A tumor prediction system based on 18F-FDG PET / CT metabolic parameters

Abstract

The present application discloses a tumor prediction system based on 18F-FDG PET / CT metabolic parameters, which specifically includes: obtaining patient information and screening out lung adenocarcinoma patients who meet preset standards; performing image acquisition on lung adenocarcinoma patients to obtain PET / CT images; manually outlining the region of interest of the PET / CT image, and obtaining various metabolic parameter values ​​of the lesion based on the region of interest; manually determining the expression results of positively stained reaction cells in the PET / CT image; analyzing the correlation between metabolic parameters and the expression results using the Spearman rank correlation coefficient; and establishing a lung adenocarcinoma tumor prediction model based on the correlation analysis results using a multifactorial Logistic regression method. It can be seen that the present invention establishes 18F-FDG PET / CT metabolic parameters to predict the expression of a new marker NPM1 in lung adenocarcinoma, and improves the preoperative prediction of high-risk lung adenocarcinoma by performing non-invasive diagnosis of lung adenocarcinoma.

Description

Technical Field

[0001] The present invention relates to the technical field of tumor diagnosis, and in particular to a tumor prediction system based on 18F-FDG PET / CT metabolic parameters. Background Art

[0002] Lung cancer is the malignant tumor with the highest morbidity and mortality, among which lung adenocarcinoma (LUAD) is the most common type. Although lung cancer has made considerable progress in disease diagnosis and tumor treatment, most patients are already in the middle and late stages at the time of diagnosis, with metastatic disease, and thus lack effective treatment, resulting in a poor prognosis for patients with lung adenocarcinoma, with an overall survival rate of less than 18% within 5 years. Early diagnosis and effective treatment of LUAD are of great significance to the prognosis of patients. Imaging methods play an important role in the early screening and late-stage disease evaluation of LUAD. First, chest CT is used for examination, and gene typing is achieved through sequencing technology. Among them, thyroid transcription factor 1 (TTF1) is a relatively mature pathological diagnostic marker for lung adenocarcinoma, and its positive expression rate in lung adenocarcinoma tumor tissue is 85%. However, the biological markers for the pathological diagnosis of lung adenocarcinoma must be based on the acquisition of tissue specimens, which takes a long time to detect and causes certain damage to the patient's body, thus limiting the guiding value of preoperative treatment. In addition, the treatment for lung cancer is mainly surgical resection of the lesion combined with radiotherapy and chemotherapy, which has strong side effects and is prone to tumor recurrence. Therefore, it is extremely important to find new markers and a non-invasive, visual method for diagnosing and treating lung cancer. Summary of the invention

[0003] Based on this, the embodiment of the present application provides a tumor prediction system based on 18F-FDG PET / CT metabolic parameters, establishes a model for using FDG PET / CT metabolic parameters to predict the expression levels of tumor diagnostic markers in patients, and solves the key technical problem of non-invasive quantitative monitoring of tumor cell protein expression by molecular imaging for tumor pathology diagnosis.

[0004] In the first aspect, an application of 18F-FDG PET / CT metabolic parameters as a pathological diagnostic marker for lung adenocarcinoma in constructing a corresponding tumor prediction system is provided, and the application includes:

[0005] 18F-FDG is used as a contrast agent in PET / CT examinations.

[0006] In a second aspect, a tumor prediction system based on 18F-FDG PET / CT metabolic parameters is provided, the system comprising:

[0007] An acquisition module is used to obtain patient information and screen out lung adenocarcinoma patients who meet the preset criteria;

[0008] An image acquisition module is used to acquire images of lung adenocarcinoma patients to obtain PET / CT images; wherein 18F-FDG is used as a developer during PET / CT examination, and the image acquisition includes performing a low-dose spiral CT scan on the lung adenocarcinoma patients, then a PET emission scan, and finally a chest thin-layer CT scan;

[0009] A delineation module is used to manually analyze the size, shape, edge, density, number, etc. of the PET / CT image, delineate a region of interest on the lesion, and automatically measure the region of interest through a preset software system to obtain various metabolic parameter values ​​of the lesion; wherein the region of interest is used to characterize the concentrated lesion in the 18F-FDG PET / CT image, and the metabolic parameters of the lesion include at least SUVmax, SUVmean, TLG, and MTV;

[0010] A determination module, used to manually determine the expression results of the positively stained reaction cells in the PET / CT image; wherein the expression results are used to characterize the expression levels of NPM1 and TTF1 proteins in the tumor and adjacent tissues;

[0011] An analysis module, used for analyzing the correlation between metabolic parameters and the expression results using the Spearman rank correlation coefficient;

[0012] A module was established to establish a prediction model for lung adenocarcinoma tumors using a multifactorial logistic regression method based on the results of correlation analysis.

[0013] Optionally, in the module, a prediction model for lung adenocarcinoma is established by using a multi-factor logistic regression method based on the correlation analysis results, including:

[0014] A multivariate logistic regression method was used to establish a model for predicting NPM1 expression in lung adenocarcinoma, and six indicators, including patient age, gender, SUVmax, SUVmean, TLG, and MTV, were set as independent factors for predicting NPM1 expression;

[0015] The receiver operating characteristic curve of the prediction model was drawn, and the area under the curve and its 95% CI were obtained. The z statistic was then used to compare the area under the curve between the modeling group and the validation group.

[0016] Youden index was used to find the optimal cutoff point of the receiver operating characteristic curve and to determine the efficacy and optimal cutoff point of PET / CT metabolic parameters in evaluating NPM1 expression.

[0017] Optionally, drawing a receiver operating characteristic curve of the prediction model, obtaining the area under the curve and its 95% CI, and then using the z statistic to compare the area under the curve of the modeling group and the validation group includes:

[0018] All statistical tests were two-sided, and P < 0.05 was considered statistically significant.

[0019] Optionally, the preset standards in the acquisition module include:

[0020] The pathological diagnosis after surgery was confirmed to be lung adenocarcinoma;

[0021] No radiotherapy or chemotherapy was performed before 18F-FDG PET / CT examination;

[0022] Surgery within 2 weeks after 18F-FDG PET / CT examination;

[0023] The tissue sample wax blocks meet the requirements for immunohistochemical staining;

[0024] With complete clinical data.

[0025] Optionally, the image acquisition module acquires images of lung adenocarcinoma patients to obtain PET / CT images, including:

[0026] The CT scanning conditions were performed according to the following parameters: tube voltage of 120 kV, tube current of 110 mA, rotation speed of 0.7 s / cycle, bed speed of 29.46 mm / s, and matrix of 512×512.

[0027] Optionally, the image acquisition module acquires images of lung adenocarcinoma patients to obtain PET / CT images, and further includes:

[0028] The PET images were corrected for scattering and attenuation. After the PET images were corrected for scattering and attenuation, they were transmitted to the PACS workstation together with the CT images for simultaneous fusion to obtain PET / CT images in the axial, coronal and sagittal planes.

[0029] Optionally, the determining module manually determines the expression results of the positive staining reaction cells in the PET / CT image, including:

[0030] The number of positively stained cells in each PET / CT image was manually determined, and the staining intensity was divided into multiple levels according to the staining situation.

[0031] Optionally, the analysis module uses the Spearman rank correlation coefficient to analyze the correlation between the metabolic parameters and the expression results, including:

[0032] SPSS17.0 and Graphpad Prism 7.0 software were used for statistical analysis, and t-test was used to analyze normally distributed continuous variables. 2 For bivariate analysis tests.

[0033] The technical solution provided in the embodiment of the present application specifically includes: using 18F-FDG as a developer during PET / CT examination, specifically, by obtaining patient information, screening out lung adenocarcinoma patients who meet the preset standards; performing image acquisition on lung adenocarcinoma patients to obtain PET / CT images; manually outlining the region of interest of the PET / CT image, and obtaining various metabolic parameter values ​​of the lesion based on the region of interest; manually determining the expression results of positively stained reaction cells in the PET / CT image; using the Spearman rank correlation coefficient to analyze the correlation between metabolic parameters and the expression results; and using a multifactor logistic regression method based on the correlation analysis results to establish a prediction model for lung adenocarcinoma tumors. It can be seen that the present invention establishes a new method for predicting the expression of a new marker NPM1 in lung adenocarcinoma by using FDG PET / CT metabolic parameters, and performing non-invasive diagnosis of lung adenocarcinoma, thereby improving the preoperative prediction of high-risk lung adenocarcinoma. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0035] Figure 1 A block diagram of a tumor prediction system based on 18F-FDG PET / CT metabolic parameters provided in an embodiment of the present application;

[0036] Figure 2 PET / CT images of different SUVmax of patients with lung adenocarcinoma in the embodiments of this application;

[0037] Figure 3 This is a schematic diagram of the expression of NPM1 and TTF1 in lung cancer tissue and adjacent tissue in the examples of this application;

[0038] Figure 4 This is a schematic diagram of the scoring of NPM1 and TTF1 in lung cancer tissue and adjacent tissue in the examples of this application;

[0039] Figure 5 This is a schematic diagram of the correlation analysis between PET / CT metabolic parameters and NPM1 expression in the examples of this application;

[0040] Figure 6 Schematic diagram of the correlation analysis between PET / CT metabolic parameters and TTF1 expression in the examples of this application. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0042] In the description of the present invention, the terms "comprises", "has" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may also include other steps or units that are not explicitly listed but are inherent to these processes, methods, products or apparatuses, or steps or units added based on further optimization schemes conceived by the present invention.

[0043] 18F-FDG positron emission tomography / computed tomography (18F-FDG PET / CT) is a non-invasive method that can show the glucose uptake in tumors. 18F-FDG refers to fluorodeoxyglucose, which is currently widely used in disease diagnosis, staging, treatment planning and treatment monitoring. The poor prognosis of many human tumors, including lung adenocarcinoma, is associated with increased 18F-FDG uptake. The maximum standardized uptake value (SUVmax), the metabolic tumor volume (MTV) and the total lesion glycolysis (TLG) are metabolic parameters used in 18F-FDG PET / CT imaging to diagnose neoplastic lesions and can reflect the glucose metabolism and load of tumor cells in the body. Studies have found that during the occurrence and development of tumors, there is abnormal expression of related genes, which causes corresponding pathological changes. The imaging manifestations based on this are all produced on the basis of pathological changes. There is a certain connection between the three, which can be studied through the correlation between FDG PET / CT metabolic parameters and tumor biological markers.

[0044] Nucleophosmin 1 (NPM1) is composed of 294 amino acids. As a multifunctional protein, it often shuttles between the nucleus and cytoplasm and participates in many important biological activities of the human body. The latest studies have shown that NPM1 is upregulated in lung cancer and may be involved in the occurrence and development of lung cancer, suggesting that NPM1 expression can be used as a diagnostic marker for LUAD.

[0045] Based on this, this application retrospectively analyzes the expression of multiple metabolic parameters and related biological indicators NPM1 and TTF1 in 43 cases of LUAD by 18F-FDG PET / CT imaging, explores the correlation between the two, provides a new method for LUAD diagnosis, and provides guidance for clinical practice. Figure 1 , which shows a block diagram of a tumor prediction system based on 18F-FDG PET / CT metabolic parameters provided in an embodiment of the present application, the system may include:

[0046] The acquisition module obtains patient information and screens out patients with lung adenocarcinoma who meet the preset criteria.

[0047] In the examples of the present application, 46 patients who underwent 18F-FDG PET / CT examination in a certain hospital and were pathologically diagnosed with lung adenocarcinoma after surgical resection were selected, and paraffin blocks of such patients were obtained for experimental needs. Patient information including general case information and pathological information were collected as required. Inclusion criteria: ① Lung adenocarcinoma was confirmed by pathology after surgery. ② No radiotherapy or chemotherapy was performed before the 18F-FDG PET / CT examination. ③ Surgery was performed within 2 weeks after the 18F-FDG PET / CT examination. ④ The tissue sample wax block met the requirements for immunohistochemistry (IHC) staining. ⑤ Complete clinical data are available. Exclusion criteria: ① Multiple primary lesions. ② Postoperative pathology is of other types, such as squamous cell carcinoma, undifferentiated carcinoma, and small cell carcinoma.

[0048] The image acquisition module acquires images of patients with lung adenocarcinoma to obtain PET / CT images.

[0049] Among them, 18F-FDG is used as the developer in PET / CT examination. Image acquisition includes low-dose spiral CT scan of patients with lung adenocarcinoma, followed by PET emission scan, and finally thin-layer CT scan of the chest.

[0050] In the present embodiment, the patient should fast for more than 6-8 hours before image acquisition, and the blood sugar level should be controlled at 7.1mmol·L -1 The injection dose of 18F-FDG (3.7-4.1MBq·kg -1), intravenous injection. After the injection of the contrast agent, the patient needs to rest in a quiet and comfortable waiting room, and the scan will be performed 50 to 60 minutes later. During this period, drink plenty of water, talk less, do not exercise vigorously, and empty the bladder before the examination.

[0051] All patients lie flat on the appropriate position on the examination bed and are examined using routine clinical protocols. The CT scanning conditions are based on the following parameters: tube voltage of 120 kV, tube current of 110 mA, rotation speed of 0.7 s / week, bed speed of 29.46 mm / s, and matrix of 512×512. The scanning range is from the top of the skull to the middle and upper end of the femur. Depending on the patient's condition, it is necessary to scan the limbs. First, a low-dose spiral CT scan is performed, followed by a PET emission scan, and finally a thin-layer CT scan of the chest. For PET image acquisition, 7 to 9 beds are generally collected, one bed every 1.2 minutes, with a layer thickness of 5 mm. After scattering and attenuation correction, the PET image is transmitted to the PACS workstation together with the CT image for simultaneous fusion to obtain PET / CT images in the axial, coronal and sagittal planes. Figure 2 , PET / CT images of different SUVmax in patients with lung adenocarcinoma are given.

[0052] The delineation module manually delineates the region of interest of the PET / CT image and obtains the values ​​of various metabolic parameters of the lesion based on the region of interest.

[0053] The region of interest is used to characterize the concentrated lesions in the 18F-FDG PET / CT images, and the metabolic parameters of the lesions include at least SUVmax, SUVmean, TLG, and MTV.

[0054] In the embodiment of the present application, two experienced nuclear medicine physicians independently read the film, for example, two physicians with rich experience above the deputy senior level are selected in the PET-CT center, and the image size, morphology, edge, density, number, etc. are analyzed, and the region of interest (ROI) is delineated for the concentrated lesions in the 18F-FDG PET / CT image to calculate SUVmax, SUVmean, TLG, and MTV; for patients in whom no obvious abnormal radioactive concentration is observed in PET imaging, the region of interest with the same size as the lesion is delineated at the corresponding position in the image according to the position of the lesion shown on the CT. The Meizi software uses a specific computer program automatic algorithm based on the delineated ROI to obtain the SUVmax, SUVmean, and MTV of the lesion, and then calculates the TLG value (TLG=ΣMTV×SUVmean). In the embodiment of the present application, the specific computer program can be any existing program algorithm that can realize the acquisition of corresponding parameters through the region of interest.

[0055] The determination module manually determines the expression results of positively stained cells in the PET / CT images, wherein the expression results are used to characterize the expression levels of NPM1 and TTF1 proteins in tumors and adjacent tissues.

[0056] like Figure 3 and Figure 4 The expression and scores of NPM1 and TTF1 in lung cancer tissue and adjacent tissue are given respectively. As can be seen from the figure, compared with the low expression group of NPM1, the high expression group has dense and deep staining, obvious cell atypia, and a large nuclear-cytoplasmic ratio, while no obvious tumor staining is seen in the adjacent adjacent tissue. Compared with the low expression group of TTF1, the high expression group has the same dense and deep staining, and a few tumor cells can still be seen in the adjacent tissue. The NPM1 score in tumor tissue (mean±SD: 1.96±0.84) is significantly higher than that in adjacent normal lung tissue (mean±SD: 0.13±0.34); similarly, the immunohistochemical staining score of TTF1 (mean±SD: 1.65±0.90) is significantly higher than that in adjacent normal lung tissue (mean±SD: 0.83±0.38). In the embodiment of the present application, the outcome of the examinee was followed up after the PET / CT examination, and the postoperative pathological results or dynamic changes of the lesions of the patient were obtained by consulting the electronic medical records of the hospital, CT follow-up images, and telephone inquiries. The postoperative wax blocks of the patient were collected, and the expression levels of NPM1 and TTF1 proteins in the tumor and adjacent tissues were detected after a series of immunohistochemical steps such as sectioning, antigen repair, quenching, blocking, antibody incubation, DAB color development, and hematoxylin counterstaining. The specific scoring criteria are as follows: under light microscopy, the yellow-brown particles specifically located in the nucleus, cell membrane or cytoplasm are positively stained cells. Each section was carefully observed under the microscope by two senior pathologists to estimate the situation of positive staining reaction cells in each section. The staining intensity is divided into 4 levels according to the staining situation: the number of positive cells in the distribution range is strongly stained>50% for 3+, 25%-50% for 2+, 10%-25% for 1+, <10% for-, 2+ and 3+ are recorded as high expression.

[0057] Table 1 shows the positive rates of NPM1 and TTF1 in lung adenocarcinoma tumor tissues and adjacent tissues. In tumor cells, both NPM1 and TTF1 are expressed in the nucleus. The positive rates of NPM1 and TTF1 in tumor tissues were 95.7% (44 / 46) and 89.1% (41 / 46), respectively. However, the positive rates of NPM1 and TTF1 in adjacent normal lung epithelial tissues were 13.0% (6 / 46) and 82.6% (38 / 46). NPM1 and TTF1 were highly expressed in lung adenocarcinoma tissues, and NPM1 was specifically expressed in tumor tissues, while adjacent tissues were rarely expressed or only weakly positive.

[0058] Table 1

[0059]

[0060] Analysis module, using Spearman rank correlation coefficient to analyze the correlation between metabolic parameters and expression results.

[0061] In the examples of the present application, SPSS 17.0 and Graphpad Prism 7.0 software were used for statistical analysis. The t-test was used to analyze normally distributed continuous variables. 2 For bivariate analysis, the Spearman rank correlation coefficient was used to analyze the relationship between PET / CT metabolic parameters and NPM1 and TTF1 expression. Figure 5 , a schematic diagram of the correlation analysis between PET / CT metabolic parameters and NPM1 expression in the embodiment of this application is given. The results show that SUVmax, SUVmean, TLG are positively correlated with the IHC grade of NPM1, with rho of 0.695, 0.679, and 0.478, respectively, and P values ​​are all less than 0.001; however, the correlation between the IHC expression grade of NPM1 and MTV is not statistically significant. Figure 6 A schematic diagram of the correlation analysis between PET / CT metabolic parameters and TTF1 expression in the examples of this application is given. The results show that MTV is negatively correlated with the IHC expression level of TTF1 (rho=-0.369, P=0.009). However, the IHC expression level of TTF1 has no correlation with SUVmax, SUVmean, and TLG.

[0062] A module was established, and a prediction model for lung adenocarcinoma tumors was established using the multivariate logistic regression method based on the correlation analysis results.

[0063] In the present application embodiment, a multi-factor Logistic regression method is used to establish a model for predicting the expression of NPM1 in lung adenocarcinoma, and the patient's age, gender, SUVmax, SUVmean, TLG, and MTV are all independent factors for predicting the expression of NPM1. The receiver operating characteristic (ROC) curve of the prediction model is drawn, and the area under the curve (AUC) and its 95% CI (confidence interval) are obtained. The z statistic is then used to compare the AUC of the modeling group and the validation group. All statistical tests are bilateral tests, and P<0.05 is considered to be statistically significant. The Youden index is used to find the optimal cutoff point of the ROC curve, and the metabolic parameters of PET / CT are determined to evaluate the efficacy and optimal cutoff point of NPM1 expression. Table 2 shows the results of multi-factor analysis of the high and low expression of NPM1 in patients with lung adenocarcinoma. The results show that SUVmax, one of the most commonly used metabolic parameters in clinical practice, is a variable that can affect and reflect NPM1. It is speculated that SUVmax can be used as the main predictor of NPM1 expression in lung adenocarcinoma.

[0064] Table 2

[0065]

[0066] In summary, it can be seen that the present invention obtains the patient's general clinical data, postoperative pathology, tumor marker expression score and PET image parameters, and uses histological diagnosis technology to compare the value of the new molecular phosphoprotein 1 (nucleophosmin1, NPM1) and TTF1 in the diagnosis of lung adenocarcinoma. NPM1 has higher sensitivity than TTF1 (95.7% vs 89.1% positive rate in tumor tissue), and higher specificity than TTF1. It can be used as a new marker for specific diagnosis of lung adenocarcinoma and a new target for treatment. The quantitative parameter SUVmax reflecting FDG uptake in tumor tissue is correlated with NPM1 expression (p<0.05), while the correlation with TTF1 expression is not statistically significant. The multivariate Logistic regression method is used to establish a prediction model, and it is found that SUVmax is the main predictor of NPM1 expression in lung adenocarcinoma. The ROC curve evaluation model is used to determine the SUVmax critical value that can best predict the high or low expression of NPM1 in lung adenocarcinoma. The ROC curve analysis shows that the area under the curve is 0.839, and the 95% CI is 0.672 to 1. The optimal cutoff value of SUVmax was 3.93, with a sensitivity of 90.9% and a specificity of 84.6% for predicting NPM1 expression. FDG PET / CT metabolic parameters were established to predict the expression of a novel marker NPM1 in lung adenocarcinoma, for non-invasive diagnosis of lung adenocarcinoma, and to improve preoperative prediction of high-risk lung adenocarcinoma.

[0067] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0068] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the patent application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be subject to the attached claims.

Claims

1. A tumor prediction system based on 18F-FDG PET / CT metabolic parameters, characterized in that: The system comprises: An acquisition module is used to obtain patient information and screen out lung adenocarcinoma patients who meet the preset criteria; An image acquisition module is used to acquire images of lung adenocarcinoma patients to obtain PET / CT images; wherein 18F-FDG is used as a developer during PET / CT examination, and the image acquisition includes performing a low-dose spiral CT scan on the lung adenocarcinoma patients, then a PET emission scan, and finally a chest thin-layer CT scan; The delineation module manually analyzes the size, shape, edge, density, and number of the PET / CT image, delineates a region of interest on the lesion, and automatically measures the region of interest through a preset software system to obtain various metabolic parameter values ​​of the lesion; wherein the region of interest is used to characterize the concentrated lesion in the 18F-FDG PET / CT image, and the metabolic parameters of the lesion include at least SUVmax, SUVmean, TLG, and MTV; A determination module, used to manually determine the expression results of the positively stained reaction cells in the PET / CT image; wherein the expression results are used to characterize the expression levels of NPM1 and TTF1 proteins in the tumor and adjacent tissues; An analysis module, used for analyzing the correlation between metabolic parameters and the expression results using the Spearman rank correlation coefficient; Establish a module for establishing a prediction model for lung adenocarcinoma tumors using a multi-factor logistic regression method based on the results of correlation analysis; In the module, a prediction model for lung adenocarcinoma was established using multivariate logistic regression method based on the correlation analysis results, including: A multivariate logistic regression method was used to establish a model for predicting NPM1 expression in lung adenocarcinoma, and six indicators, including patient age, gender, SUVmax, SUVmean, TLG, and MTV, were set as independent factors for predicting NPM1 expression; The receiver operating characteristic curve of the prediction model was drawn, and the area under the curve and its 95% CI were obtained. The z statistic was then used to compare the area under the curve between the modeling group and the validation group. Youden index was used to find the optimal cutoff point of the receiver operating characteristic curve and to determine the efficacy and optimal cutoff point of PET / CT metabolic parameters in evaluating NPM1 expression.

2. The system according to claim 1, characterized in that The method of drawing the receiver operating characteristic curve of the prediction model, obtaining the area under the curve and its 95% CI, and then using the z statistic to compare the area under the curve of the modeling group and the verification group includes: All statistical tests were two-sided, and P < 0.05 was considered statistically significant.

3. The system according to claim 1, characterized in that The preset standards in the acquisition module include: The pathological diagnosis after surgery was confirmed to be lung adenocarcinoma; No radiotherapy or chemotherapy was performed before 18F-FDG PET / CT examination; Surgery within 2 weeks after 18F-FDG PET / CT examination; The tissue sample wax blocks meet the requirements for immunohistochemical staining; With complete clinical data.

4. The system according to claim 1, characterized in that The image acquisition module acquires images of lung adenocarcinoma patients to obtain PET / CT images including: The CT scanning conditions were performed according to the following parameters: tube voltage of 120 kV, tube current of 110 mA, rotation speed of 0.7 s / cycle, bed speed of 29.46 mm / s, and matrix of 512×512.

5. The system according to claim 1, characterized in that The image acquisition module acquires images of lung adenocarcinoma patients to obtain PET / CT images, and also includes: The PET images were corrected for scattering and attenuation. After the PET images were corrected for scattering and attenuation, they were transmitted to the PACS workstation together with the CT images for simultaneous fusion to obtain PET / CT images in the axial, coronal and sagittal planes.

6. The system according to claim 1, characterized in that The determination module manually determines the expression results of the positive staining reaction cells in the PET / CT image, including: The number of positively stained cells in each PET / CT image was manually determined, and the staining intensity was divided into multiple levels according to the staining situation.

7. The system according to claim 1, characterized in that SPSS 17.0 software was used for statistical analysis in the analysis module, including: t test for analyzing normally distributed continuous variables, χ 2 For bivariate analysis tests, the Spearman rank correlation coefficient was used to analyze the correlation between metabolic parameters and the expression results.

Images