Prediction device and prediction model for predicting probability of complications after pulmonary lobectomy or sub-pulmonary lobectomy and application of prediction device and prediction model

Abstract

The invention relates to the technical field of medical treatment, in particular to a prediction device and a prediction model for predicting the probability of complications after pulmonary lobectomy or sub-pulmonary lobectomy and application of the prediction device and the prediction model. Aiming at prediction errors of an AS method and a QCT traditional mode at present, the invention finds that FEV1QCT (postoperative FEV1 value estimated based on QCT) is an independent prediction factor of short-term pulmonary complications after lung cancer lobectomy or sub-lobectomy through research, and a model based on QCT has relatively high prediction performance; the specific technical scheme of the prediction device for predicting the probability of the complications after the pulmonary lobe or sub-pulmonary lobe resection, the prediction model comprising the column diagram and used for predicting the probability of the complications after the pulmonary lobe or sub-pulmonary lobe resection and the application of the prediction model are obtained and provided, and the prediction accuracy of the PC after the pulmonary segmental resection is improved. Personalized selection of appropriate operation schemes and postoperative management schemes is facilitated, and the living quality and prognosis of patients are improved.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to a prediction index and a prediction model for predicting the probability of complications after lobectomy or sublobar resection of lung cancer. Background Art

[0002] Lung cancer is the leading cause of cancer-related death worldwide. Historically, lobectomy has been considered a treatment option for early-stage non-small cell lung cancer (NSCLC). In recent years, several retrospective analyses have reported that for early-stage peripheral NSCLC, sublobar resection is no less effective than lobectomy in terms of oncological prognosis and may also preserve more lung function after surgery. However, lung tissue resection can lead to decreased lung function and significantly increase the risk of postoperative complications, particularly pulmonary complications (PC), such as pulmonary infection, persistent air leak, pulmonary infarction, and pleural effusion. Postoperative pulmonary complications are a major factor affecting treatment efficacy and prognosis, and severe pulmonary complications can sometimes even lead to death. It has been reported that the occurrence of postoperative pulmonary complications is closely correlated with residual lung function. Therefore, preoperative assessment of postoperative pulmonary function in lung cancer patients may help predict pulmonary complications after lobectomy or sublobar resection, which is of great significance for improving patients' quality of life after surgery.

[0003] Currently, the most widely used method for predicting lung function after pneumonectomy is the anatomical segmentation (AS) method. This method predicts residual lung function after pneumonectomy based on preoperative pulmonary function tests (PFTs) and clinical anatomical segmentation. The AS method assumes that each lung segment contributes equally to lung function. Therefore, when the function of each lung segment varies due to lung tissue development or disease, the AS method can produce significant errors.

[0004] Quantitative computed tomography (QCT) can identify and measure functional lung tissue, avoiding the inclusion of nonfunctional lung tissue. Multiple studies have demonstrated a good correlation between QCT lung volume indices and PFT results. However, previous studies have focused primarily on QCT findings at the whole lung, left and right lung, and lobar levels, inevitably leading to bias in quantifying residual lung function after resection. Currently, advanced QCT lung volume analysis software can automatically identify and segment lobes and even segments, then measure the functional volume of each lobe and segment. This convenient and rapid measurement eliminates manual delineation and individual bias, laying the foundation for improving the accuracy of postoperative lung function prediction. Segmental QCT volume analysis has demonstrated significant value in preoperative planning for segmentectomy. However, research on its application in predicting lung function and complications after sublobar resection is relatively limited. Summary of the Invention

[0005] In order to fill the above technical gap, the present invention has found that FEV1QCT (Postoperative FEV1 value estimated based on QCT) is an independent predictor of short-term pulmonary complications after lobectomy or sublobar resection for lung cancer, and the QCT-based model has high predictive performance; thereby providing a prediction device, prediction model and application thereof for predicting the probability of complications after lobectomy or sublobar resection.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] The present invention provides a prediction device for predicting the probability of complications after lobectomy or sublobar resection, comprising a data acquisition module and a calculation module; the data acquisition module is used to obtain the patient's preoperative forced expiratory volume in the first second, the functional lung volume and the total functional lung volume removed by CT scan; the calculation module is used to calculate the postoperative forced expiratory volume in the first second FEV1 based on the data acquired by the data acquisition module QCT , FEV1 QCT Used to predict the probability of complications after lobectomy or sublobar resection.

[0008] Furthermore, the calculation module calculates the postoperative forced expiratory volume in the first second FEV1QCT by a QCT method.

[0009] Furthermore, FEV1 QCT = preoperative FEV1 × [1-(RFLV / TFLV)]; where preoperative FEV1 is the forced expiratory volume in 1 second obtained through preoperative pulmonary function testing, RFLV is the resected functional lung volume obtained through CT image analysis, and TFLV is the total functional lung volume obtained through CT image analysis.

[0010] Furthermore, the present invention also provides a prediction model for predicting the probability of complications after lobectomy or sublobectomy, comprising a prediction module for predicting the probability of complications after lobectomy or sublobectomy based on the forced expiratory volume in the first second FEV1 as described above. QCT Predicting the probability of complications after lobectomy or sublobar resection.

[0011] Furthermore, the prediction module includes a nomogram tool module.

[0012] Furthermore, the nomogram tool module includes:

[0013] The first line is a score scale with a score range of 0 to 100;

[0014] The second row is the patient's age, including age <61 and ≥61, with corresponding scores of 0 and 8, respectively;

[0015] The third row is the surgical method, including sublobar resection and lobectomy, with corresponding scores of 0 and 0.5, respectively;

[0016] The fourth row is the ratio of RFLV to TFLV, ranging from 5 to 17.5, corresponding to a score of 0 to 7, where 12.5 corresponds to 5 points;

[0017] The fifth line is FEV1 QCT The value range is 5 to 0.5, and the corresponding score is 0 to 100;

[0018] The sixth row is the total score, which is the sum of the scores from the second to the fifth row, ranging from 0 to 120;

[0019] The seventh row is the probability of pulmonary complications, with a total score range of 72 to 104.5, corresponding to a probability of 0.1 to 0.8.

[0020] Furthermore, the present invention also provides an electronic device comprising at least one processor and a memory in communication with the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to run the data acquisition module and the calculation module of the prediction device for predicting the probability of complications after lobectomy or sublobectomy as described above, and obtain the forced expiratory volume in the first second FEV1 after surgery. QCT .

[0021] Furthermore, the present invention also provides an application of the aforementioned prediction model in predicting the probability of complications after lobectomy or sublobar resection for non-diagnostic purposes.

[0022] The present invention has the following beneficial effects:

[0023] 1. FEV1 of the present invention QCT Provides quantitative parameters for predicting postoperative complications based on FEV1 QCT The prediction model and its corresponding nomogram provide valuable clinical applications for risk stratification and personalized perioperative management.

[0024] 2. Although QCT-based prediction of postoperative pulmonary function can be used to predict complications after lobectomy, the relationship between QCT parameters and PC after segmental resection remains largely unexplored in existing studies. This may be because most QCT techniques used are primarily based on lobe-level quantification and manual measurement, which can lead to inaccurate segmental estimation. The present invention, however, enables precise segmental-level quantitative analysis, effectively resolving this issue and improving the accuracy of PC prediction after segmental resection.

[0025] 3. Currently, models based on different types of patient data are rarely used. To improve the clinical application value, a nomogram was established based on the prediction model. Its high performance will help to personalize the selection of appropriate surgical plans, postoperative management, and improve patients' quality of life and prognosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are intended to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0027] Figure 1 A diagram showing the patient screening process of the present invention;

[0028] Figure 2 This is an example diagram of post-processing of CT reconstructed images of a first patient according to the present invention;

[0029] Figure 3 This is an example diagram of post-processing of CT reconstructed images of a second patient according to the present invention;

[0030] Figure 4 This is an example diagram of post-processing of CT reconstructed images of a third patient of the present invention;

[0031] Figure 5 It is the ROC curve analysis diagram of the present invention;

[0032] Figure 6 It is a nomogram based on the clinical-QCT model of the present invention; DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0034] Example 1

[0035] Example 1 provides a prediction device for predicting the probability of complications after lobar or sublobar resection, including a data acquisition module and a calculation module; the data acquisition module is used to obtain the patient's preoperative forced expiratory volume in the first second, functional lung volume resected by CT scan, and total functional lung volume; the calculation module is used to calculate the postoperative forced expiratory volume in the first second FEV1QCT based on the data acquired by the data acquisition module, and FEV1QCT is used to predict the probability of complications after lobar or sublobar resection.

[0036] Furthermore, the calculation module calculates the postoperative forced expiratory volume in the first second FEV1QCT by a QCT method.

[0037] Furthermore, FEV1QCT = preoperative FEV1 × [1-(RFLV / TFLV)]; where preoperative FEV1 is the forced expiratory volume in one second obtained through preoperative pulmonary function testing, RFLV is the resected functional lung volume obtained by CT image analysis, and TFLV is the total functional lung volume obtained by CT image analysis.

[0038] The contents of the data acquisition module and calculation module in this implementation are obtained based on clinical experimental research. Specific clinical experimental research includes:

[0039] 1 Research Methodology:

[0040] This retrospective study was approved by the Institutional Review Board of the Union Hospital Affiliated to Fujian Medical University (No. 2023QH016). Due to the retrospective nature of this study, written informed consent was not required.

[0041] 1.1 Patients

[0042] A total of 141 patients with lung cancer who underwent thoracoscopic lobectomy or sublobar resection in our hospital were consecutively enrolled. All patients had preoperative chest CT scan and PFT results. Exclusion criteria: 1) patients with a history of thoracotomy; 2) patients with lung resection combined with lung transplantation; 3) patients with unhealed preoperative lung infection; 4) patients with severe chest deformity; 5) patients with an interval of more than one week between CT scan and PFT; 6) patients with severe motion artifacts on CT images; 7) patients with incomplete clinical data. Finally, 79 patients were included in the present invention ( Figure 1 ).

[0043] We used the hospital medical record database to retrieve data on the patients' clinical characteristics, including age, sex, height, weight, smoking history, pulmonary function test results, surgical method, and postoperative complications. One-month postoperative follow-up was performed to determine the occurrence of postoperative complications. Postoperative complications included pulmonary infection, intrathoracic hemorrhage, pneumothorax, atelectasis, persistent air leak, pleural effusion, acute pulmonary embolism, and respiratory failure. Postoperative pulmonary complications were determined according to the Clavien-Dindo classification. The Clavien-Dindo classification categorizes postoperative complications into five different grades. Grade 1: Abnormal changes requiring antiemetics, antipyretics, analgesics, electrolytes, and physical therapy. Wound infections requiring open drainage but not requiring surgical, endoscopic, or radiological treatment were included. Grade 2: Requirement of medical treatment with medications other than those permitted for grade 1 complications. Blood transfusion and total parenteral nutrition were also included. Grade 3: Requirement of surgical, endoscopic, or radiological intervention. Grade 3a: Intervention not requiring general anesthesia. Grade 3b: Intervention requiring general anesthesia. Grade 4: Life-threatening complications requiring intensive care unit (IC / ICU) management (including central nervous system complications). Grade 4a: Single organ dysfunction (including dialysis). Grade 4b: Multiple organ dysfunction. Grade 5: Patient death. In this application, PC+ refers to patients who develop pulmonary complications of grade 1 or higher.

[0044] Methods: Seventy-nine patients with lung cancer who underwent thoracoscopic lobectomy or sublobar resection underwent chest computed tomography (CT) and pulmonary function tests (PFTs) one week before surgery. Patients were divided into a PC+ group (n=16) and a PC- group (n=63) based on whether pulmonary complications (PC) occurred within 30 days after surgery. Total functional lung volume (TFLV) and resected functional lung volume (RFLV) were calculated using the QCT method. Predicted postoperative pulmonary function indices, including total lung capacity (TLC), forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), and diffusing capacity for carbon monoxide (DLCO), were calculated using QCT and anatomical segmentation (AS) methods. Factors associated with pulmonary complications were screened by univariate analysis to construct clinical, clinical-PFT, clinical-AS, and clinical-QCT models.

[0045] 1.2 Pulmonary function test

[0046] Patients underwent PFT within one week before surgery. Measurements included forced expiratory volume in one second (FEV1), forced vital capacity (FVC), total lung capacity (TLC), and diffusing capacity for carbon monoxide (DLCO).

[0047] 1.3 CT examination plan

[0048] CT examination was performed using a 256-slice CT scanner (Revolution CT, GE Healthcare, USA). The patient first underwent a plain scan, followed by a dual-phase enhanced scan, and the plain scan images were used for QCT analysis. The plain scan parameters were as follows: the patient was in the supine position with both hands raised, and the entire lung was scanned from the apex to the diaphragm during end-inspiratory breath-hold. The tube voltage was 120 kV, the automatic tube current was 50-500 mA, the noise index (NI) was 12.0, the rotation time was 0.5 s, the pitch was 0.984:1, and the slice thickness was 5.0 mm. The reconstruction parameters were as follows: adaptive statistical iterative reconstruction technology (ASIR-V) 50%, slice thickness 1.25 mm, inner lung core, window width 1500 HU, window position -600 HU.

[0049] 1.4 Image Analysis

[0050] The reconstructed images from the plain CT scans were imported into artificial intelligence (AI) analysis software (InferVisual SurgeryPlanning, Infer Medical Technology Co., Ltd., China). The software automatically segmented the lobes and segments of the bilateral lungs to obtain the total functional lung volume (TFLV) and the resected functional lung volume (RFLV). The automatic segmentation and calculation results of the functional lung volumes of the lobes and segments of three patients are listed in the present invention, as shown in the attached figure. Figure 2 -Attached Figure 4 As shown in the attached Figure 2-4The red arrow indicates the location of the lesion. (A) Preoperative chest CT lung axial window image; (B) 3D images generated by quantitative CT lung volume analysis software.

[0051] Figure 2 A 73-year-old woman presented with severe short-term PC (grade III) and underwent preoperative chest CT scan and quantitative CT pulmonary function analysis. QCT :0.96L; FVC QCT :0.99L.

[0052] Figure 3 The patient was a 79-year-old woman. Preoperative chest CT scan and quantitative CT pulmonary function analysis showed mild short-term PC (grade I). FEV1 QCT :1.53L; FVC QCT :1.79L.

[0053] Figure 4 The patient was a 57-year-old male with no short-term pulmonary complications after surgery. He had a preoperative chest CT scan and quantitative CT pulmonary function analysis. QCT :2.29L; FVC QCT :2.69L.

[0054] 1.5 Prediction of postoperative pulmonary function

[0055] Postoperative lung function was estimated using the AS method. The postoperative prediction formula of the AS method is as follows:

[0056] TLC AS =Preoperative TLC×[1-(na) / (42-a)],

[0057] FVC AS =Preoperative FVC×[1-(na) / (42-a)],

[0058] FEV1 AS =Preoperative FEV1×[1-(na) / (42-a)],

[0059] DLCO AS =Preoperative DLCO×[1-(na) / (42-a)],

[0060] where n is the total number of subsegments in the resected lobe, assuming that the number of subsegments in the right upper, middle, and lower lobes is 6, 4, and 12, respectively, and the number of subsegments in the left upper and lower lobes is 10, and a is the number of subsegments obstructed by the tumor.

[0061] The postoperative pulmonary function indices predicted by QCT were calculated as follows:

[0062] TLC QCT = preoperative TLC × [1-(RFLV / TFLV)],

[0063] FVC QCT = preoperative FVC × [1-(RFLV / TFLV)],

[0064] FEV1 QCT = preoperative FEV1 × [1-(RFLV / TFLV)],

[0065] DLCO QCT = preoperative DLCO × [1-(RFLV / TFLV)].

[0066] 1.6 Statistical analysis

[0067] Data were analyzed using the SPSS statistical software package (version 27, SPSS Inc., Chicago, USA). Categorical variables were expressed as numbers and percentages and compared using the chi-square test or Fisher's exact test. Continuous variables were expressed as mean ± standard deviation and analyzed using the Student t test and Mann-Whitney test. Outcomes were categorized according to surgical approach. Multivariate logistic regression analysis and Poisson regression analysis with robust variance were used to identify independent risk factors for pulmonary complications. Univariate analysis was performed to screen factors associated with pulmonary complications to construct a model including clinical, clinical-PFT, clinical-AS, and clinical-QCT. Receiver operating characteristic (ROC) analysis was performed to evaluate predictive performance. A nomogram was drawn using R 4.1.3 software to visually display the clinical-QCT model. A two-sided P < 0.05 was considered statistically significant.

[0068] 2 Results

[0069] 2.1 Patients

[0070] Patient characteristics are summarized in Table 1. The cohort included 34 men and 45 women (mean age, 59.66 ± 12.85 years; range, 23–84 years). Surgical procedures included lobectomy (n = 29) and sublobar resection (n = 50). Patients were divided into a PC+ group (n = 16) and a PC− group (n = 63) based on the development of pulmonary complications within 30 days after surgery.

[0071] Table 1 Patient characteristics (n=79)

[0072] feature value Gender (n[%]) male 34(43.04％) female 45(56.96％) Age (years, mean ± standard deviation) 59.66±12.85 Height (cm, median [interquartile range Q1, Q3]) 162.00(150.00，174.00) Body weight (kg, mean ± standard deviation) 60.64±8.78 Smoking index (median [Q1, Q3]) 0(0，0) Hypertension (n[%]) 29(36.71％) Diabetes (n[%]) 13(16.46％) Surgical method (n[%]) Lobectomy 29(36.70％) Sublobar resection 50(63.30％) Postoperative respiratory complications (n[%]) PC- 63(79.75％) PC+ 16(20.25％) lung infection 7(43.75％) Acute pulmonary embolism 1(6.25％) Pulmonary infection with chest hemorrhage 2(12.50％) Pulmonary infection with pneumothorax 1(6.25％) Pulmonary infection with atelectasis 1(6.25％) Pulmonary infection combined with acute pulmonary embolism 1(6.25％) Pulmonary infection with persistent air leak 1(6.25％) Pulmonary infection with respiratory failure 1(6.25％) Pulmonary infection with pneumothorax and pleural effusion 1(6.25％) Postoperative hospital stay (days, median [Q1, Q3]) 4.00(3.00，6.00) PC- 4.00(3.00，5.00) PC+ 10.00(6.00，16.00) Clavien-Dindo classification in PC+ (n[%]) Level I 9(56.25％) Level II 4(25.00％) Level III 2(12.5％) Level IV 1(6.25％) V-level none

[0073] PC-, no pulmonary complications, PC+, with pulmonary complications. Smoking index is the number of cigarettes smoked per day multiplied by the number of years of smoking.

[0074] Comparison of clinical, PFT, AS, and QCT results between the PC+ and PC- groups

[0075] The comparison of clinical, PFT, AS and QCT results between the two groups is shown in Table 2. The age, number of resected lung segments, and RFLV / TFLV of patients in the PC+ group were significantly higher than those in the PC- group (P < 0.05); FEV1 and FVC in the PC+ group were significantly higher than those in the PC- group (P < 0.05). QCT , FEV1 QCT , FVC AS and FEV1 AS lower than those in the PC-group (P<0.05).

[0076] Table 2 Comparison of clinical, PFT, AS and QCT results between the two groups

[0077]

[0078] AS, anatomic segmentation; QCT, quantitative computed tomography; PC+, presence of pulmonary complications; TFLV, total functional lung volume; RFLV, resected functional lung volume; TLC, total lung capacity; FVC, forced vital capacity; FEV1, forced expiratory volume in 1 second; DLCO, diffusing capacity for carbon monoxide. The smoking index was calculated as the number of cigarettes smoked per day multiplied by the number of years of smoking.

[0079] 2.3 Independent risk factors for PC+

[0080] The results of univariate logistic regression, multivariate logistic regression and Poisson regression with robust variance are shown in Table 3. Univariate analysis showed that age, surgical method, FVC, FEV1, number of resected lung segments, FVC AS , FEV1 AS , RFLV / TFLV, FVC QCT , FEV1 QCT Multivariate analysis and Poisson regression showed that FEV1 QCT (odds ratio: 0.043, P = 0.001; relative risk: 0.261, P < 0.001) was an independent predictor of PC+.

[0081] Table 3 Univariate logistic regression, multivariate logistic regression, and Poisson regression with robust variance analysis of factors associated with short-term postoperative pulmonary complications

[0082]

[0083] FVC, forced vital capacity; FEV1, forced expiratory volume in 1 second; AS, anatomic segmentation; QCT, quantitative computed tomography; PC+, presence of pulmonary complications; TFLV, total functional lung volume; RFLV, resected functional lung volume; 95% CI, 95% confidence interval; OR, odds ratio; RR, relative risk.

[0084] 2.4 Prediction Performance

[0085] The ROC analysis of the four prediction models is summarized in Table 4 and Figure 5 The clinical model included age and surgical method, the clinical-PFT model included age, surgical method and FEV1, and the clinical-AS model included age, surgical method and FEV1. AS , the clinical-QCT model includes age, surgical method, RFLV / TFLV and FEV1 QCT Comparison of the area under the curve (AUC): Clinical-QCT (0.823) > Clinical-PFT (0.806) > Clinical-AS (0.801) > Clinical (0.757) model. As can be seen from the above, the clinical-QCT model has the best predictive performance.

[0086] Based on the clinical-QCT model, the present invention constructs a nomogram prediction model for predicting the probability of complications after lobectomy or sublobar resection, as shown in the attached figure. Figure 6 shown.

[0087] Table 4 ROC analysis of prediction model

[0088]

[0089]

[0090] ROC, receiver operating characteristic; AUC, area under the curve; 95% CI, 95% confidence interval; PFT, pulmonary function test; AS, anatomic segmentation; QCT, quantitative computed tomography.

[0091] Comparison of clinical, PFT, AS, and QCT findings between the PC+ and PC- groups (based on lobectomy versus sublobar resection)

[0092] The stratified comparison of clinical, PFT, AS and QCT results between the two groups according to surgical method is shown in Table 5. For patients who underwent lobectomy, the age of the PC+ group was older than that of the PC- group (P < 0.05); the FEV1 of the PC+ group was higher than that of the PC- group (P < 0.05); QCT and DLCO QCT For patients who underwent sublobar resection, the FEV1 in the PC+ group was lower than that in the PC- group (P<0.05). AS and FEV1 QCT Compared with the PC- group (P < 0.05), the univariate logistic regression analysis according to surgical approach is shown in Table 6. For patients undergoing sublobar resection, univariate analysis showed that FEV1AS and FEV1QCT were associated with PC+.

[0093] Table 5 Comparison of clinical, PFT, AS, and QCT results between the two groups (based on lobectomy and sublobar resection)

[0094]

[0095]

[0096] AS, anatomic segmentation; QCT, quantitative computed tomography; PC-, no pulmonary complications; PC+, pulmonary complications; TFLV, total functional lung volume; RFLV, resected functional lung volume; TLC, total lung capacity; FVC, forced vital capacity; FEV1, forced expiratory volume in 1 second; DLCO, diffusing capacity for carbon monoxide. The smoking index was calculated as the number of cigarettes smoked per day multiplied by the number of years of smoking.

[0097] Table 6 Univariate logistic regression analysis of factors associated with short-term postoperative pulmonary complications (based on lobectomy and sublobar resection)

[0098]

[0099] FEV1, forced expiratory volume in 1 second; AS, anatomic segmentation; QCT, quantitative computed tomography; DLCO, diffusing capacity for carbon monoxide; 95% CI, 95% confidence interval; OR, odds ratio.

[0100] In summary, the present invention investigates the value of postoperative pulmonary segment function indices estimated by preoperative quantitative computed tomography (QCT) in predicting short-term complications after lobectomy or sublobar resection in patients with lung cancer. QCT ) is an independent predictor of pulmonary complications (PC), and the clinical model based on QCT (clinical-QCT model) showed higher predictive performance than the clinical model based on anatomical segmentation (clinical-AS model) and the clinical model based on pulmonary function test (clinical-PFT model). The predictive performance was further improved when all factors related to pulmonary complications were combined. In existing studies, although some postoperative pulmonary function predicted by QCT has been used to predict short-term complications after lobectomy, the relationship between QCT parameters and pulmonary complications after subsegmentectomy is still rarely involved. This may be because most commonly used QCT techniques are mainly based on quantitative analysis and manual measurement at the lobe level, which may lead to inaccurate results in lung segment assessment.

[0101] The present study showed that the forced expiratory volume in one second (FEV1) was lower in the pulmonary complication-positive group (PC+ group) than in the pulmonary complication-negative group (PC- group). In addition, preoperative FEV1 was an independent predictor of short-term postoperative pulmonary complications. In this study, the clinical-QCT model combining clinical indicators and QCT had the highest efficacy. A nomogram was established based on this model, such as Figure 6 The high performance of our clinical-QCT prediction model will further facilitate personalized selection of appropriate surgical options, postoperative management, and improve patients' quality of life and prognosis.

[0102] Example 2

[0103] This embodiment provides an electronic device, comprising at least one processor and a memory in communication with the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to run the data acquisition module and the calculation module of the device for predicting the probability of complications after lobectomy or sublobectomy described in Example 1, and obtain the forced expiratory volume in the first second FEV1 after surgery. QCT .

[0104] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A prediction device for predicting the probability of complications after lobectomy or sublobar resection, characterized in that: It includes a data acquisition module and a calculation module; The data acquisition module is used to obtain the patient's preoperative forced expiratory volume in one second, functional lung volume removed by CT scan, and total functional lung volume; The calculation module is used to calculate the forced expiratory volume in the first second FEV1 after surgery based on the data obtained by the data acquisition module QCT , FEV1 QCT Used to predict the probability of complications after lobectomy or sublobar resection.

2. The prediction device for predicting the probability of complications after lobectomy or sublobar resection according to claim 1, characterized in that: The calculation module calculates the forced expiratory volume in the first second FEV1 after surgery using the QCT method QCT .

3. The prediction device for predicting the probability of complications after lobectomy or sublobar resection according to claim 2, characterized in that: FEV1 QCT = preoperative FEV1 × [1-(RFLV / TFLV)]; Among them, preoperative FEV1 is the forced expiratory volume in the first second obtained through pulmonary function test before surgery, RFLV is the resected functional lung volume obtained by CT image analysis, and TFLV is the total functional lung volume obtained by CT image analysis.

4. A prediction model for predicting the probability of complications after lobectomy or sublobar resection, characterized in that: A prediction module is included, wherein the prediction module is used to determine the forced expiratory volume in the first second FEV1 after surgery according to claim 1. QCT Predicting the probability of complications after lobectomy or sublobar resection.

5. The prediction model for predicting the probability of complications after lobectomy or sublobar resection according to claim 4, characterized in that: The prediction module includes a nomogram tool module.

6. The prediction model for predicting the probability of complications after lobectomy or sublobar resection according to claim 5, characterized in that: The nomogram tool module includes: The first line is a score scale with a score range of 0 to 100; The second row is the patient's age, including age <61 and ≥61, with corresponding scores of 0 and 8, respectively; The third row is the surgical method, including sublobar resection and lobectomy, with corresponding scores of 0 and 0.5, respectively; The fourth row is the ratio of RFLV to TFLV, ranging from 5 to 17.5, corresponding to a score of 0 to 7, where 12.5 corresponds to 5 points; The fifth line is FEV1 QCT The value range is 5 to 0.5, and the corresponding score is 0 to 100; The sixth row is the total score, which is the sum of the scores from the second to the fifth row, ranging from 0 to 120; The seventh row is the probability of pulmonary complications, with a total score range of 72 to 104.5, corresponding to a probability of 0.1 to 0.

8.

7. An electronic device, characterized in that: The invention comprises at least one processor and a memory in communication with the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor can run the data acquisition module and the calculation module described in any one of claims 1 to 3.

8. Use of the prediction model according to any one of claims 4 to 5 in predicting the probability of complications after lobectomy or sublobar resection for non-diagnostic purposes.

Images