correlation (COR), and the fifth step can use Random Forest (RF).

[0020] Preferably, the method may further include the step of training a prediction model using medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome was alleviated after bariatric surgery, or medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome was not alleviated after bariatric surgery.

[0021] Preferably, the prediction model may be any one of LR (logistic regression), RF (random forest), and SVM (support vector machine).

[0022] In addition, the present invention provides a device for predicting remission of metabolic syndrome after bariatric surgery, comprising: a processor including one or more cores; and a memory; The processor includes, and is further characterized by receiving preoperative abdominal and pelvic CT images of one or more obese patients, setting a region of interest centered on the circumference of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images, extracting radiomics features within the region of interest through radiomics analysis, selecting statistically significant features among the extracted radiomics features using logistic regression analysis, training a prediction model with radiomics features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome has been relieved after bariatric surgery or radiomics features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome has not been relieved after bariatric surgery, and inputting the selected radiomics features into the prediction model to predict remission of metabolic syndrome after bariatric surgery.

[0023] In addition, the present invention is further characterized by a computer program that includes instructions stored in a computer-readable storage medium to cause a computer to perform the following operations, wherein the operations include: receiving preoperative abdominal and pelvic CT images of one or more obese patients; setting a region of interest centered on the circumference of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images and extracting radiomics features within the region of interest through radiomics analysis; selecting statistically significant features among the extracted radiomics features using logistic regression analysis; and training a prediction model with the selected radiomics features. Effects of the invention

[0024] The present invention has the advantage of being able to predict whether metabolic syndrome will be alleviated after bariatric surgery by analyzing skeletal muscle CT images of obese patients.

[0025] The present invention has the advantage of being able to predict the prognosis of metabolic syndrome after bariatric surgery prior to the surgery, thereby providing objective criteria for patients to choose bariatric surgery. Brief explanation of the drawing

[0026] Figure 1 shows a flowchart of a method for predicting remission of metabolic syndrome after bariatric surgery according to an embodiment of the present invention. FIG. 2 shows a schematic diagram of a method for predicting remission of metabolic syndrome after bariatric surgery according to an embodiment of the present invention. Figure 3 shows a flowchart for selecting obese patients used in learning a prediction model according to an embodiment of the present invention. FIG. 4 shows a flowchart of the step of selecting a significant feature according to an embodiment of the present invention. Figure 5 shows a graph comparing the AUC of each combination using the cross-validation method according to an embodiment of the present invention. Figure 6 shows a configuration diagram of a device for predicting remission of metabolic syndrome after bariatric surgery according to an embodiment of the present invention. FIG. 7 shows a schematic diagram of a computing environment according to an embodiment of the present invention. Specific details for implementing the invention

[0027] The present invention will be described in detail below with reference to the contents described in the attached drawings. However, the present invention is not limited or restricted by exemplary embodiments. Identical reference numerals in each drawing indicate components that perform substantially the same function.

[0028] The purpose and effects of the present invention may be naturally understood or become clearer through the following description, and the purpose and effects of the present invention are not limited solely to the description below. Furthermore, in describing the present invention, if it is determined that a detailed description of known technology related to the present invention may unnecessarily obscure the essence of the present invention, such detailed description will be omitted.

[0029] The terms used in this invention are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the description of the invention, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0030] Terms such as "first," "second," etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component.

[0031] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which this invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this invention.

[0032] In interpreting the components, they are interpreted to include a margin of error even without a separate explicit indication. In the case of descriptions regarding temporal relationships, for example, where the temporal sequence is described using 'after,' 'following,' 'next,' 'before,' etc., cases that are not continuous are included unless 'immediately' or 'directly' is used.

[0033] Hereinafter, the technical configuration of the present invention will be described in detail with reference to the attached drawings.

[0034] FIG. 1 shows a flowchart of a method for predicting remission of metabolic syndrome after bariatric surgery according to an embodiment of the present invention. Referring to FIG. 1, the method for predicting remission of metabolic syndrome after bariatric surgery may include the steps of receiving a CT image (S100), extracting medical image information systems (radiomics) features (S200), selecting significant features (S300), training a prediction model (S400), and predicting remission of metabolic syndrome after bariatric surgery (S500).

[0035] To predict metabolic syndrome remission after bariatric surgery, a CT radiomics-based predictive model can be utilized to determine whether preoperative skeletal muscle characteristics can predict postoperative remission. This method predicts the likelihood of metabolic syndrome remission by analyzing skeletal muscle CT images of obese patients. Since this method allows for the prediction of the prognosis of postoperative metabolic syndrome prior to the surgery, it can provide objective criteria for patients to make informed decisions regarding the procedure.

[0036] FIG. 2 shows a schematic diagram of a method for predicting remission of metabolic syndrome after bariatric surgery according to an embodiment of the present invention. Referring to FIG. 2, the method for predicting remission of metabolic syndrome after bariatric surgery receives abdominal and pelvic CT images and analyzes radiomics data of the received CT images to manually segment the entire skeletal muscle included at the level of the third lumbar vertebra in the non-contrast CT image performed prior to bariatric surgery and extract internal radiomics features. The extracted radiomics features may include intensity features, texture features, and deep features with applied image filters, which are histograms of CT attenuation indices (Hounsfield units) of image pixels within the skeletal muscle. The method for predicting remission of metabolic syndrome after bariatric surgery may perform feature selection to extract statistically meaningful values ​​from the extracted radiomics feature values. Ultimately, a method to predict metabolic syndrome remission after bariatric surgery involves generating a prediction model using medical imaging information systems (radiomics) features and then predicting metabolic syndrome remission after bariatric surgery.

[0037] FIG. 3 shows a flowchart for selecting obese patients to be used for training a prediction model according to an embodiment of the present invention. Referring to FIG. 3, abdominal and pelvic CT scans for one year prior to and after surgery were collected from 83 out of 124 patients who underwent bariatric surgery from January 2019 to October 2019 for the training and evaluation of the prediction model according to the present invention. Among the 83 patients, 3 patients with poor CT image quality and 1 patient with insufficient clinical data were excluded. As a result, data for training and evaluating the prediction model was obtained from abdominal and pelvic CT images of 79 patients. Of the 79 patients, 49 (62%) underwent sleeve gastrectomy, 15 (19%) underwent Roux-en-Y gastric bypass, and 15 (19%) underwent sleeve gastrectomy combined with duodenojejunostomy.

[0038] Table 1 below shows the basic characteristics of the data for training and evaluating the predictive model, and presents the results of comparing clinical baselines between those who experienced metabolic syndrome remission after non-reverse metabolic surgery and those whose metabolic syndrome persisted. Referring to Table 1, most of the 79 morbidly obese patients had diabetes (72 patients, 91.1%) and hypertension (61 patients, 77.2%), and according to the new definition, all were metabolically unhealthy. In Table 1, BMI (body mass index) is body mass index, SBP (systolic blood pressure) is systolic blood pressure, DBP (diastolic blood pressure) is diastolic blood pressure, FBS (Fasting blood sugar) is fasting blood sugar, HDL (High-density lipoprotein) is high-density lipoprotein, LDL (Low-density lipoprotein) is low-density lipoprotein, HbA1c (glycosylated hemoglobin) is glycated hemoglobin, AST (aspartate aminotransferase) is aspartate aminotransferase, ALT (alanine aminotransferase) is alanine aminotransferase, eGFR (estimated glomerular filtration rate) is estimated glomerular filtration rate, and MS (metabolic syndrome) is metabolic syndrome. Hypertension is defined as a blood pressure (BP) of 130 / 85 mmHg or higher, taking antihypertensive drugs, or having a fasting blood glucose level of 126 mg / dL or higher. Type 2 diabetes is defined as an HbA1c level of 6.5% or higher, or using diabetes medication.

[0039] Baseline characteristics Total (N=79) Remission of MS (N=29) Persistent MS (N=50) P-value Age (yr) 36 [28-50] 29 [26 -42] 43.5 [32.25-51] 0.0014 Gender; Female 53 (67.09%) 21 (72.41%) 32 (64%) 0.6039 BMI (kg / m2) 39.5 [35.63-46.05] 37.71 [35.57-42.44] 41 [35.78-48.87] 0.1005 Height (cm) 165.3 [161.6-171.3] 165.5 [161.4-168.3] 165.2 [161.85-172.55] 0.7719 Weight (kg) 108 [97.05-129.8] 105.3 [97.5-123] 114.85 [96.97-137.7] 0.1947 WC (Waist circumference) (cm) 118.32 ± 13.72 112.96 ± 11.08 121.43 ± 14.23 0.0044 WHR (Waist-Hip ratio) 1.03 ± 0.07 1 ± 0.08 1.04 ± 0.06 0.0156 > 0.95 (women), 1.03 (men) 66 (83.54%) 22 (75.86%) 44 (88%) 0.211 SBP (mmHg) 130 [121.5-146.5] 125 [120-138] 133 [124.25-149.5] 0.0134 DBP (mmHg) 79 [70-83.5] 78 [70-80] 80 [70-86.5] 0.3923 Hypertension* 61 (77.22%) 16 (55.17%) 45 (90%) 0.001 FBS (mg / dL) 117 [103.5-140] 110 [97-118] 124 [108-154] 0.0018 HDL cholesterol (mmol / L) 44-60 44-61 44-58.75 0.9189 HDL < 40 (men), <50 (women) 22 / 79 (27.85%) 8 / 29 (27.59%) 14 / 50 (28%) >0.99 LDL cholesterol (mmol / L) 119.58 ± 34.85 129.14 ± 35.33 114.04 ± 33.68 0.0678 Triglycerides (mmol / L) 124-221 139-200 105-223.75 0.9351 TG >= 150 or medication 42 / 79 (53.16%) 16 / 29 (55.17%) 26 / 50 (52%) 0.9693 HbA1c (%) 5.9 [5.35-6.75] 5.6 [5.2-6.1] 6 [5.5-7.18] 0.0139 Type 2 Diabetes** 72 (91.14%) 29 (100%) 43 (86%) 0.0432 AST (IU / L) 30 [19-42.5] 30 [22-42] 28 [17.5-48.25] 0.7951 ALT (IU / L) 38 [20-57] 42 [28-59] 31.5 [19-57] 0.222 eGFR (mL / min) 95.98-120.72 106.6-123.38 91.27-114.49 0.0031 eGFR < 60 (mL / min) 3 (3.8%) 0 (0%) 3 (6%) 0.2941

[0040] Table 2 below compares postoperative factors based on whether metabolic syndrome was alleviated or persisted. Referring to Table 2, bariatric surgery reduced body weight, BMI, diabetes prevalence, and Waist-Hip ratio (WHR) over one year postoperatively, and improved the postoperative lipid profile. However, bariatric surgery did not change the estimated glomerular filtration rate (eGFR) postoperatively.

[0041] Post operative factors Total (N=79) Remission of MS (N=29) Persistent MS (N=50) P-value Weight (kg) 78.8 [69.6-95] 73.1 [63.7-85.6] 82.75 [75.82-99.98] 0.0028 Delta weight 29.4 [24-40.9] 32.9 [27.2-39.8] 27.8 [22.33-41.25] 0.1604 Total weight loss (%) 28.38 ± 8.22 (%) 31.49 ± 7.87 (%) 26.57 ± 7.95 (%) 0.0098 BMI (kg / m2) 28.48 [26.28-32.3] 27.35 [24.32-29.32] 30.25 [27.16-34.66] <0.001 WHR (Waist-Hip ratio) 0.95 (women), 1.03 (men) 27 (34.18%) 0 (0%) 27 (54%) <0.001 SBP (mmHg) 120 [113-133] 115 [112-120] 127.5 [120-139.5] <0.001 DBP (mmHg) 72.89 ± 12.2 66.41 ±10.16 76.64 ± 11.78 <0.001 Hypertension* 32 (40.51%) 0 (0%) 32 (64%) <0.001 FBS (mg / dL) 97 [91-104.5] 94 [88-98] 99 [93.25-105.75] 00096 HbA1c (%) 5.3 [4.9-5.45] 5.3 [4.8-5.3] 5.3 [5-5.5] 0.1145 Type 2 Diabetes** 12 (15.19%) 0 (0%) 12 (15.19%) 0.0029 HDL cholesterol (mmol / L) 64.22 ± 14.1 65.59 ±13.98 63.42 ±14.25 0.5123 LDL cholesterol (mmol / L) 80-121 91-123 76.5-115.75 0.1776 Triglycerides (mmol / L) 68.5-126.5 64-108 73-140.5 0.0207 AST (IU / L) 19 [15-24.5] 19 [13-22] 19 [17-25] 0.2157 ALT (IU / L) 18 [13-26] 16 [11-26] 19.5 [13-25.25] 0.4419 eGFR (mL / min) 92.32-118.62 103.29-121.96 92.01-113.66 0.0751

[0042] The step of receiving CT images (S100) may receive preoperative abdominal and pelvic CT images of one or more obese patients. The abdominal and pelvic CT images received in the step of receiving CT images (S100) may be scanned using a 64-slice scanner.

[0043] The step of receiving CT images (S100) can receive preoperative abdominal and pelvic CT images of obese patients and transmit them to the step of extracting medical image information systems (radiomics) features (S200). The step of receiving CT images (S100) can transmit the original images to the step of selecting a context range using a wireless or wired network (S120), and when using a wireless network, a broad mobile communication network such as CDMA (Code Division Multiple Access), WCDMA (Wideband CDMA), LTE (Long Term Evolution), or Wi-Fi can be used.

[0044] Obesity refers to a body mass index (BMI) exceeding 30 kg / m². In the present invention, the alleviation of metabolic syndrome is defined as metabolic health, which may be determined according to a new definition. According to the new definition of the present invention, an obese patient may be defined as a patient who meets one or more of the following criteria: blood pressure of 130 / 85 mmHg or higher or the use of antihypertensive drugs; a fasting blood glucose level of 126 mg / dL or higher; a hemoglobin A1c (HbA1c) level of 6.5% or higher or the use of diabetes drugs; and a waist-to-hip ratio (WHR) of 0.95 or higher for women or 1.03 or higher for men. Preferably, if none of the above criteria are met, it may be judged as metabolic health.

[0045] The step (S200) of extracting medical imaging information systems (radiomics) features involves setting a region of interest centered on the perimeter of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images, and extracting medical imaging information systems (radiomics) features within the region of interest through medical imaging information systems (radiomics) analysis.

[0046] The step (S200) of extracting medical imaging information (radiomics) features can extract a region of interest centered on the entire abdominal muscle area including the psoas muscle, paravertebral muscles, quadratus lumborum, rectus abdominis, transverse abdominis, or internal and external oblique muscles in each slice across the entire third lumbar vertebra (L3).

[0047] The step (S200) of extracting radiomics features may utilize an artificial intelligence model capable of image processing for segmenting (setting regions of interest) and extracting radiomics features from non-contrast enhanced axial preoperative abdominal and pelvic CT images, and preferably, a special software program (syngo.viaFrontier) may be used.

[0048] The step of extracting medical imaging information systems (radiomics) features (S200) can set the region of interest using an artificial intelligence model trained on abdominal and pelvic CT images in which a skilled musculoskeletal radiologist with 8 years of experience after a fellowship and no access rights to clinical data marks the region of interest across the entire lumbar spine (L3).

[0049] The step (S200) of extracting medical imaging information systems (radiomics) features uses linear interpolation to set the spatial resolution of the CT image to 1*1*1mm 3 It can be resampled.

[0050] FIG. 4 shows a flowchart of the step (S300) of selecting a significant feature according to an embodiment of the present invention. Referring to FIG. 4, the step (S300) of selecting a significant feature may include a first step (S310), a second step (S320), a third step (S330), a fourth step (S340), and a fifth step (S350).

[0051] The step of selecting significant features (S300) can select statistically significant features among the medical image information systems (radiomics) features extracted using logistic regression analysis. The step of selecting significant features (S300) can improve the performance of the prediction model by removing noise and prevent overfitting in the prediction model. The step of selecting significant features (S300) can be designed as an optimal feature selection framework consisting of a total of 5 steps.

[0052] Step 1 (S310) can generate a feature map using univariate logistic regression. Here, univariate logistic regression refers to logistic regression when there is only one dependent variable, and in the present invention, the dependent variable may be whether metabolic syndrome is alleviated after bariatric surgery. Step 1 (S310) can store the medical imaging information systems (radiomics) features extracted in the step of extracting medical imaging information systems (radiomics) features (S200) in an initial candidate feature set. Step 1 (S310) can generate a feature map for the medical imaging information systems (radiomics) features included in the candidate feature set. The feature map stores all information generated during the process of selecting significant features for each feature. Step 1 (S310) can perform univariate analysis on the relationship between each feature and the output value. The feature map can store the category and univariate analysis results for each feature. For example, the first step (S310) may store statistical correlations for each feature (e.g., p-value, AUC, odds ratio (OR), etc.) and basic information of the corresponding feature (category of medical imaging information systems (radiomics) feature, name of the feature) in the feature map. The category of medical imaging information systems (radiomics) feature may be classified into morphological features, intensity, or texture, etc.

[0053] Step 1 (S310) updates the feature map during the selection process of medical imaging information systems (radiomics) features and accumulates other information. That is, Step 1 (S310) can reflect the verification or evaluation results performed in Step 2 (S320) through Step 4 (S340) in the feature map.

[0054] Step 2 (S320) can verify the possibility of association between selected significant features and remission of metabolic syndrome after bariatric surgery. Step 2 (S320) examines the basic possibility of significance based on the relationship between the feature (selected significant feature) and the output value (remission of metabolic syndrome after bariatric surgery) for each of the medical imaging information systems (radiomics) features included in the candidate feature set. Step 2 (S320) can filter (remove) features with low correlation to the output value. Step 2 (S320) determines the possibility for each feature based on the p-value or AUC, and features with low probability can be removed from the candidate feature set. For example, Step 2 (S320) can remove features from the candidate feature set where the p-value is greater than or equal to a first threshold (e.g., 0.5) or the AUC is less than a second threshold (e.g., 0.7). In other words, based on the second step (S320) p-value or AUC, only features with a high correlation to the model's output value can be retained in the candidate feature set.

[0055] Step 3 (S330) can evaluate redundancy or correlation between significant features verified as potentially associated in Step 2 using absolute correlation (COR) or variance inflation factor (VIF). Here, correlation refers to a statistical measure that expresses the range in which two variables are in a linear relationship, and the variance inflation factor refers to a criterion for determining whether independent variables have a problem of multicollinearity in multiple regression analysis. That is, Step 3 (S330) can filter features based on the collinearity between features regarding the medical imaging information systems (radiomics) features remaining in the candidate feature set. Therefore, Step 3 (S330) can calculate an indicator representing the degree of correlation between features. Step 3 (S330) can remove features from the candidate feature set in which the absolute correlation is greater than or equal to the third threshold or the variance inflation factor is greater than or equal to the fourth threshold. That is, the third step (S330) can retain a candidate feature set containing only features with low correlation (i.e., independent) based on absolute correlation or variance expansion factors.

[0056] Step 4 (S340) can evaluate the relevance or importance of significant features verified as potentially associated in Step 2 (S320). That is, Step 4 (S340) can perform significance-based filtering based on the relationship between the feature and the output value for each of the medical imaging information systems (radiomics) features remaining in the candidate feature set. At this time, Step 4 (S340) can select features based on the odds ratio along with the "p-value or AUC." For example, Step 4 (S340) can remove features from the candidate feature set where (i) the p-value is greater than or equal to the 5th threshold or the AUC is less than the 6th threshold, and simultaneously (ii) the odds ratio (OR) is 1 (removing features where the increase or decrease in variable values ​​does not affect the output of the learning model). That is, step 4 (S340) can retain only the features in the candidate feature set that have a high correlation with the output value of the learning model based on "p-value or AUC" and, at the same time, the increase or decrease in variable values ​​affects the learning model.

[0057] Step 5 (S350) can select optimal features from a feature map using LASSO (Least Absolute Shrinkage and Selection Operator), EN (Elastic-Net), and RF (Random Forest). Step 5 (S350) can select a certain number of features located at the top in the classification using RF as final medical image information systems (radiomics) features.

[0058] The step (S200) of extracting medical imaging information systems (radiomics) features according to an embodiment of the present invention can extract 854 medical imaging information systems (radiomics) features and 55 categories from a dataset consisting of abdominal and pelvic CT images of 79 patients.

[0059] The step of selecting significant features (S300) can generate a feature map containing six combinations. The six combinations are COR and LASSO, VIF and LASSO, COR and EN, VIF and EN, CON and RF, and VIF and RF. In the step of selecting significant features (S300), when COR is used for feature selection, 43 features are selected with LASSO, 35 with EN, and 10 with RF. In the step of selecting significant features (S300), when VIF is used for feature selection, 26 features are selected with LASSO, 30 with EN, and 10 with RF.

[0060] The step of training a prediction model (S400) can train the prediction model using medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome was alleviated after bariatric surgery, or medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome was not alleviated after bariatric surgery.

[0061] The entire dataset used in the prediction model was randomly split into training (70%) and validation (30%) subsets. The step of training the prediction model (S400) performed a random search using normalization, standardization, and 5 iterations of 5 cross-validations to optimize training and validation.

[0062] The prediction model can be any one of LR (logistic regression), RF (random forest), and SVM (support vector machine). The step of training the prediction model (S400) can compare the number of selected features through various combinations and evaluate the impact on the performance of the prediction model based on the number of features.

[0063] Table 3 below shows the prediction performance of each combination using the cross-validation method, and Figure 5 shows a graph comparing the AUC of each combination using the cross-validation method according to an embodiment of the present invention.

[0064] Referring to Table 1 and Figure 5, the LR model achieved the best performance with a training AUC of 0.744 and a test AUC of 0.800 in the COR and RF combination. The RF model recorded a training AUC of 1.000 and a test AUC of 0.675 with the same COR and RF combination. The SVM model showed the best performance in the VIF and RF combination, recording a training AUC of 0.819 and a test AUC of 0.783.

[0065] That is, the prediction model according to the present invention showed that the LR and SVM models exhibited excellent and acceptable prediction performance (AUCs of 0.800 and 0.783, respectively), and when using the COR and RF combination and the VIF and RF combination, the RF model showed fair prediction performance (AUC of 0.675). Preferably, the embodiment according to the present invention may use an LR prediction model combining absolute correlation (COR) and RF (Random Forest), which has the best performance.

[0066] The step (S500) for predicting remission of metabolic syndrome after bariatric surgery can predict remission of metabolic syndrome after bariatric surgery by inputting selected medical imaging information systems (radiomics) features into a prediction model.

[0067] FIG. 6 shows a configuration diagram of a device (100) for predicting remission of metabolic syndrome after bariatric surgery according to an embodiment of the present invention. Referring to FIG. 6, the configuration of the device (100) for predicting remission of metabolic syndrome after bariatric surgery is merely a simplified example. In one embodiment of the present invention, the device (100) for predicting remission of metabolic syndrome after bariatric surgery may include other configurations for performing the computing environment of the device (100), and only some of the disclosed configurations may constitute the device (100).

[0068] A device (100) for predicting remission of metabolic syndrome after bariatric surgery may include a processor (110) including one or more cores, memory (120), and a network (130).

[0069] The processor (110) may be composed of one or more cores and may include processors for data analysis and deep learning, such as a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), and a tensor processing unit (TPU) of a computing device. The processor (110) may read a computer program stored in memory (120) and perform data processing for machine learning according to one embodiment of the present disclosure. According to one embodiment of the present disclosure, the processor (110) may perform operations for learning a neural network. The processor (110) may perform calculations for learning a neural network, such as processing input data for learning in deep learning (DL), extracting features from input data, calculating errors, and updating the weights of the neural network using backpropagation. At least one of the CPU, GPGPU, and TPU of the processor (110) may process the learning of a network function. For example, a CPU and a GPGPU can work together to process the learning of a network function and data classification using the network function. Additionally, in one embodiment of the present disclosure, processors of a plurality of computing devices can be used together to process the learning of a network function and data classification using the network function. Furthermore, a computer program executed on a computing device according to one embodiment of the present disclosure may be a CPU, GPGPU, or TPU executable program.

[0070] The processor (110) can receive preoperative abdominal and pelvic CT images of one or more obese patients. The processor (110) can perform the step (S100) of receiving the aforementioned CT images.

[0071] The processor (110) can set a region of interest centered on the circumference of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images, and extract medical imaging information systems (radiomics) features within the said region of interest through medical imaging information systems (radiomics) analysis. The processor (110) can perform the step (S200) of extracting the aforementioned medical imaging information systems (radiomics) features.

[0072] The processor (110) can select statistically significant features among the medical image information systems (radiomics) features extracted using logistic regression analysis. The processor (110) can perform the step (S300) of selecting the aforementioned significant features.

[0073] The processor (110) can train a prediction model using medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome has been alleviated after bariatric surgery, or medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome has not been alleviated after bariatric surgery. The processor (110) can perform the step (400) of training the aforementioned prediction model.

[0074] The processor (110) can predict remission of metabolic syndrome after bariatric surgery by inputting selected medical imaging information systems (radiomics) features into a prediction model. The processor (110) can perform the step (S500) of predicting remission of metabolic syndrome after bariatric surgery as described above.

[0075] The memory (120) can store any form of information generated or determined by the processor (110) and any form of information received by the network (130).

[0076] The memory (120) may include at least one type of storage medium among flash memory type, hard disk type, multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic disk, and optical disk. The computing device (100) may operate in conjunction with web storage that performs the storage function of the memory (120) on the internet. The description of the memory described above is merely an example, and the present disclosure is not limited thereto.

[0077] The network (130) may use any type of known wired or wireless communication system. The network (130) may receive endoscopic images, etc. from a related device or related system.

[0078] The network (130) can transmit and receive information, user interfaces, etc., processed by the processor (110) through communication with other terminals. For example, the network (130) can provide a user interface generated by the processor (100) to a client (e.g., a user terminal). In addition, the network (130) can receive external input from a user authorized as a client and transmit it to the processor (110). At this time, the processor (110) can process operations such as outputting, modifying, changing, or adding information provided through the user interface based on the external input from the user received from the network (130).

[0079] Meanwhile, a device (100) for predicting remission of metabolic syndrome after bariatric surgery according to one embodiment of the present disclosure may include a server as a computing system that transmits and receives information through communication with a client. At this time, the client may be any type of terminal capable of accessing the server.

[0080] In an additional embodiment, the device (100) for predicting remission of metabolic syndrome after bariatric surgery may include any type of terminal that receives data resources generated from any server and performs additional information processing.

[0081] A computer program for predicting remission of metabolic syndrome after bariatric surgery, which is another embodiment of the present invention, may include the operation of receiving a CT image, the operation of extracting medical imaging information systems (radiomics) features, the operation of selecting significant features, the operation of training a prediction model, and the operation of predicting remission of metabolic syndrome after bariatric surgery. The computer program for predicting remission of metabolic syndrome after bariatric surgery may include instructions stored on a computer-readable storage medium to cause a computer to perform the following operations.

[0082] The operation of receiving CT images may receive preoperative abdominal and pelvic CT images of one or more obese patients. The operation of receiving CT images refers to the operation performed in the aforementioned step of receiving CT images.

[0083] The operation of extracting radiomics features involves setting a region of interest centered on the perimeter of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images, and extracting radiomics features within the said region of interest through radiomics analysis. The operation of extracting radiomics features refers to the operation performed in the aforementioned step of extracting radiomics features.

[0084] The operation of selecting significant features can select statistically significant features among the medical image information systems (radiomics) features extracted using logistic regression analysis. The operation of selecting significant features refers to the operation performed in the aforementioned step of selecting significant features.

[0085] The operation of training the prediction model can train the prediction model using selected medical imaging information systems (radiomics) features. The operation of training the prediction model refers to the operation performed in the aforementioned step of training the prediction model.

[0086] The action of predicting metabolic syndrome remission after bariatric surgery refers to the action performed in the aforementioned step of predicting metabolic syndrome remission after bariatric surgery.

[0087] FIG. 7 shows a schematic diagram of a computing environment according to an embodiment of the present invention.

[0088] Although the present disclosure has been described as generally being implementable by a computing device, those skilled in the art will understand that the present disclosure may be implemented in combination with computer-executable instructions and / or other program modules that can be executed on one or more computers, and / or as a combination of hardware and software.

[0089] Generally, a program module includes routines, programs, components, data structures, etc., that perform a specific task or implement a specific abstract data type. Furthermore, those skilled in the art will be well aware that the method of the present disclosure may be implemented in other computer system configurations, including single-processor or multi-processor computer systems, minicomputers, mainframe computers, as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, etc. (each of which may be connected to and operated with one or more associated devices).

[0090] The embodiments described in this disclosure may also be implemented in a distributed computing environment in which tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

[0091] Computers typically include various computer-readable media. Any medium accessible by a computer may be a computer-readable medium, and such computer-readable media include volatile and non-volatile media, transitory and non-transitory media, and removable and non-removable media. By example, but not limiting, computer-readable media may include computer-readable storage media and computer-readable transmission media. Computer-readable storage media include volatile and non-volatile media, transitory and non-transitory media, and removable and non-removable media implemented by any method or technique for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer-readable storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, DVD (digital video disk) or other optical disk storage devices, magnetic cassettes, magnetic tapes, magnetic disk storage devices or other magnetic storage devices, or any other media that can be accessed by a computer and used to store desired information.

[0092] Computer-readable transmission media typically include all information transmission media that implement computer-readable instructions, data structures, program modules, or other data, etc., on a modulated data signal, such as a carrier wave or other transport mechanism. The term modulated data signal means a signal in which one or more of the characteristics of the signal are set or modified to encode information within the signal. By example, not limiting, computer-readable transmission media include wired media, such as wired networks or direct-wired connections, and wireless media, such as acoustic, RF, infrared, and other wireless media. Any combination of the media described above is also considered to be within the scope of computer-readable transmission media.

[0093] An exemplary environment for implementing various aspects of the present disclosure, including a computer (1000), is shown, wherein the computer (1000) includes a processing unit (1020), a system memory (1030), and a system bus (1010). The system bus (1010) connects system components, including the system memory (1030) (but not limited thereto), to the processing unit (1020). The processing unit (1020) may be any processor among various commercial processors. Dual processors and other multiprocessor architectures may also be used as the processing unit (1020).

[0094] The system bus (1010) may be any of several types of bus structures that can be additionally interconnected to a local bus using any of the memory bus, peripheral bus, and various commercial bus architectures. The system memory (1030) includes read-only memory (ROM) (1034) and random access memory (RAM) (1032). The basic input / output system (BIOS) is stored in non-volatile memory (1034), such as ROM, EPROM, EEPROM, etc., and this BIOS includes basic routines that help transfer information between components within the computer (1000) at times such as during startup. The RAM (1032) may also include high-speed RAM, such as static RAM, for caching data.

[0095] The computer (1000) also includes an internal hard disk drive (HDD) (1050) (e.g., EIDE, SATA)—this internal hard disk drive (1050) may also be configured for external use within a suitable chassis (not shown)—a magnetic floppy disk drive (FDD) (1060) (e.g., for reading from or writing to a removable diskette), and an optical disk drive (1070) (e.g., for reading from a CD-ROM disk or reading from or writing to other high-capacity optical media such as a DVD). The hard disk drive (1050), the magnetic disk drive (1060), and the optical disk drive (1070) may each be connected to the system bus (1010) via a hard disk drive interface, a magnetic disk drive interface, and an optical drive interface. Interfaces for implementing external drives include at least one or both of the Universal Serial Bus (USB) and IEEE 1394 interface technologies.

[0096] These drives and associated computer-readable media provide non-volatile storage of data, data structures, computer-executable instructions, etc. In the case of a computer (1000), the drives and media correspond to storing any data in a suitable digital format. Although the description of computer-readable media above refers to HDDs, removable magnetic disks, and removable optical media such as CDs or DVDs, those skilled in the art will know that other types of computer-readable media, such as zip drives, magnetic cassettes, flash memory cards, cartridges, etc., may also be used in exemplary operating environments and that any of these media may contain computer-executable instructions for performing the methods of the present disclosure.

[0097] A number of program modules, including an operating system (1092), one or more application programs (1094), other program modules (1096), and a database (1098), may be stored in the drive and RAM (1032). All or part of the operating system, applications, modules, and / or data may also be cached in RAM (1032). It will be well known that the present disclosure may be implemented in various commercially available operating systems or combinations of operating systems.

[0098] The user can input commands and information into the computer (1000) through one or more wired / wireless input devices (1042), such as pointing devices like a keyboard and a mouse. Other input devices (not shown) may include a microphone, an IR remote control, a joystick, a game pad, a stylus pen, a touch screen, etc. These and other input devices are often connected to the processing unit (1020) via an input / output interface (1040) connected to the system bus (1010), but may also be connected via other interfaces such as a parallel port, an IEEE 1394 serial port, a game port, a USB port, an IR interface, etc.

[0099] A monitor or other type of display device is also connected to the system bus (1010) via an interface such as a video adapter. In addition to the monitor, the computer generally includes other peripheral output devices (not shown), such as speakers, a printer, and so on.

[0100] A computer (1000) may operate in a networked environment using a logical connection to one or more remote computers, such as remote computer(s) (1082), via wired and / or wireless communication. The remote computer(s) (1082) may be a workstation, a computing device computer, a router, a personal computer, a portable computer, a microprocessor-based entertainment device, a peer device, or other conventional network node, and may include many or all of the components generally described for the computer (1000). The logical connection includes a wired / wireless connection to a local area network (LAN) and / or a larger network, e.g., a wide area network (WAN). Such LAN and WAN networking environments are common in offices and companies and facilitate enterprise-wide computer networks, such as intranets, all of which may be connected to a global computer network, e.g., the Internet.

[0101] When used in a LAN networking environment, the computer (1000) is connected to a local network (not shown) via a wired and / or wireless communication network interface or adapter (not shown). The adapter (not shown) may facilitate wired or wireless communication to the LAN (not shown), and the LAN (not shown) may also include a wireless access point installed therein to communicate with the wireless adapter (not shown). When used in a WAN networking environment, the computer (1000) may include a modem (not shown), be connected to a communication computing device on the WAN (not shown), or have other means to establish communication over the WAN (not shown), such as through the Internet. The modem (not shown), which may be internal or external and wired or wireless, is connected to the system bus (1010) via a serial port interface (not shown). In a networked environment, the program modules described for the computer (1000) or parts thereof may be stored in a remote memory / storage device (not shown). You will be well aware that the illustrated network connection is exemplary and that other means of establishing communication links between computers can be used.

[0102] The computer (1000) operates to communicate with any wireless device or object that is deployed and operated via wireless communication, for example, a printer, scanner, desktop and / or portable computer, PDA (portable data assistant), communication satellite, any equipment or place associated with a wireless detectable tag, and a telephone. This includes at least Wi-Fi and Bluetooth wireless technologies. Accordingly, the communication may be a predefined structure as in a conventional network, or simply ad hoc communication between at least two devices.

[0103] Wi-Fi (Wireless Fidelity) enables connectivity to the Internet and other sources without wires. Wi-Fi is a wireless technology, similar to a cell phone, that allows devices, such as computers, to transmit and receive data indoors and outdoors—that is, anywhere within the coverage area of ​​a base station. Wi-Fi networks use a wireless technology called IEEE 802.11 (a, b, g, etc.) to provide secure, reliable, and high-speed wireless connections. Wi-Fi can be used to connect computers to each other, to the Internet, and to wired networks (using IEEE 802.3 or Ethernet). Wi-Fi networks can operate in unlicensed 2.4 and 5 GHz wireless bands, for example, at data rates of 11 Mbps (802.11a) or 54 Mbps (802.11b), or in products that include both bands (dual band).

[0104] Those skilled in the art of the present disclosure will understand that information and signals may be represented using any various different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced in the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

[0105] Those skilled in the art will understand that the various exemplary logic blocks, modules, processors, means, circuits, and model steps described in connection with the embodiments disclosed herein may be implemented by electronic hardware, various forms of programs or design code (referred to herein as software for convenience), or a combination of all such. To clearly illustrate this interoperability between hardware and software, various exemplary components, blocks, modules, circuits, and steps have been generally described above in relation to their functions. Whether such functions are implemented as hardware or software depends on the design constraints imposed on the specific application and the overall system. Those skilled in the art may implement the functions described in various ways for each specific application, but such implementation decisions should not be interpreted as being outside the scope of the present disclosure.

[0106] The various embodiments presented herein may be implemented as methods, devices, or articles manufactured using standard programming and / or engineering techniques. The term "article manufactured" includes a computer program, a carrier, or a medium accessible from any computer-readable storage device. For example, computer-readable storage media include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical discs (e.g., CDs, DVDs, etc.), smart cards, and flash memory devices (e.g., EEPROMs, cards, sticks, key drives, etc.). Additionally, the various storage media presented herein include one or more devices and / or other machine-readable media for storing information.

[0107] It should be understood that the specific order or hierarchy of steps in the presented processes is merely an example of exemplary approaches. It should be understood that, based on design priorities, the specific order or hierarchy of steps in the processes may be rearranged within the scope of this disclosure. The appended method claims provide elements of various steps in a sample order, but do not imply being limited to the specific order or hierarchy presented.

[0108] Description of the presented embodiments is provided so that a person skilled in the art may use or practice the present disclosure. Various modifications to these embodiments will be apparent to a person skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present disclosure is not limited to the embodiments presented herein, but should be interpreted in the broadest possible scope consistent with the principles and novel features presented herein.

[0109] The embodiments of the present invention described above are not implemented only through devices and methods, but may also be implemented through a program that realizes a function corresponding to the configuration of the embodiments of the present invention, or a recording medium on which such a program is recorded. Such a recording medium may be executed not only on a server but also on a user terminal.

[0110] Although embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art using the basic concept of the present invention as defined in the following claims also fall within the scope of the present invention.

Claims

Claim 1 A method for predicting remission of metabolic syndrome after bariatric surgery performed by a computing device, comprising: receiving preoperative abdominal and pelvic CT images of one or more obese patients; setting a region of interest centered on the circumference of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images and extracting radiomics features within the region of interest through radiomics analysis; selecting statistically significant features among the extracted radiomics features using logistic regression analysis; and inputting the selected radiomics features into a prediction model to predict remission of metabolic syndrome after bariatric surgery; wherein the step of extracting radiomics features is to extract a region of interest centered on the entire abdominal muscle area including the psoas muscle, paravertebral muscles, quadratus lumborum, rectus abdominis, transversus abdominis, or internal and external oblique muscles in each slice across the entire third lumbar vertebra (L3). Claim 2 A method according to claim 1, wherein the obese patient is defined as a patient having at least one of the following: blood pressure of 130 / 85 mmHg or higher or using antihypertensive drugs, fasting blood glucose level of 126 mg / dL or higher, hemoglobin A1c (HbA1c) level of 6.5% or higher or using diabetes drugs, and a waist-to-hip ratio (WHR) of 0.95 or higher for women or 1.03 or higher for men. Claim 3 delete Claim 4 In claim 1, the step of extracting the medical imaging information systems (radiomics) features comprises using linear interpolation to set the spatial resolution of the CT image to 1*1*1mm 3 A method of resampling. Claim 5 A method according to claim 1, wherein the step of selecting the significant features comprises a first step of generating a feature map using univariate logistic regression analysis. Claim 6 A method according to claim 5, wherein the step of selecting the significant features further comprises a second step of verifying the possibility of association between the selected significant features and metabolic syndrome remission after bariatric surgery. Claim 7 A method according to claim 6, wherein the step of selecting the significant features further comprises a third step of evaluating the redundancy or correlation between the significant features verified as potentially associated in the second step using the absolute correlation (COR) or variance inflation factor (VIF). Claim 8 A method according to claim 7, wherein the step of selecting the significant features further comprises a fourth step of evaluating the relevance or importance of the significant features verified as having a possible association in the second step. Claim 9 A method according to claim 8, wherein the first step is to reflect the verification or evaluation results performed in the second to fourth steps into the feature map. Claim 10 In claim 8, the step of selecting the significant features further comprises a fifth step of selecting optimal features from the feature map using LASSO (Least Absolute Shrinkage and Selection Operator), EN (Elastic-Net), and RF (Random Forest). Claim 11 A method according to claim 10, wherein the third step above uses absolute correlation (COR) and the fifth step above uses RF (Random Forest). Claim 12 The method of claim 1 further comprises the step of training a prediction model using medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome has been alleviated after bariatric surgery, or medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome has not been alleviated after bariatric surgery. Claim 13 In paragraph 12, the method wherein the prediction model is any one of LR (logistic regression), RF (random forest), and SVM (support vector machine). Claim 14 A device for predicting remission of metabolic syndrome after bariatric surgery, comprising one or more cores; and memory; the processor receives preoperative abdominal and pelvic CT images of one or more obese patients, sets a region of interest centered on the circumference of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images, extracts medical imaging information systems (radiomics) features within the region of interest through medical imaging information systems (radiomics) analysis, selects statistically significant features among the extracted medical imaging information systems (radiomics) features using logistic regression analysis, trains a prediction model using medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome has been remissioned after bariatric surgery or medical imaging information systems (radiomics) features extracted from preoperative abdominal and pelvic CT images of obese patients whose metabolic syndrome has not been remissioned after bariatric surgery, and inputs the selected medical imaging information systems (radiomics) features into the prediction model to predict remission of metabolic syndrome after bariatric surgery, wherein the extraction of the medical imaging information systems (radiomics) features is a third A device for extracting a region of interest centered on the entire abdominal muscle area including the psoas muscle, paraspinal muscles, quadratus lumborum, rectus abdominis, transverse abdominis, or internal and external oblique muscles in each slice across the entire lumbar spine (L3). Claim 15 A computer program stored on a computer-readable storage medium comprising instructions that cause a computer to perform the following operations, wherein the operations include: receiving preoperative abdominal and pelvic CT images of one or more obese patients; setting a region of interest centered on the perimeter of the entire abdominal muscle area in each slice of the received abdominal and pelvic CT images and extracting radiomics features within the region of interest through radiomics analysis; selecting statistically significant features among the extracted radiomics features using logistic regression analysis; and training a prediction model with the selected radiomics features; wherein the operation of extracting radiomics features is to extract a region of interest centered on the entire abdominal muscle area including the psoas muscle, paravertebral muscles, quadratus lumborum, rectus abdominis, transverse abdominis, or internal and external oblique muscles in each slice across the entire third lumbar vertebra (L3).

Images