A method, apparatus and program product for discriminating between systemic lupus erythematosus with lymphadenopathy and lymphoma

Abstract

The application relates to the field of intelligent medical treatment, in particular to a method, equipment and program product for identifying systemic lupus erythematosus with lymph node enlargement and lymphoma. The method comprises the following steps: acquiring a PET / CT image of a patient; performing ROI region division on the image to obtain an ROI region; performing metabolic characteristic calculation on the ROI region to obtain a metabolic characteristic, wherein the metabolic characteristic is the ratio of SUVmax-blo and SUVmax-liv; and identifying systemic lupus erythematosus with lymph node enlargement and lymphoma based on the metabolic characteristic to obtain an identification result. The application can accurately identify systemic lupus erythematosus with lymph node enlargement and lymphoma, and has good clinical value.

Description

Technical Field

[0001] This application relates to the field of intelligent healthcare, specifically to a method, device, program product, and computer-readable storage medium for differentiating between systemic lupus erythematosus with lymphadenopathy and lymphoma. Background Technology

[0002] Systemic lupus erythematosus (SLE) is an autoimmune disease characterized by the presence of multiple autoantibodies in the serum and multi-system involvement. Its pathogenesis is not fully understood, and its clinical manifestations are highly heterogeneous. Patients often experience significant inflammatory responses and tissue damage, even life-threatening ones, due to the production of large amounts of autoantibodies and inflammatory mediators. More than 30% of SLE patients develop lymphadenopathy (LAD) during the course of the disease, typically affecting the neck and axilla, followed by the abdomen, pelvis, and mediastinum. Except for malignant tumors, histopathological examination often reveals reactive hyperplasia and histiocytic necrotizing lymphadenitis (HNL). LAD is a non-specific manifestation of SLE, and in clinical practice, it is necessary to be alert to the possibility of LAD arising from co-existing tumors or infections. It is noteworthy that the incidence of malignant tumors in SLE patients (3.2%-13.8%) is significantly higher than in the general population, with the risk of developing lymphoma reaching 4-7 times that of healthy individuals. Therefore, for SLE patients with LAD, a comprehensive assessment is necessary, including detailed medical history taking, thorough physical examination, and imaging, pathological, and etiological testing, to identify secondary causes such as infection and tumors. Currently, the fusion of PET and CT combines functional metabolic information with anatomical structure localization, and its clinical application is increasingly widespread. A single whole-body scan can simultaneously obtain high-resolution anatomical images from CT and metabolic activity images from PET, providing a unique advantage for comprehensive disease assessment. However, current approaches for SLE-LAD patients... 18 Studies on F-FDG PET / CT imaging features are relatively scarce, mainly consisting of case reports and lacking systematic cohort studies. In particular, the differences in metabolic patterns among different pathological subtypes have not been clarified, and the quantitative relationship between key metabolic parameters and disease activity needs to be elucidated. Summary of the Invention

[0003] To address the above problems, this invention provides a method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma, specifically comprising: Acquire patient PET / CT images; The image is divided into Regions of Interest (ROIs) to obtain ROI regions; Metabolic features are calculated for the ROI region, and the metabolic features are the ratio of SUVmax-blo to SUVmax-liv. Based on the aforementioned metabolic characteristics, the differential diagnosis between systemic lupus erythematosus with lymphadenopathy and lymphoma was obtained.

[0004] Optionally, the metabolic characteristics are replaced by the hottest lesion TLR, and the results of the identification of systemic lupus erythematosus with lymphadenopathy or lymphoma are obtained by identifying the hottest lesion TLR.

[0005] Optionally, the metabolic characteristics also include the hottest lesion TLR, and the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma is obtained by the ratio of the hottest lesion TLR, SUVmax-blo, and SUVmax-liv.

[0006] Optionally, the calculation process for the hottest lesion TLR is as follows: obtain the metabolic parameters of the liver, and calculate the hottest lesion TLR by combining the metabolic parameters of the liver with the metabolic parameters of the ROI region.

[0007] Optionally, the metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax, and the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma is obtained by using the ratio of SUVmax-blo and / or hottest lesion SUVmax, SUVmax-blo and SUVmax-liv.

[0008] Optionally, the metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax, and the identification results of systemic lupus erythematosus with lymphadenopathy or lymphoma are obtained by differentiating SUVmax-blo and / or hottest lesion SUVmax and hottest lesion TLR.

[0009] Optionally, the metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax, and the identification results of systemic lupus erythematosus with lymphadenopathy or lymphoma are obtained by using SUVmax-blo and / or hottest lesion SUVmax, the ratio of SUVmax-blo and SUVmax-liv, and the hottest lesion TLR.

[0010] Optionally, the method further includes immune characteristics, obtaining immune data from the patient to obtain immune characteristics, and differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma based on the metabolic and immune characteristics to obtain a differentiation result; the immune characteristics include IgA, C4 and / or C3.

[0011] Optionally, the method further includes clinical features, obtaining clinical data of the patient to obtain clinical features, and differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma based on the metabolic features and clinical features to obtain the differentiation result; the clinical features include any one or more of the following: SLEDAI-2K score and platelet count, absolute lymphocyte count, and maximum standardized uptake value of spleen; Optionally, the differentiation between systemic lupus erythematosus with lymphadenopathy and lymphoma can be obtained based on the aforementioned metabolic, immune, and clinical characteristics.

[0012] Optionally, the identification is performed by an identification model, and the metabolic characteristics are input into the identification model to obtain the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma. Optionally, the discrimination model may employ any one or more of the following: convolutional neural network, residual network, XGBoost, SVM, random forest, and logistic regression; Optionally, the training process of the discrimination model is as follows: acquire a PET / CT image dataset, divide the PET / CT image dataset into ROI regions to obtain ROI regions, extract metabolic features from the ROI regions and calculate the metabolic features, input the metabolic features into the discrimination model to be trained, until the loss function remains unchanged, and obtain the discrimination model; Optionally, the metabolic and immune features are input into a second differential diagnosis model to obtain the differential diagnosis result of systemic lupus erythematosus with lymphadenopathy or lymphoma; the second differential diagnosis model is obtained by training on the metabolic and immune features. Optionally, the metabolic and clinical features are input into a third differential diagnosis model to obtain the differential diagnosis results for systemic lupus erythematosus with lymphadenopathy or lymphoma; the third differential diagnosis model is obtained through training on the metabolic and clinical features. Optionally, the metabolic, immune, and clinical features are input into a fourth differential diagnosis model to obtain the differential diagnosis results for systemic lupus erythematosus with lymphadenopathy or lymphoma; the fourth differential diagnosis model is obtained by training on the metabolic, immune, and clinical features.

[0013] Optionally, the ROI region is the mediastinal blood pool, and metabolic characteristics are calculated from the mediastinal blood pool. Optionally, metabolic characteristics can be calculated for the luminal region of the mediastinal blood pool ascending aorta or aortic arch to obtain metabolic features. The purpose of this invention is to provide a computer program product comprising a computer program or instructions, wherein the computer program or instructions are executed by a processor to implement the above-described method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma.

[0014] The purpose of this invention is to provide a computer device comprising a memory, a processor, and a computer program or instructions stored in the memory, wherein the computer program or instructions are executed by the processor to implement the above-described method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma.

[0015] The purpose of this invention is to provide a computer-readable storage medium having a computer program or instructions stored thereon, which are executed by a processor to implement the above-described method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma.

[0016] Advantages of this invention: 1. SLE patients often experience lymphadenopathy due to immune dysregulation, and SLE itself increases the risk of lymphoma (the incidence rate is 4-7 times higher than in the general population). Traditional imaging techniques (such as CT / MRI) are difficult to distinguish between inflammatory lymph nodes in SLE and lymphoma, leading to over-biopsy or missed diagnosis. Therefore, based on research, this invention proposes a metabolic indicator that can identify SLE-LD (systemic lupus erythematosus with lymphadenopathy) and lymphoma: the ratio of SUVmax-blo to SUVmax-liv. This indicator can effectively differentiate between these two diseases, providing a clinically significant technical solution.

[0017] 2. To differentiate between SLE-LD and lymphoma, this invention also proposes other indicators to be used in conjunction with the ratios of SUVmax-blo and SUVmax-liv for differentiation, including any one or more of the following: hottest lesion TLR, SUVmax-blo, hottest lesion SUVmax, IgA, C4, C3, SLEDAI-2K score and platelet count, absolute lymphocyte count, and maximum standardized uptake value of the spleen; providing better accuracy and reliability for disease differentiation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic flowchart of a method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of population screening provided in an embodiment of the present invention; Figure 3 Provided for embodiments of the present invention 18 F-FDG PET / CT findings and distribution range; Figure 4 This invention provides a comparison and distribution of the hottest lesion parameters in lymph nodes of SLE-LAD patients using 18F-FDG PET / CT. A represents a comparison of the SUVmax and location of the most metabolically active lesions in SLE-LAD patients. B represents a schematic diagram of the location of metabolically active lesions in SLE-LAD patients. Figure 5 A heatmap showing the correlation between general information, clinical data, and 18F-FDG PET / CT metabolic indicators of SLE-LAD patients provided in this embodiment of the invention. Figure 6 Correlation analysis of clinical data and imaging indicators of SLE-LAD patients provided in embodiments of the present invention; Figure 7 Correlation analysis of laboratory and imaging parameters in SLE-LAD patients provided in this embodiment of the invention: (AH) analysis of the correlation between SLEDAI-2K and imaging parameters; (IO) analysis of the correlation between C3 and imaging parameters.

[0020] Figure 8 Correlation analysis of laboratory and imaging parameters in SLE-LAD patients provided in this embodiment of the invention includes: (A-F) Correlation analysis of blood cell count with imaging parameters; (GL) Correlation analysis of lymphocyte percentage with imaging parameters.

[0021] Figure 9 Correlation analysis of laboratory and imaging parameters in SLE-LAD patients provided in this embodiment of the invention: (A) Correlation analysis of C3 with imaging parameters. (BI) Correlation analysis of C4 with imaging parameters. (JO) Correlation analysis of IgM with imaging parameters.

[0022] Figure 10 This invention provides a correlation analysis of laboratory and imaging parameters in SLE-LAD patients, including: (AB) Correlation analysis of IgM with imaging parameters; (CJ) Correlation analysis of IgA with imaging parameters; and (KO) Analysis of the correlation between IgG and imaging parameters.

[0023] Figure 11 A comparative analysis of laboratory and imaging characteristics between the SLE group and the lymphoma group provided in this embodiment of the invention; Figure 12 ROC curves for SUVmax-blo / SUVmax-liv, hottest lesion TLR, and combined predictive factors for diagnosing lymphoma provided in embodiments of the present invention; Figure 13 This is a schematic diagram of a system for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma, provided in an embodiment of the present invention. Figure 14 A schematic diagram of a computer device provided for an embodiment of the present invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0025] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as S101, S102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0026] Figure 1 The schematic diagram of the method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma provided in this embodiment of the invention specifically includes: S1: Acquire patient PET / CT images; In one specific embodiment, this study employed a single-center retrospective case-control design, enrolling 15 SLE patients who underwent 18F-FDG PET / CT examinations at the Second Hospital of Hebei Medical University between June 2021 and October 2024 as the study group. All cases met the 2019 European League Against Rheumatism / American College of Rheumatology (EULAR / ACR) SLE classification criteria and had concurrent LAD on PET / CT examination. An additional 30 patients with pathologically confirmed lymphoma (all stage III and IV) admitted during the same period were selected as the control group. The two groups were matched 1:3 in terms of age and sex (e.g., ...). Figure 2 This study was approved by the ethics committee of the Second Hospital of Hebei Medical University (ethics number: 2023-R610) and informed consent was obtained from the patients.

[0027] In one specific embodiment, demographic data of 15 SLE patients, including gender, age, and disease duration, were retrieved from the electronic medical record system. The SLE Disease Activity Index 2000 (SLEDAI-2K) was used for scoring, and clinical manifestations, including general condition, skin and mucous membrane involvement, arthritis, kidney involvement, nervous system involvement, serositis, and hematologic system involvement, were collected. Baseline laboratory results were also collected, including complete blood count, liver and kidney function tests, immunoglobulin (Ig), complement (C), autoantibodies (including antinuclear antibodies, anti-dsDNA antibodies, anti-SSA antibodies, anti-SSB antibodies, and anti-RNP antibodies), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP). Imaging data included echocardiography, lymph node ultrasound, lung CT, and PET / CT scans. Observation and recording were then performed. 18 F-FDG PET / CT shows the extent of lesion involvement, lymph node distribution, related metabolic parameters, and pathological data.

[0028] In one specific embodiment, all patients 18 All F-FDG PET / CT examinations were performed in the Department of Nuclear Medicine of our hospital, using a Philips Vereos PET / CT imaging system from the Netherlands. 18F-FDG was provided by Nanjing Jiangyuan Andico Positron Research and Development Co., Ltd., and the scanning range extended from the top of the skull to the mid-thigh. Following standard procedures, patients fasted for ≥6 hours and had a blood glucose level <8.0 mmol / L before the examination. The injection dose was 0.15 MBq / kg. During the imaging agent uptake period after injection, patients were instructed to drink 500-1000 mL of water and rest quietly for 60 minutes. 3D acquisition mode was used, with an acquisition time of 1.0 min per bed. CT scan parameters: tube voltage 140 kV, tube current automatically set by the system. After computer processing, 18F-FDG tomographic images in the transverse, coronal, and sagittal planes were obtained, including PET, CT, and fused images.

[0029] S2: Divide the image into ROI regions to obtain ROI regions; In one embodiment, the ROI region is the mediastinal blood pool, and metabolic characteristics are calculated from the mediastinal blood pool. Preferably, metabolic characteristics are obtained by calculating the metabolic characteristics of the luminal region of the mediastinal blood pool ascending aorta or aortic arch.

[0030] In one specific embodiment, two nuclear medicine physicians analyzed the examination images, observing abnormal radioactive uptake foci in lymph nodes throughout the body, and recording the location, size, number, CT imaging features, and maximum standardized uptake value (SUVmax) of the lesions. Simultaneously, the SUVmax of the liver blood pool (SUVmax-liv) and the mediastinal blood pool (SUVmax-blo) were recorded for each subject; when spleen or bone marrow involvement was suspected, the SUVmax of the spleen and bone marrow were also recorded. Furthermore, this invention defines the lymph node with the highest SUVmax within the scan range as the "hottest lesion," and the lymph node with the lowest SUVmax in the same patient as the "coldest lesion." The target-to-bloodpool ratio (TBR) and the target-to-liver ratio (TLR) of the lesions (abnormally uptake lymph nodes, spleen, and bone marrow) to the mediastinal blood pool were calculated.

[0031] S3: The metabolic features of the ROI region are calculated to obtain the metabolic features, which are the ratio of SUVmax-blo to SUVmax-liv. In one embodiment, the metabolic characteristic is replaced by the hottest lesion TLR, and the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma is obtained by identifying the hottest lesion TLR.

[0032] In one embodiment, the metabolic characteristics also include the hottest lesion TLR, and the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma is obtained by the ratio of the hottest lesion TLR, SUVmax-blo and SUVmax-liv. Optionally, the calculation process for the hottest lesion TLR is as follows: obtain the metabolic parameters of the liver, and calculate the hottest lesion TLR by combining the metabolic parameters of the liver with the metabolic parameters of the ROI region.

[0033] In one embodiment, the metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax, and the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma is obtained by using the ratio of SUVmax-blo and / or hottest lesion SUVmax, SUVmax-blo and SUVmax-liv.

[0034] In one embodiment, the metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax, and the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma is obtained by using SUVmax-blo and / or hottest lesion SUVmax and hottest lesion TLR.

[0035] In one embodiment, the metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax. The identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma is obtained by using SUVmax-blo and / or hottest lesion SUVmax, the ratio of SUVmax-blo and SUVmax-liv, and the hottest lesion TLR.

[0036] In one embodiment, the method further includes immune characteristics, obtaining immune data from the patient to obtain immune characteristics, and differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma based on the metabolic and immune characteristics to obtain a differentiation result; the immune characteristics include IgA, C4 and / or C3.

[0037] Among them, the metabolic characteristics are the ratio of SUVmax-blo and SUVmax-liv and / or any one or more of the following: SUVmax-blo and / or hottest lesion SUVmax, hottest lesion TLR.

[0038] In one embodiment, the method further includes clinical features, obtaining clinical data of the patient to obtain clinical features, and differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma based on the metabolic features and clinical features to obtain a differentiation result; the clinical features include any one or more of the following: SLEDAI-2K score and platelet count, absolute lymphocyte count, and maximum standardized uptake value of spleen.

[0039] Among them, the metabolic characteristics are the ratio of SUVmax-blo and SUVmax-liv and / or any one or more of the following: SUVmax-blo and / or hottest lesion SUVmax, hottest lesion TLR.

[0040] In one embodiment, the differentiation between systemic lupus erythematosus with lymphadenopathy and lymphoma is obtained based on the metabolic, immune, and clinical characteristics.

[0041] S4: Based on the aforementioned metabolic characteristics, the differentiation between systemic lupus erythematosus with lymphadenopathy and lymphoma was obtained.

[0042] In one embodiment, the identification is performed by an identification model, and the metabolic characteristics are input into the identification model to obtain the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma. Optionally, the discrimination model may employ any one or more of the following: convolutional neural network, residual network, XGBoost, SVM, random forest, and logistic regression.

[0043] In one embodiment, the training process of the discrimination model is as follows: acquire a PET / CT image dataset, divide the PET / CT image dataset into ROI regions to obtain ROI regions, extract metabolic features from the ROI regions and calculate the metabolic features, input the metabolic features into the discrimination model to be trained, until the loss function remains unchanged, and obtain the discrimination model.

[0044] In one embodiment, the metabolic and immune features are input into a second differential diagnosis model to obtain a differential diagnosis result for systemic lupus erythematosus with lymphadenopathy or lymphoma; the second differential diagnosis model is obtained by training on the metabolic and immune features.

[0045] In one embodiment, the metabolic and clinical features are input into a third differential diagnosis model to obtain a differential diagnosis result for systemic lupus erythematosus with lymphadenopathy or lymphoma; the third differential diagnosis model is obtained by training on the metabolic and clinical features.

[0046] In one embodiment, the metabolic features, immune features, and clinical features are input into a fourth differential diagnosis model to obtain a differential diagnosis result for systemic lupus erythematosus with lymphadenopathy or lymphoma; the fourth differential diagnosis model is obtained by training on the metabolic features, immune features, and clinical features.

[0047] In one specific embodiment, IBM SPSS 26.0 software and GraphPad Prism 10.1.2 were used for data analysis. For continuous data, normally distributed data were expressed as mean ± standard deviation (Mean ± SD), and t-tests were used for comparisons between two groups, while parametric tests were used for comparisons between three or more groups. Non-normally distributed data were expressed as median (M(Q1, Q4)), and non-parametric tests (Mann-Whitney U test, Kruskal-Wallis H test, etc.) were used for comparisons between groups. Count data were described as frequency (percentage) (n (%)). Correlation analysis was performed using Pearson's test and Spearman's test. Univariate logistic regression analysis was used to screen indicators for potential differentiation from lymphoma. A multivariate model was established using stepwise regression. Adjusted odds ratio (OR) and 95% confidence interval (CI) were calculated. Receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was calculated to evaluate the diagnostic efficacy of the combined model. All tests were two-sided, and p < 0.05 was considered statistically significant.

[0048] In one specific embodiment, after screening out cases of EB virus infection (n=1), Castleman disease (n=1), and malignant tumors (n=3, 1 case of lymphoma, 1 case of lung cancer, and 1 case of thyroid cancer), 10 SLE-LAD patients were finally included for systematic analysis (baseline characteristics are detailed in Table 1).

[0049] The patients' mean age was 42.20 ± 20.10 years, with females predominating (80%). The mean body mass index (BMI) was 21.84 ± 4.23, and the median disease duration was 4.79 ± 5.49 years. 40% were treatment-naïve patients. All previously treated patients had used glucocorticoids, and 66.67% had used hydroxychloroquine sulfate and / or immunosuppressants. 40% of patients had hypertension, and none had a history of diabetes or familial autoimmune disease. The initial departmental distribution showed that 60% of patients first consulted rheumatology, while the remaining patients consulted hematology, infectious diseases, respiratory medicine, or pediatrics.

[0050] The most common clinical manifestation was hematologic involvement (100%), followed by skin and mucous membrane lesions (80%). The hematologic involvement included leukopenia (50%), hemolytic anemia (20%), and thrombocytopenia (30%). Other clinical manifestations included fever (60%) and renal involvement (50%), with only one patient experiencing neurological involvement. Disease activity assessment showed a median SLEDAI-2K score of 10.8 in the 10 patients, with 80% exhibiting a SLEDAI-2K score ≥8, indicating moderate disease activity.

[0051] Table 1. Demographic data and clinical manifestations of patients with SLE-LAD

[0052] Note: BMI: Body Mass Index; SLEDAI-2K: Systemic Lupus Erythematosus Disease Activity Index - 2000.

[0053] In one specific embodiment, the laboratory parameters of 10 SLE-LAD patients are shown in Table 2. SLE-LAD patients exhibited hematological abnormalities: a neutrophil-lymphocyte ratio (NLR) of 5.68±5.92, an absolute lymphocyte count of 0.92±0.46×10⁹ / L, a hemoglobin level of 104.90±15.11 g / L, and a median platelet count of 140.50×10⁹ / L. Regarding inflammatory markers, CRP, ESR, and ferritin levels were 24.00±30.15 mg / L, 72.20±50.34 mm / h, and 483.75 ng / ml, respectively. In terms of coagulation, the D-dimer level in SLE-LAD patients was 0.66±0.38 ng / ml. Regarding immunological markers, complement C3 and C4 levels in SLE-LAD patients were 0.78±0.40 g / L and 0.14±0.07 g / L, respectively. IgG, IgA, and IgM levels were 13.95 g / L, 3.59 g / L, and 0.80 g / L, respectively. Among autoantibodies, high-titer antinuclear antibodies (≥1:1000) were observed in 60% of patients, anti-SSA antibodies were positive in 30% of patients, and the positive rate for specific antibodies, anti-Sm antibodies, was 10%. Additionally, 30% of patients were positive for anticardiolipin antibodies, 70% were positive for anti-β2GP1 antibodies, and 40% were positive for lupus anticoagulants.

[0054] In one specific embodiment, among the 10 SLE-LAD patients included in this invention, lymph node involvement was extensive, and the 18F-FDG PET / CT findings and distribution range were as follows. Figure 3 Patient 1's pathological diagnosis was HNL. PET / CT showed multiple lymph nodes of varying sizes in the neck, axilla, and groin, with mildly / significantly increased metabolic activity. The largest lymph node in the right axilla had an SUVmax of 11.8 ( Figure 3 Patient 2's pathological findings showed reactive lymphoid hyperplasia (BD). Figure 3 E, F). PET / CT showed multiple lymph nodes of varying sizes with mildly increased metabolism in the superior and inferior mediastinum (SUVmax 2.3-9.4), accompanied by mildly increased metabolic activity in the spleen and bone marrow. Patient 3 was diagnosed with SLE complicated by diffuse large B-cell lymphoma (E, F). Figure 3The pelvic masses (G, H) presented as irregular, heterogeneous masses (SUVmax 8.5-70.0). Scattered hypermetabolic lymph nodes were observed alongside the mass, and mild hypermetabolic lymph nodes were also found in the neck and mediastinum (see Table 3). Deep lymph node involvement was present in all patients, including the chest (90%) and abdominopelvic region (70%). Involvement of the iliac vessels (80%), hilar lymph nodes (70%), and mediastinal lymph nodes (70%) was most common, while the involvement of the retroperitoneal (30%) and intermuscular chest wall lymph nodes (40%) was relatively low. Superficial lymphadenopathy was seen in 80% of patients, involving the neck (80%), axilla (80%), and groin (80%) regions. Affected cervical lymph nodes could be distributed in various regions of the neck. Focal lymph node distribution patterns were predominant (90%), while confluent lymph node aggregation was rare (10%).

[0055] This invention analyzed the PET / CT metabolic parameters of 10 SLE-LAD patients (as shown in Table 4). The results showed significant heterogeneity in the metabolism of lymph node lesions: the SUVmax of the coldest lesion was 2.37±1.16, while the SUVmax of the hottest lesion was as high as 10.37±2.28, suggesting hypermetabolism in some lesions. Using the liver as a reference, the TLRs of the coldest and hottest lesions were 1.11±0.69 and 4.62±1.37, respectively; using the mediastinal blood pool as a reference, the TBRs of the coldest and hottest lesions were 2.01±1.35 and 8.22±2.76, respectively, indicating that the uptake intensity of hypermetabolic lesions can reach 4-8 times that of background tissue. Furthermore, some patients exhibited abnormal metabolism in the bone marrow and / or spleen, with a bone marrow SUVmax of 7.14±4.00 and a spleen SUVmax of 4.10±1.02, suggesting activation of hematopoietic and immune functions in the bone marrow and spleen in the context of SLE.

[0056] Table 2 Laboratory Indicators of SLE-LAD Patients

[0057] Note: Nlr: Neutrophil to Lymphocyte ratio; ESR: Erythrocyte sedimentation rate; Crp: C-reactive protein; C3: Complement component 3; C4: Complement component 4; IgM: Immunoglobulin M; IgA: Immunoglobulin A; IgG: Immunoglobulin G; IIF: Indirect immunofluorescence assay; ELISA: Enzyme-linked immunosorbent assay.

[0058] Table 3. Lymph node distribution in SLE-LAD patients

[0059] Table 4 Lymph nodes in SLE-LAD patients 18 F-FDG PET / CT related parameters

[0060] Note: SUVmax-liv: SUVmax of liver; SUVmax-blo: SUVmax of blood pool; TLR: target-to-liver ratio, TBR: target-to-blood pool ratio.

[0061] Each SLE-LAD patient had one hottest lesion, with an SUVmax ranging from 7.3 to 14.0, primarily located in the abdominis (40%), followed by the axilla (30%), chest (20%), and neck (10%). Figure 4 B). In the comparison between the hottest lesion SUVmax and its location in SLE-LAD patients ( Figure 4 (A) There was no significant difference in the SUVmax of the hottest lesion at different locations, but the SUVmax of the hottest lesion in the superficial part was higher than that in the deep part (P=0.049).

[0062] In one specific embodiment, among the 15 SLE patients of the present invention, 11 patients underwent lymph node biopsy or surgery (thoracoscopic lobectomy, thyroidectomy) to determine the lesion type. Among them, 2 patients had HNL pathological type, 4 patients had reactive lymph node hyperplasia, 3 patients had tumors, namely thyroid cancer, lung cancer, and lymphoma, 1 patient had Castleman's disease, and 1 patient's pathological biopsy indicated EB virus infection.

[0063] The clinical data of 7 patients with reactive hyperplasia, HNL, and lymphoma are shown in Table 5. The SLE patient with lymphoma was 64 years old, had a SLEDAI-2K score of 1, presented with fever, an erythrocyte sedimentation rate (ESR) of 105 mm / h, a CRP of 79 mg / L, and SUVmax values ​​for the coldest and hottest lesions of 4.1 and 70.0, respectively. The patients with HNL were 17 and 16 years old, with SLEDAI-2K scores of 15 and 8, respectively, both with leukopenia, ESRs of 29 mm / h and 32 mm / h, CRPs of 3.72 mg / L and 6.13 mg / L, respectively, SUVmax values ​​for the coldest and hottest lesions of 1.3 and 1.5, respectively, and SUVmax values ​​for the hottest and hottest lesions of 10.1 and 11.8, respectively. Four patients with reactive hyperplasia were aged between 46 and 71 years. Two patients had a SLEDAI-2K score of 4, while the other two had scores of 11 and 12, respectively. Only one patient had fever. The erythrocyte sedimentation rate (ESR) of the four patients ranged from 67 to 151 mm / h. The CRP levels were 1.85 mg / L, 38.3 mg / L, 44.5 mg / L, and 98.6 mg / L, respectively. The SUVmax of the coldest lesion ranged from 1.7 to 4.9, while the SUVmax of the hottest lesion ranged from 12.6 to 14.4, respectively.

[0064] Table 5. Clinical data of SLE patients with pathological types of reactive hyperplasia, HNL, and lymphoma.

[0065] Note: SLEDAI-2K: Systemic lupus erythematosus disease activity index - 2000; ESR: Erythrocyte sedimentation rate; Crp: C-reactive protein; '+' indicates yes, '-' indicates no.

[0066] A correlation analysis was performed on the general information, clinical data, and 18F-FDG PET / CT metabolic parameters of 10 SLE-LAD patients, such as... Figures 5-10 .

[0067] Regarding disease activity, the SLEDAI-2K score was negatively correlated with patients' platelet count and absolute lymphocyte count (r=-0.902, P<0.001; r=-0.654, P=0.031), such as Figure 6 A and B were negatively correlated with complement C3 and C4 (r=-0.698, P=0.020; r=-0.442, P=0.192), as... Figure 6 C and D. The SLEDAI-2K score was significantly positively correlated with spleen SUVmax (r=0.959, P=0.041), such as Figure 6The E value may be related to the active glucose metabolism in the spleen, which is dominated by lymphocytes, during the active phase of the disease. Furthermore, the maximum SUVmax of the hottest lymph node lesions was positively correlated with the maximum SUVmax of bone marrow (r=0.838, P=0.076), such as... Figure 6 The F value suggests a link between lymph node hypermetabolism and bone marrow hematopoiesis or immune activation.

[0068] Regarding metabolic parameters in PET / CT imaging, SUVmax-liv and SUVmax-blo were positively correlated with the percentage of lymphocytes (r=0.805, P=0.005; r=0.686, P=0.029), such as... Figure 6 The values ​​of G and H suggest that SUVmax-liv and SUVmax-blo may reflect the body's basal glucose metabolism. SUVmax-blo was negatively correlated with the SLEDAI-2K score (r=-0.632, P=0.050), as shown in D of score 7. Furthermore, the TLR of the coldest lesion, the TLR of the hottest lesion, the TBR of the coldest lesion, and the TBR of the hottest lesion were all positively correlated with white blood cell count (r=0.530, P=0.115; r=0.684, P=0.029; r=0.414, P=0.234; r=0.512, P=0.130), as shown in D of score 7. Figure 6 The IL level was positively correlated with the absolute lymphocyte count, but not with the absolute lymphocyte count (P>0.05). IgG levels were positively correlated with the coldest lesion SUVmax, coldest lesion TLR, and coldest lesion TBR (r=0.733, P=0.016; r=0.513, P=0.130; r=0.590, P=0.072). Figure 6 MO suggests that B cell activation leading to increased plasma cell secretion is related to active glucose metabolism in SLE lymph nodes.

[0069] A comparative analysis of laboratory and 18F-FDG PET / CT imaging parameters between SLE and lymphoma patients was performed (Table 6). The complement C4 level in the SLE group was significantly lower than that in the lymphoma group (0.14±0.07 vs. 0.23±0.10 g / L, P=0.041). Figure 11 The median IgA level was elevated (median 3.59 vs. 1.57 g / L, P = 0.012). Figure 11 K). In terms of metabolic imaging parameters, the lymphoma group had a higher maximum lesion SUVmax (18.32±12.19 vs. 10.37±2.28, P=0.002) Figure 11 C), higher SUV max-blo (1.68±0.34 vs. 1.36±0.39, P=0.017) Figure 11(A) and higher SUVmax-blo / SUVmax-liv (0.67±0.10 vs. 0.58±0.10, P=0.014) Figure 11 (F). Furthermore, the TLR of the most active lesions was significantly elevated in the lymphoma group (7.42±5.19 vs. 4.62±1.37, P=0.011). Figure 11 (D). In contrast, there were no significant differences between groups in parameters such as white blood cell count, hemoglobin, platelet level, complement C3, IgM, IgG, and the coldest lesion SUVmax, coldest lesion TLR, and TBR (P>0.05). Ferritin levels in the SLE group showed an increasing trend (P=0.054), suggesting that the degree of systemic inflammation in SLE patients may be higher than that in lymphoma patients.

[0070] Table 6. Differences in laboratory and imaging characteristics between the SLE group and the lymphoma group.

[0071] Note: C3 / C4: complement components 3 / 4; Ig: immunoglobulin; Suvmax-Liv: liver suvmax; SUVmax-blo: blood pool SUVmax; TLR: target / liver ratio; TBR: target / blood pool ratio; T: T-test; Z: Mann-Whitney test. P<0.05, P<0.01.

[0072] In one specific embodiment, univariate and multivariate logistic regressions were performed on the 18F-FDG PET / CT related indicators of the SLE group and the lymphoma group, as shown in Tables 7 and 8. In the univariate analysis, SUVmax-blo (OR=13.30, 95%CI: 1.32~134.31, P=0.028) and SUVmax-blo / SUVmax-liv (OR=69235.83, 95%CI: 4.78~1003073215.05, P=0.023) were significantly associated with lymphoma. The hottest lesion SUVmax showed a trend association (OR=1.11, P=0.081), while SUVmax-liv (OR=2.38, P=0.296) and the coldest lesion SUVmax (OR=1.07, P=0.782) were not significantly associated.

[0073] In multivariate analysis, SUVmax-blo / SUVmax-liv were independent predictors of lymphoma (OR=46254.73, 95%CI: 3.23~662530305.19, P=0.028).

[0074] Based on the logistic regression analysis results, the formula for calculating the joint predictor factor is as follows: Joint predictor factor = -6.748 + 10.742 × SUVmax-blo / SUVmax-liv + 0.205 × hottest lesion TLR. Subsequently, the ROC curve for predicting lymphoma was plotted and the AUC was calculated (as shown in Table 9). Figure 12 (As shown). In the single-parameter prediction model, the AUC of SUVmax-blo / SUVmax-liv was 0.777 (95% CI: 0.585–0.969), with a sensitivity of 70.0% (95% CI: 0.416–0.984), a specificity of 83.3% (95% CI: 0.700–0.967), and an optimal cutoff value of 0.584. However, the AUC for the hottest lesion TLR was 0.620 (95% CI: 0.448–0.792). The combined predictor of SUVmax-blo / SUVmax-liv and the hottest lesion TLR had an AUC of 81.3% (95% CI: 0.657–0.970), a sensitivity of 90.0% (95% CI: 0.714–1.000), a specificity of 66.7% (95% CI: 0.498–0.835), and an optimal cutoff value of 1.499. The most active lesion TLR has higher sensitivity when used alone, but its specificity is slightly insufficient. SUVmax-blo / SUVmax-liv has high specificity but poor sensitivity when used alone. However, the combination of predictive factors improves sensitivity without reducing specificity and has potential value in differentiating SLE-LAD patients from lymphoma patients.

[0075] Table 7 18 Univariate logistic regression analysis of F-FDG PET / CT related indicators and lymphoma

[0076] Note: Suvmax-Liv: liver suvmax; SUVmax-blo: blood pool SUVmax; TLR: target liver ratio; TBR: target blood pool ratio; OR: odds ratio; CI: confidence interval.

[0077] Table 8 18 Multivariate logistic regression analysis of F-FDG PET / CT related indicators and lymphoma

[0078] Note: Suvmax-Liv: liver suvmax; SUVmax-blo: blood pool SUVmax; TLR: target liver ratio; OR: odds ratio; CI: confidence interval.

[0079] Table 9. Diagnostic value of ROC for relevant predictive factors

[0080] Note: Suvmax-Liv: liver SUVmax; SUVmax-blo: blood pool SUVmax; TLR: target / liver ratio; TBR: target / blood pool ratio.

[0081] In one specific embodiment, SLE combined with LAD not only complicates the patient's clinical presentation but also increases the difficulty of differential diagnosis and management with other diseases. Currently, clinical research on SLE-LAD is relatively limited, with few series of studies on the application of 18F-FDG PET / CT in this population, mainly consisting of case reports. In 2004, Geoffrey et al. found significant FDG uptake on 18F-FDG PET / CT imaging in a 16-year-old SLE-LAD patient, similar to the imaging features of PET in lymphoma. Subsequently, case reports of the application of 18F-FDG PET / CT in SLE-LAD have emerged, providing important imaging evidence for the systematic evaluation of SLE and allowing for optimized biopsy strategies by locating hypermetabolic lymph nodes, guiding pathological biopsy in SLE-LAD. PET / CT-guided targeted biopsy can improve diagnostic efficiency, but further research is needed to differentiate SLE-LAD subtypes (such as HNL, reactive hyperplasia) based solely on metabolic characteristics; currently, it still requires combination with pathological biopsy for differentiation.

[0082] This invention analyzed the clinical manifestations, laboratory tests, and 18F-FDG PET / CT metabolic parameters of 10 SLE-LAD patients (excluding those with infection or tumors) from a single center and compared them with 30 lymphoma patients. The results showed that SLE-LAD is often accompanied by hematologic involvement, skin and mucous membrane damage, and moderate disease activity; and lymphoma patients had higher lymph node PET / CT metabolic parameter values. Furthermore, the diagnostic model combining SUVmax-blo / SUVmax-liv and the hottest lesion TLR in 18F-FDG PET / CT has potential value in differentiating SLE-LAD patients from lymphoma patients.

[0083] The SUV value, a commonly used metabolic parameter in PET / CT, is easily affected by various factors, including the activity of the injected imaging agent, the dosage of the injected imaging agent, the PET equipment and the measurement and reconstruction methods, as well as the individual characteristics of the examinee. This leads to limited comparability of SUV values ​​between different studies. This invention reduces the interference of the aforementioned confounding factors to a certain extent by applying the ratio of the SUVmax of the target region to the SUVmax of the background region, i.e., the target-to-background ratio.

[0084] This invention demonstrates that 18F-FDG PET / CT metabolic parameters have significant clinical value in the differential diagnosis of SLE and LAD. The SUVmax and TLR of the hottest lesion were significantly higher in the lymphoma group than in the SLE group, reflecting the metabolic differences in malignant proliferation and inflammatory activation, suggesting that 18F-FDG PET / CT can provide important clues for the differential diagnosis of SLE and lymphoma. Considering that SUV values ​​are easily affected by multiple factors, this invention introduces the SUVmax-blo / SUVmax-liv ratio for logistic regression analysis, followed by multivariate logistic regression analysis using meaningful and interesting correction indicators. Logistic regression analysis shows that SUVmax-blo / SUVmax-liv is an independent predictor of distinguishing between SLE and lymphoma. The combined predictor obtained by combining SUVmax-blo with the TLR of the hottest lesion shows high differential diagnostic performance (AUC=0.813, sensitivity 90.0%, specificity 66.7%). This suggests that the combined application of corrected 18F-FDG PET / CT metabolic parameters has potential value in the differential diagnosis of SLE and lymphoma, and is expected to serve as a non-invasive diagnostic tool to guide clinical practice.

[0085] The method of this invention uses computer-assisted identification, and doctors or other medical personnel refer to the identification results to give a final diagnosis.

[0086] The present invention also discloses a computer program product or system, including a computer program that, when executed by a processor, implements the above-described method steps.

[0087] Figure 13 The schematic diagram of a system for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma provided in this embodiment of the invention specifically includes: Acquisition Unit: Acquires patient PET / CT images; ROI unit: The image is divided into ROI regions to obtain ROI regions; Feature unit: The metabolic features are calculated by performing metabolic feature calculation on the ROI region, and the metabolic features are the ratio of SUVmax-blo to SUVmax-liv; Differentiation Unit: Based on the aforementioned metabolic characteristics, the differentiation between systemic lupus erythematosus with lymphadenopathy and lymphoma is obtained.

[0088] Figure 14 An embodiment of the present invention provides a schematic diagram of a computer device, specifically including: A memory and a processor; the memory is used to store program instructions; the processor is used to invoke the program instructions, when any one of the program instructions is executed, as described above in the method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma.

[0089] The present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, provides any of the above-described methods for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma.

[0090] The verification results of this verification embodiment show that assigning inherent weights to indications can improve the performance of this method compared to the default settings. Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for example, the division of units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be electrical, mechanical, or other forms. The units described as separate components may or may not be physically separated; the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of this embodiment. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0091] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0092] The computer device provided by the present invention has been described in detail above. For those skilled in the art, there will be changes in the specific implementation and application scope based on the ideas of the embodiments of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma, characterized in that, include: Acquire patient PET / CT images; The image is divided into Regions of Interest (ROIs) to obtain ROI regions; Metabolic features are calculated for the ROI region, and the metabolic features are the ratio of SUVmax-blo to SUVmax-liv. Based on the aforementioned metabolic characteristics, the differential diagnosis between systemic lupus erythematosus with lymphadenopathy and lymphoma was obtained.

2. The method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma according to claim 1, characterized in that, The metabolic characteristics are replaced with the hottest lesion TLR, and the results of the differentiation of systemic lupus erythematosus with lymphadenopathy or lymphoma are obtained by the hottest lesion TLR. Optionally, the metabolic characteristics also include the hottest lesion TLR, and the identification result of systemic lupus erythematosus with lymphadenopathy or lymphoma is obtained by the ratio of the hottest lesion TLR, SUVmax-blo and SUVmax-liv; Optionally, the calculation process for the hottest lesion TLR is as follows: obtain the metabolic parameters of the liver, and calculate the hottest lesion TLR by combining the metabolic parameters of the liver with the metabolic parameters of the ROI region.

3. The method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma according to claim 1 or 2, characterized in that, The metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax. The results of the identification of systemic lupus erythematosus with lymphadenopathy or lymphoma are obtained by using the ratio of SUVmax-blo and / or hottest lesion SUVmax, SUVmax-blo and SUVmax-liv. Optionally, the metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax, and the identification results of systemic lupus erythematosus with lymphadenopathy or lymphoma are obtained by differentiating SUVmax-blo and / or hottest lesion SUVmax and hottest lesion TLR. Optionally, the metabolic characteristics also include SUVmax-blo and / or hottest lesion SUVmax, and the identification results of systemic lupus erythematosus with lymphadenopathy or lymphoma are obtained by using SUVmax-blo and / or hottest lesion SUVmax, the ratio of SUVmax-blo and SUVmax-liv, and the hottest lesion TLR.

4. The method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma according to any one of claims 1-3, characterized in that, The method also includes immune characteristics, obtaining immune data from patients to obtain immune characteristics, and differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma based on the metabolic and immune characteristics to obtain the differentiation results; the immune characteristics include IgA, C4 and / or C3.

5. The method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma according to any one of claims 1-4, characterized in that, The method also includes clinical features, obtaining clinical data from patients to obtain clinical features, and differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma based on the metabolic features and clinical features to obtain the differentiation results; the clinical features include any one or more of the following: SLEDAI-2K score and platelet count, absolute lymphocyte count, and maximum standardized uptake value of spleen. Optionally, the differentiation between systemic lupus erythematosus with lymphadenopathy and lymphoma can be obtained based on the aforementioned metabolic, immune, and clinical characteristics.

6. The method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma according to any one of claims 1-5, characterized in that, The identification is performed through an identification model. The metabolic characteristics are input into the identification model to obtain the identification results of systemic lupus erythematosus with lymphadenopathy or lymphoma. Optionally, the discrimination model may employ any one or more of the following: convolutional neural network, residual network, XGBoost, SVM, random forest, and logistic regression; Optionally, the training process of the discrimination model is as follows: acquire a PET / CT image dataset, divide the PET / CT image dataset into ROI regions to obtain ROI regions, extract metabolic features from the ROI regions and calculate the metabolic features, input the metabolic features into the discrimination model to be trained, until the loss function remains unchanged, and obtain the discrimination model; Optionally, the metabolic and immune features are input into a second differential diagnosis model to obtain the differential diagnosis result of systemic lupus erythematosus with lymphadenopathy or lymphoma; the second differential diagnosis model is obtained by training on the metabolic and immune features. Optionally, the metabolic and clinical features are input into a third differential diagnosis model to obtain the differential diagnosis results for systemic lupus erythematosus with lymphadenopathy or lymphoma; the third differential diagnosis model is obtained through training on the metabolic and clinical features. Optionally, the metabolic, immune, and clinical features are input into a fourth differential diagnosis model to obtain the differential diagnosis results for systemic lupus erythematosus with lymphadenopathy or lymphoma; the fourth differential diagnosis model is obtained by training on the metabolic, immune, and clinical features.

7. The method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma according to claim 1, characterized in that, The ROI region is the mediastinal blood pool, and metabolic characteristics are obtained by calculating the metabolic characteristics of the mediastinal blood pool. Optionally, metabolic characteristics can be calculated for the luminal region of the mediastinal blood pool ascending aorta or aortic arch.

8. A computer program product comprising a computer program or instructions, characterized in that, The computer program or instructions are executed by a processor to implement the method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma as described in any one of claims 1-7.

9. A computer device comprising a memory, a processor, and a computer program or instructions stored in the memory, characterized in that, The computer program or instructions are executed by a processor to implement the method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, The computer program or instructions are executed by a processor to implement the method for differentiating systemic lupus erythematosus with lymphadenopathy from lymphoma as described in any one of claims 1-7.

Images