18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery

18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article 18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery Gong Zhang, Chengkai Bao, Yanzhe Liu, Zizheng Wang, Lei Du, Yue Zhang, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2578400/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 25 May, 2023 Read the published version in EJNMMI Research → Version 1 posted 3 You are reading this latest preprint version Abstract Background ：The determination of pathological grading has a guiding significance for the treatment of pancreatic ductal adenocarcinoma（PDAC）patients. However, there is a lack of an accurate and safe method to obtain pathological grading before surgery. The aim of this study is to develop a deep learning（DL）model based on 18F-FDG-PET/CT for a fully automatic prediction of preoperative pathological grading of pancreatic cancer. Results ：A total of 370 PDAC patients from January 2016 to September 2021 were collected retrospectively. All patients underwent 18F-FDG-PET/CT examination before surgery and obtained pathological results after surgery. A DL model for pancreatic cancer lesion segmentation was first developed using 100 of these cases and applied to the remaining cases to obtain lesion regions. After that, all patients were divided into training set, validation set and test set according to the ratio of 5:1:1. A predictive model of pancreatic cancer pathological grade was developed using the features computed from the lesion regions obtained by the lesion segmentation model and key clinical characteristics of the patients. Finally, the stability of the model was verified by 7-fold cross-validation.The Dice score of the developed PET/CT based tumour segmentation model for PDAC was 0.89. The area under curve (AUC) of the PET/CT-based DL model developed on the basis of the segmentation model was 0.74, with an accuracy, sensitivity, and specificity of 0.72, 0.73, and 0.72, respectively. After integrating key clinical data, the AUC of the model improved to 0.77, with its accuracy, sensitivity, and specificity boosted to 0.75, 0.77, and 0.73, respectively. Conclusion ：To the best of our knowledge, this is the first deep learning model to end-to-end predict the pathological grading of PDAC in a fully automatic manner, which is expected to improve clinical decision making. deep learning pancreatic cancer PET/CT pathological grading prediction model Figures Figure 1 Figure 2 Figure 3 Figure 4 Background Pancreatic cancer is a common malignancy and is the fourth most deadly cancer in the world 1 , killing approximately 480,000 people worldwide each year. In the next decade, pancreatic cancer is likely to become the second leading cause of death 2 . Surgery is currently the only treatment that may cure pancreatic cancer. Statistics show 3 that 10% of patients with indications for surgical resection have a 5-year survival rate of 24.6% (and 2.9%) for patients who have (and have not) undergone PDAC resection in Stage I. Predicting pathological grading of pancreatic cancer is an important part of the diagnosis and treatment of pancreatic cancer. Pathological differentiation of pancreatic cancer helps assess the extent, depth, and metastatic status of pancreatic cancer, which is an important basis for determining the best treatment plan and predicting prognosis and plays an important role in guiding surgery and corresponding adjuvant therapy for precise individualized treatment. A study by Golan et al. showed that well-differentiated PDAC was associated with long-term survival after surgery 4 . In contrast, poor differentiation is an independent prognostic factor affecting overall survival 5 . For patients with poorly differentiated PDAC, neoadjuvant therapy may provide longer survival than direct surgery 6-8 . The only method currently available to determine PDAC grading preoperatively is ultrasound or CT-guided puncture biopsy. Tumor tissue columns obtained in this manner do not reliably reflect the structural features of the entire lesion due to their high heterogeneity 9 . A study by Larghi et al. showed that the preoperative grading of EUS-FNB had an accuracy of 56%, sensitivity of 41%, and specificity of 78% 10 . Therefore, a safe and accurate preoperative method for determining the degree of differentiation of PDAC is needed. The determination of the pathological grading of PDAC relies on pathological slices as pathological examination is the gold standard for diagnosing the disease. However, since pathological tissue is obtained through invasive puncture or surgery, the pathological results have a pronounced lag. Currently, it is common to diagnose PDAC using medical images, such as CT, MRI, or PET, which can be acquired noninvasively; the diagnosis results are obtained in a shorter time. However, due to low image resolution, the information in medical images is not as clear as that in pathological slices. Also, the variations in equipment and operator lead to unstable imaging results, or at least not as stable as pathological examination. Therefore, how to automatically predict the pathological grading of PDAC through imaging data is a challenging task and has not been investigated in the literature, to the best of our knowledge. In this paper, we attempt to bridge this gap. Deep learning is a machine learning method that automatically learns features and classifies through the design and use of multi-layer networks 11,12 . In recent years, deep learning has been widely applied in medical image analysis, including PET/CT image analysis 13 . PET/CT is a whole-body functional imaging examination, which reflects the malignancy or benignity of lesions through the metabolic activity of cells. Wang et al. 14 studied the use of deep learning models to segment lung cancer in PET/CT images and achieved a high accuracy. Chao et al. 15 used a dual-energy CT based deep learning radiomics model to classify PDAC's lymph node metastasis (LNM) status, and the model's AUC, sensitivity, and specificity were 0.87, 76% and 83%, respectively. These studies indicate that deep learning models based on PET/CT have a high accuracy and sensitivity. However, current research on pathological grading of PDAC is still limited. Therefore, our goal is to establish a fully automated deep learning model based on 18 F-FDG-PET/CT for predicting preoperative pathological grading of PDAC. Materials And Methods Fig. 1 presents the schematic workflow of the proposed DL model based on PET/CT for pathological grading of patients with PDAC, which consists of multiple processing stages. Below, we elaborate the details related to the workflow, starting with study population, image labeling, to model construction, and finally model testing. Study population Patients who underwent pancreatic surgery at the PLA General Hospital from January 2016 to September 2021 and obtained pathological confirmation of PDAC were collected and included in the study according to the inclusion and exclusion criteria, and 370 patients were finally included. Inclusion criteria: (i) PDAC was pathologically confirmed by radical pancreatic resection; (ii) PDAC was confirmed by pathological biopsy of non-radical pancreatic surgery; (iii) PET/CT of the pancreas was performed within 1 month before surgery. Exclusion criteria: (i) patients had adjuvant treatment such as radiotherapy, chemotherapy and intervention before surgery; (ii) PET/CT images were of poor quality (tumour and borders could not be distinguished with the naked eye or there were artifacts interfering) and could not be used to analyze patients; (iii) other malignant tumours were combined; (iv) there was no significant FDG uptake at the tumour site (SUVmax < 2.5); (v) pathological findings and images could not correspond. Clinical data such as a patient’s age, gender, preoperative CA199 level, tumour location, tumour size (long and short diameter) on PET/CT images, and SUVmax values were also collected. PET-CT image labeling process Supplementary Method 1.1 provides detailed information about the PET/CT scanning protocol. The regions of abnormal 18 F-FDG uptake on PET and density abnormality on CT are localized as the lesion region as follows. After the PET/CT image fusion is completed, two experienced PET/CT diagnostic physicians use 3D slicers (version 5.1.0, https://www.slicer.org) software with a threshold of 40% SUVmax to draw out the ROI (Region of Interest) of the target lesion, and all discrepancies are confirmed through discussion. All examination images are interpreted by two senior nuclear medicine specialists (with more than 5 years of PET interpretation experience), including the location of tumor lesion, its relationship with surrounding tissue, the presence of lymph node metastasis, the presence of distant metastasis, and the performance under different sequences. Constructing the lesion segmentation model The whole process of building the deep model for lesion segmentation is shown in Figure 1A. 100 cases of annotated PET/CT images of pancreatic cancer were input into the segmentation model for training. The PET-CT images were first pre-processed: a. Window width and window level (350, 40) were applied to intercept the gray value; b. Each pair of 3D CT series (512*512*H CT ) and 3D PET series (of size 96*96*H PET , 128*128*H PET , 168*168*H PET, 170*170*H PET ) were uniformly resized to 256*256*H PET ; c. For each slice, the gray scale was normalized [0,1]; d. 3 PET slices and 3 CT slices centered around the corresponding location form an 6-channel input to the model. Model construction: The 6-channel input has a PET part and a CT part, each fed into a 2D-Unet network (Fig.S1.) branch with no shared parameters. The feature vectors of the two 2D-Unets are then concatenated to pass through convolutions, which output the final lesion segmentation mask. The learning rate was set to 1 × 10 -5 and the parameters were updated using the Adam optimiser. Post-processing the segmentation result: a. A pre-trained model of organ segmentation was added to nnUnet 16 to provide a coarse segmentation of the abdominal organs. The predicted segmentation of pancreatic organ was then fused with the tumour segmentation results to enhance the lesion segmentation performance; b. Medical image analysis techniques including erosion, expansion, and SUVmax 40% threshold segmentation are further applied to obtain the final lesion segmentation results. Building PDAC pathological grade classification models Due to the low prevalence of pathological samples with extreme pathological differentiation grades in the clinic, the grades with few samples were merged in this study and all samples were set to two predictive labels: low grade or high grade. Highly, moderate-highly, and moderately differentiated pathologies were defined as low grade; undifferentiated, lowly, and moderate-lowly differentiated were defined as high grade (Fig. S2.). This is similar to the classification method of Wasif and Rochefort et al. 17,18 According to the segmentation result, the lesion regions were cropped out of the 3D data of PET, CT, and segmentation Mask, respectively, and three aligned copies of size 64*64*16 (length*width*height) were obtained. The CT data was intercepted with a window width and window level (350, 40) and normalized to [0,1], and the PET data was normalized to [0,1]. The cropped PET, CT, and Mask were concatenated in the channel dimension to obtain a tensor of size 3*64*64*16 (number_of_channels*length*width*height). The tensor was fed into a Unet3D-based Encoder to extract image feature vectors as shown in Figure 1B. Cases with clinical data missing ratios greater than 20% were excluded from our study. A total of 21 clinical variables were collected to build predictive models based on clinical experience and literature reports. Subsequently, the individual clinical data was analysed for significance using the Random Forest method (Fig.S3.). Eleven important clinical characteristics including age, BMI, SUVmax, ALT, AST, total bilirubin, direct bilirubin, blood glucose, CEA, CA125 and CA199 were kept. Finally, the clinical data feature vectors were extracted using the MLP through the multilayer perceptron. The part was shown in Figure 1C-D. Both image features or clinical data features can be used to obtain prediction results for their respective modalities through the fully connected (FC) layer. To obtain better prediction performance, we replaced the last FC layer with a TMC (Trusted Multi-view Classification) 19 to integrate image features and clinical data features and constructed a PET/CT + Clinical data model. TMC is a new multi-view classification algorithm that dynamically integrates different views at an evidence level to promote classification reliability by considering evidence from each view（Supplementary Method 1.2）. The learning rate was set to 1×10 -5 and the parameters in the feature extractor were updated using the Adam optimiser. Seven-fold cross validation for model testing We used a 7-fold cross-validation to better evaluate the generalization ability of the model. This is shown in Figure 2. We divided 370 patients into 7 folds, of which 5 folds were the training set for the model in each training round, 1 fold was the internal validation set and 1 fold was used as the test set to test the final performance of the model. The next round was trained by changing the order of the training, validation and test folds. The final model is obtained by averaging the results of the 7 folds. Statistical analysis The clinical data was statistically processed using SPSS 22.0 statistical analysis software: normally distributed measures were expressed as x±s and comparisons between groups were made using the Student-t test. Skewed measures were expressed as median (range) and comparisons of count data were made using the 𝑋^2 test or Fisher's exact probability method. Dice score was used to evaluate the pancreatic lesion segmentation model. Accuracy, sensitivity, and specificity of the test dataset results were calculated using receiver operator characteristic curve (ROC) for the classification models. P values less than 0.05 were considered statistically significant. Results Patient baseline characteristics From January 2016 to September 2021, 613 consecutive patients with PDAC were retrospectively recruited to our cancer center. Of these, 370 patients (164 women and 206 men; mean age 60.08 ± 9.36 years) were finally screened. These patients were divided into two cohorts based on pathological grading. There were 190 cases in the LG group and 180 cases in the HG group. Table 1 summarizes the baseline characteristics of the patients in the LG and HG groups. Table.1 Baseline characteristics of patients Variable Low grade (n = 190) High grade (n =180) P-value Demographics Age (y), mean ± SD 59.90±9.22 60.28±9.52 0.698 Gender 0.066 Female 93（49%） 71（39%） Male 97（51%） 109（61%） BMI† 23.4（21.47-25.9） 23.44（22.03-25.13） 0.449 Smoke 59（31%） 60（33%） 0.639 Drink alcohol 45（24%） 60（33%） 0.04 Abdominal discomfort 88（46%） 99（55%） 0.095 Weight loss（＞5kg） 39（21%） 57（30%） 0.015 PET/CT report Tumor size (mm)† Max 31.5（26.0-45.0） 31（27.0-42.0） 0.728 Min 26（20.0-35.5） 25（19.5-33.0） 0.254 Tumor SUVmax† 5.75（4.4-7.95） 7.5（5.6-9.6） 0.001 Location 0.01 Head-Neck 103（54%） 121（67%） Body-Tail 87（46%） 59（33%） Pathological report Neuroaggression 113（59%） 109（61%） 0.832 Cancerembolus 15（8%） 38（21%） 0.001 Lymph node metastasis 58（31%） 58（32%） 0.834 Laboratory findings † ALT(U/L) 22.2（11.5-58.7） 17.2（13.2-98.5） 0.159 AST(U/L) 17.70（13.5-49.3） 18.75（13.5-50.8） 0.369 Total-bilirubin(umol/L) 12.4（8.7-18.6） 15.7（8.7-32.7） 0.007 Direct-bilirubin(umol/L) 4.4（2.9-8.8） 4.9（2.7-20.5） 0.037 Glucose (mmol/L) 5.97（5.2-7.0） 5.48（4.8-6.9） 0.162 CEA (μg/L) 2.77（1.8-5.0） 3.02（2.0-4.7） 0.325 CA125(U/mL) 14.46（8.7-25.4） 18.96（11.7-34.7） 0.004 CA199(U/mL) 170.25（47.8-611.2） 229（70.5-738.4） 0.027 † Data in parentheses are the interquartile range Abbreviation: BMI , Body Mass Index; ALT , alanine aminotransferase; AST , aspartate amino transferase; CEA , Carcinoma Embryonic Antigen; CA 125 , carbohydrate antigen 125; CA 199 , carbohydrate antigen 199 No significant differences in clinical characteristics were observed between these two cohorts. Alcohol consumption rates were 24.0% (45/190) and 33.3% (60/180) respectively, with a significant difference between the two cohorts (p=0.04). A total of 21.0% (39 out of 190) of the LG group and 30% (87 out of 180) of the HG group had significant weight loss, with a statistically significant difference between the two groups (p=0.015). All PET/CT reports were independently reviewed and assessed by two experienced PET/CT diagnosticians. The median value of SUVmax was 5.75 (2.2-36.0) in the LG group and 7.5 (3.2-31.1) in the HG group, with a statistically significant difference between the two groups (p=0.001). Tumours in the HG group were more likely to be found in the head and neck of the pancreas than in the LG group 67% vs 54% (p=0.01), a statistically significant difference between the two groups. In laboratory tests, total and direct bilirubin levels were lower in the LG group than in the HG group (12.4 vs 15.7, p=0.007; 4.4 vs 4.9, p=0.037), with a statistically significant difference between the two groups. In terms of tumour marker detection, the LG group had lower levels of CA-125 and CA199 than the HG group (14.46 vs 18.96, p=0.004; 170.25 vs 229, p=0.027). Performance of lesion segmentation model The Dice score for the lesion segmentation Unet model in Table.S1. is 0.72. The Dice score for Unet prediction with guidance of organ location (Unet+OL), that is, with the addition of nnUnet-based organ segmentation, was increased to 0.76. The Dice score for Unet+OL prediction with post-processing (Unet+OLP), which is Unet+OL with the addition of post-processing such as erosion, expansion, and threshold segmentation, was improved to 0.89, which is a significant advantage compared to the Unet and Unet+OL models. As can be seen in Figure 3, in the three validated cases, the region of lesions output by Unet+OLP was closer to the Ground Truth (GT) labeling than that of Unet+OL. Performance of PDAC pathological grade classification model Regarding testing performance, Figure 4 shows that the AUC of the clinical data model was 0.95 in the training cohort, 0.68 in the validation cohort and 0.68 in the test cohort, while the AUC of the PET/CT model was 0.99 in the training cohort, 0.72 in the validation cohort and 0.74 in the test cohort, which was better than the clinical model. In order to improve the efficacy and accuracy of the model, we combined the clinical model with the PET/CT DL model to build a PET/CT + Clinical data model. The AUC of the PET/CT+ Clinical data model reached 0.99, 0.74, and 0.77 in the training, validation, and test cohorts, respectively. It can be seen in Table 2 that the accuracy, sensitivity, and specificity of the PET/CT model were 72%, 73%, and 72%, respectively. The accuracy, sensitivity, and specificity of the clinical data model were 66%, 67%, and 66%, respectively. The accuracy, sensitivity, and specificity of PET/CT+ Clinical data model were 75%, 77%, and 73% respectively. The model that integrates both image and clinical data achieved the best performance. Table.2 The performance comparison of different models Models Cohorts AUC ACC SENS SPEC PPV NPV PET/CT Train 0.99 0.97 0.97 0.97 0.97 0.96 Val 0.72 0.73 0.72 0.74 0.75 0.70 Test 0.74 0.72 0.73 0.72 0.72 0.72 Clinical Train 0.95 0.91 0.90 0.91 0.92 0.89 Val 0.68 0.67 0.66 0.69 0.75 0.59 Test 0.68 0.66 0.67 0.66 0.69 0.63 PET/CT+Clinical Train 0.99 0.98 0.98 0.97 0.98 0.98 Val 0.73 0.76 0.75 0.76 0.78 0.73 Test 0.77 0.75 0.77 0.73 0.73 0.76 AUC, area under receiver operating characteristic curve; ACC, accuracy; SENS, sensitivity; SPEC, specificity; PPV, positive predictive value; NPV, negative predictive value Discussion Currently, research on deep learning models for PDAC mainly focuses on the disease's differential diagnosis, preoperative staging, and prognostic analysis. Wei et al. 20 used a combination of machine learning and deep learning algorithms to extract features from PET/CT images to predict the difference between PDAC and autoimmune pancreatitis, developing a multi-domain fusion model with an overall performance of AUC, accuracy, sensitivity, and specificity of 0.96, 0.90, 0.88, and 0.93, respectively. Bian et al. 21 developed and validated an automated preoperative AI algorithm for tumor and lymph node segmentation in CT imaging to predict LN metastasis in PDAC patients. Lee et al. 22 developed a deep learning model based on clinical data to predict postoperative survival in pancreatic cancer patients, and the model's performance in predicting 2-year OS was comparable to AJCC (AUC, 0.67; P=0.35), and it was better than AJCC in predicting 1-year recurrence free survival (AUC, 0.54; P=0.049). Yao et al. 23 used deep learning to analyze preoperative multi-phase CT and developed imaging-derived biomarkers for predicting overall survival in PDAC, which can be used to predict OS in resectable PDAC patients. Research has found that the pathological grade of PDAC is largely determined by the fibrous matrix quality in its stroma. Tumors with lower differentiation have more fibrous matrix and occupy more of the contrast agent 24 . This provides a principle for pathological grading of PDAC through imaging studies. For example, Tikhonova et al. 25 used a machine learning algorithm to establish a diagnostic model for image-based PDAC grading based on preoperative CT, using data from 91 patients to establish a diagnostic model. The AUC for pathological grading ≥ 2 (or 3) was 0.75 (or 0.66). Na et al. 24 developed and validated a radiological feature based on contrast-enhanced computed tomography for preoperative prediction of histological grading of PDAC, and the AUC of the final validation set was 0.770. The above studies are all based on CT. PET/CT, integrating PET and CT into the same device and program, characterizes the lesion from different aspects and provides metabolic information from the former and detailed anatomical information from the latter, which makes the PET/CT image to have both good clarity and a strong ability to distinguish between lesion tissue and normal tissue 26 . Some research has shown that, based on 102 patients with histologically confirmed PDAC, FDG uptake is related to the invasiveness of pancreatic cancer, and SUVmax is significantly related to pathological grading 27 . Therefore, our goal is to unleash the potential of PET/CT in pathological grading of PDAC. In this study, in order to achieve the prediction of fully automatic PDAC pathological grading and reduce the impact of confounding factors in PET/CT images as much as possible, we first developed a deep learning model for PDAC lesion segmentation. Due to the presence of FDG uptake in organs such as the liver, adrenal gland, small intestine, and bladder in addition to the pancreatic lesion, the model showed abnormal segmentation in the initial training (Figure S4). In order to increase the accuracy of segmentation, nnUnet's organ segmentation pre-training model was added to provide a rough segmentation of abdominal organs, which are filtered out to enhance the performance of pancreatic lesion segmentation. While nnUnet 16 is an excellent image segmentation model that demonstrated good performance in the field of medical image segmentation especially with a large amount of training images, it is difficult to directly obtain a good segmentation model due to the small number of pancreatic cancer sections in this study. We first trained the nnUnet model of pancreatic organ segmentation using abdominal CT sections of healthy people, preserved the parameters of the model, and applied it to the target cases. The presence of tumors greatly reduced the segmentation results of the nnUnet model, but the preliminary localization of the patient's pancreas can still be obtained. The addition of nnUnet as a filter increased the Dice score of the model from 0.72 to 0.76. In the post-processing, we added corrosion, expansion, SUVmax 40% threshold segmentation and other post-operations, and finally increased the Dice score to 0.86 (Table.S1.). Through these steps, the segmentation model's performance was able to achieve an acceptable level. In building a classification model for PET/CT images, to incorporate more effective information while minimizing the effect of segmentation errors, we cropped out a 3D patch from the raw image centered around the segmented mask. The accuracy of the final PET/CT-based classification model was 72%, sensitivity was 73%, and specificity was 72%. To further boost the classification performance, we resorted to clinical indicators that are important for revealing pancreatic cancer characteristics. For example, CA199 is closely related to the prognosis of pancreatic cancer patients 28,29 . A study found that CA199 produced by pancreatic cancer cell lines in vitro is associated with histological differentiation in nude mice in vivo 30 . Therefore, we extracted key clinical data indicators such as CA199 and combined them with the PET/CT model to optimize the prediction accuracy. In our early experiment, we used fully connected layers to connect PET/CT with the extracted clinical features, and achieved a minor performance improvement; in some folds in cross validation, the combined model performed worser than the original PET/CT model. Therefore, in order to better integrate the features of PET/CT and clinical data, we used TMC 19 to improve the reliability of classification. This model parameterizes different data and combines them based on Dempster-Shafer theory, which improves the reliability and robustness of the classification model and improves the performance of the model (refer to Supplementary Method 1.2). Finally, to test the generalization ability of the model, we utilized 7-fold cross-validation. The final PET/CT + Clinical data model achieved an accuracy of 75%, sensitivity of 77%, and specificity of 73%. The deep learning model of PET/CT + Clinical data has a significant improvement compared to the traditional EUS-FNB 10 with an accuracy of 56% and sensitivity of 41%. Our research has a few limitations. Firstly, our data is from a single center and lacks external datasets for validation and evaluation. However, because the medical images were acquired using three different PET machines, it increases the generalizability. Additionally, using 7-fold cross-validation ensured the stability of the final model, which partially compensated for the lack of external validation sets. Secondly, all the data included in this study were from surgical patients, so the pathological differentiation was concentrated; however, the data for patients with excessive malignancy was relatively scarce, which did not make the model to show more significant discriminability. Finally, this study only explored the relationship between imaging features and pathological differentiation of PDAC but did not investigate the survival outcome of the patients, which is more concerned by patients and surgeons. Further research is needed to study the survival of patients. Our developed deep learning model can automatically analyze PET/CT images without human intervention, reducing subjective errors and improving the accuracy and reliability of grading. It is important to note that deep learning models cannot completely replace professional knowledge of imaging and pathology. Before using deep learning models to predict pathological grading, professional knowledge needs to be combined for interpretation and evaluation to ensure the accuracy and reliability of the results. In conclusion, deep learning-based PET/CT has a great potential in pathological grading of pancreatic cancer, and more clinical studies are needed to prove its safety and effectiveness before deep learning models can replace traditional pathological staging methods. Conclusions To the best our knowledge, this is the first report of using a DL model for preoperative prediction of PDAC pathological grading using PET/CT. The model's predictive performance was improved by combining features of PET/CT and key clinical data. Abbreviations PDAC Pancreatic ductal adenocarcinoma DL Deep learning AUC Area under curve SUV Standardized uptake value LNM Lymph node metastasis LG Low grade HG High grade FC Fully connected TMC Trusted Multi-view Classification ROC Receiver operator characteristic curve GT Ground truth OL Organ location OLP Organ location with post-processing ACC Accuracy SENS Sensitivity SPEC Specificity PPV Positive predictive value NPV Negative predictive value Declarations Supplementary Information Additional file 1: Fig.S1. Unet image segmentation network. Fig.S2. Distribution data of pathological differentiation degree in medical records. Fig.S3. Random forest analysis. Fig.S4. Examples of segmentation model before and after adding nnUet. Table.S1. The performance comparison of segmentation process. Acknowledgements Not applicable. Author Contributions All authors were involved in the study conception and design. GZ, YZL, ZZW, LD, FW, BXX were involved in acquisition of data. GZ, CKB, YZ, S. KZ participated in the development and testing of deep learning models. GZ, CKB, ZZW, YZ were involved in analysis and interpretation of the data. GZ, CKB and S. KZ were involved in drafting of the manuscript. RL is responsible for review, editing and supervision. All authors read and approved the final manuscript. Funding This work was supported by the National Key R&D Program of China (2021ZD0113301)； Availability of data and materials The relevant images and clinical data from this study are not available because they contain private patient information. However, such data can be obtained through agency approval and signed data use agreements and/or signed material transfer agreements. Ethics approval and consent to participate Approved by the Ethics Review Committee of the local institution of the PLA General Hospital, and the requirement of written informed consent was waived. All procedures involving human participants in this study comply with institutional and/or national Research Council ethical standards as well as the 1964 Declaration of Helsinki and its later amendments or similar ethical standards. Consent for publication Not applicable. Competing interests The authors declare no competing interests. References Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA: a cancer journal for clinicians. 2018;68(1):7-30. Hartwig W, Werner J, Jäger D, Debus J, Büchler MW. Improvement of surgical results for pancreatic cancer. The Lancet Oncology. 2013;14(11):e476-e485. Strobel O, Neoptolemos J, Jäger D, Büchler MW. Optimizing the outcomes of pancreatic cancer surgery. Nature reviews Clinical oncology. 2019;16(1):11-26. Golan T, Sella T, Margalit O, et al. Short- and Long-Term Survival in Metastatic Pancreatic Adenocarcinoma, 1993-2013. Journal of the National Comprehensive Cancer Network : JNCCN. 2017;15(8):1022-1027. Han SH, Heo JS, Choi SH, et al. Actual long-term outcome of T1 and T2 pancreatic ductal adenocarcinoma after surgical resection. International journal of surgery (London, England). 2017;40:68-72. de Geus SW, Eskander MF, Bliss LA, et al. Neoadjuvant therapy versus upfront surgery for resected pancreatic adenocarcinoma: A nationwide propensity score matched analysis. Surgery. 2017;161(3):592-601. Shubert CR, Bergquist JR, Groeschl RT, et al. Overall survival is increased among stage III pancreatic adenocarcinoma patients receiving neoadjuvant chemotherapy compared to surgery first and adjuvant chemotherapy: An intention to treat analysis of the National Cancer Database. Surgery. 2016;160(4):1080-1096. Versteijne E, Vogel JA, Besselink MG, et al. Meta-analysis comparing upfront surgery with neoadjuvant treatment in patients with resectable or borderline resectable pancreatic cancer. The British journal of surgery. 2018;105(8):946-958. Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. The New England journal of medicine. 2012;366(10):883-892. Larghi A, Correale L, Ricci R, et al. Interobserver agreement and accuracy of preoperative endoscopic ultrasound-guided biopsy for histological grading of pancreatic cancer. Endoscopy. 2015;47(4):308-314. Elemento O, Leslie C, Lundin J, Tourassi G. Artificial intelligence in cancer research, diagnosis and therapy. Nature reviews Cancer. 2021;21(12):747-752. Kleppe A, Skrede OJ, De Raedt S, Liestøl K, Kerr DJ, Danielsen HE. Designing deep learning studies in cancer diagnostics. Nature reviews Cancer. 2021;21(3):199-211. Choi H. Deep Learning in Nuclear Medicine and Molecular Imaging: Current Perspectives and Future Directions. Nuclear medicine and molecular imaging. 2018;52(2):109-118. Wang S, Mahon R, Weiss E, et al. Automated Lung Cancer Segmentation Using a PET and CT Dual-Modality Deep Learning Neural Network. International journal of radiation oncology, biology, physics. 2023;115(2):529-539. An C, Li D, Li S, et al. Deep learning radiomics of dual-energy computed tomography for predicting lymph node metastases of pancreatic ductal adenocarcinoma. European journal of nuclear medicine and molecular imaging. 2022;49(4):1187-1199. Isensee F, Petersen J, Klein A, et al. nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation. 2018. Wasif N, Ko CY, Farrell J, et al. Impact of tumor grade on prognosis in pancreatic cancer: should we include grade in AJCC staging? Annals of surgical oncology. 2010;17(9):2312-2320. Rochefort MM, Ankeny JS, Kadera BE, et al. Impact of tumor grade on pancreatic cancer prognosis: validation of a novel TNMG staging system. Annals of surgical oncology. 2013;20(13):4322-4329. Han Z, Zhang C, Fu H, Zhou JT. Trusted Multi-View Classification with Dynamic Evidential Fusion. IEEE transactions on pattern analysis and machine intelligence. 2022;Pp. Wei W, Jia G, Wu Z, et al. A multidomain fusion model of radiomics and deep learning to discriminate between PDAC and AIP based on (18)F-FDG PET/CT images. Japanese journal of radiology. 2022:1-11. Bian Y, Zheng Z, Fang X, et al. Artificial Intelligence to Predict Lymph Node Metastasis at CT in Pancreatic Ductal Adenocarcinoma. Radiology. 2023;306(1):160-169. Lee W, Park HJ, Lee HJ, et al. Preoperative data-based deep learning model for predicting postoperative survival in pancreatic cancer patients. International journal of surgery (London, England). 2022;105:106851. Yao J, Cao K, Hou Y, et al. Deep Learning for Fully Automated Prediction of Overall Survival in Patients Undergoing Resection for Pancreatic Cancer: A Retrospective Multicenter Study. Annals of surgery. 2022. Chang N, Cui L, Luo Y, Chang Z, Yu B, Liu Z. Development and multicenter validation of a CT-based radiomics signature for discriminating histological grades of pancreatic ductal adenocarcinoma. Quantitative imaging in medicine and surgery. 2020;10(3):692-702. Tikhonova VS, Karmazanovsky GG, Kondratyev EV, et al. Radiomics model-based algorithm for preoperative prediction of pancreatic ductal adenocarcinoma grade. European radiology. 2022. Beyer T, Antoch G, Müller S, et al. Acquisition protocol considerations for combined PET/CT imaging. Journal of nuclear medicine : official publication, Society of Nuclear Medicine. 2004;45 Suppl 1:25s-35s. Wang Z, Chen JQ, Liu JL, Qin XG, Huang Y. FDG-PET in diagnosis, staging and prognosis of pancreatic carcinoma: a meta-analysis. World journal of gastroenterology. 2013;19(29):4808-4817. Williams JL, Kadera BE, Nguyen AH, et al. CA19-9 Normalization During Pre-operative Treatment Predicts Longer Survival for Patients with Locally Progressed Pancreatic Cancer. Journal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract. 2016;20(7):1331-1342. Chen Y, Shao Z, Chen W, et al. A varying-coefficient cox model for the effect of CA19-9 kinetics on overall survival in patients with advanced pancreatic cancer. Oncotarget. 2017;8(18):29925-29934. Iwamura T, Taniguchi S, Kitamura N, et al. Correlation between CA19-9 production in vitro and histological grades of differentiation in vivo in clones isolated from a human pancreatic cancer cell line (SUIT-2). Journal of gastroenterology and hepatology. 1992;7(5):512-519. Supplementary Files Additionalfile1.docx Cite Share Download PDF Status: Published Journal Publication published 25 May, 2023 Read the published version in EJNMMI Research → Version 1 posted Reviewers agreed at journal 17 Feb, 2023 Editor assigned by journal 14 Feb, 2023 First submitted to journal 13 Feb, 2023 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-2578400","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":176798736,"identity":"391bc330-0990-4ca0-a24d-e39353ebd963","order_by":0,"name":"Gong Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYDACCRiDv7HhwAcDGzkStEgcPnhwRkGaMQlaGNKSD/N8OJxIUIf87OZnD7/8ssmTdzhjcNjGgDmBgf3w0Q34tDDOOWZuLNuXVmx4uMfgcI4BWx4DT1raDXxamCUSzKQlew4nbmw4A9LCU8wgwWOGVwubRPo3qJYcg8MWBhKJDYS08EjkmEl++HE4cT5DWsJhBgMDwlokJHLKpBkb0hI3SBw+cLDHIMGYjZBf5Gekb5P88ccmcX5/Y/OHH3/+y/GzHz6GVwsIMPO2MTAYHID5jpByEGD88QdoXQMxSkfBKBgFo2BEAgAeV1ExL/kPDgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-5079-2064","institution":"Medical School of Chinese PLA: Chinese PLA General Hospital","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Gong","middleName":"","lastName":"Zhang","suffix":""},{"id":176798737,"identity":"c2ea78f7-3943-4046-af0c-3bbfe69bd636","order_by":1,"name":"Chengkai Bao","email":"","orcid":"","institution":"University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Chengkai","middleName":"","lastName":"Bao","suffix":""},{"id":176798738,"identity":"e9e5c6b6-ca51-4f47-83bc-10629bc91f5f","order_by":2,"name":"Yanzhe Liu","email":"","orcid":"","institution":"Chinese PLA General Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yanzhe","middleName":"","lastName":"Liu","suffix":""},{"id":176798739,"identity":"ee443f6a-6025-43ca-80fb-3e6a0dc15b24","order_by":3,"name":"Zizheng Wang","email":"","orcid":"","institution":"5th Medical Center of Chinese PLA General Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zizheng","middleName":"","lastName":"Wang","suffix":""},{"id":176798740,"identity":"196b99d5-9ea8-48a4-b2c9-a014f9b24f4a","order_by":4,"name":"Lei Du","email":"","orcid":"","institution":"Chinese PLA General Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Lei","middleName":"","lastName":"Du","suffix":""},{"id":176798741,"identity":"21632b02-7c03-4abc-af97-6d8b732ae95e","order_by":5,"name":"Yue Zhang","email":"","orcid":"","institution":"University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yue","middleName":"","lastName":"Zhang","suffix":""},{"id":176798742,"identity":"54b03e3c-1c71-4bb4-931f-91859d23210f","order_by":6,"name":"Fei Wang","email":"","orcid":"","institution":"Chinese PLA General Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Fei","middleName":"","lastName":"Wang","suffix":""},{"id":176798743,"identity":"f18a1330-f051-4e39-b774-71564b5e60e9","order_by":7,"name":"Baixuan Xu","email":"","orcid":"","institution":"Chinese PLA General Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Baixuan","middleName":"","lastName":"Xu","suffix":""},{"id":176798744,"identity":"1bc1849e-8111-4f6b-87fe-926c473c40cb","order_by":8,"name":"S. Kevin Zhou","email":"","orcid":"","institution":"University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"S.","middleName":"Kevin","lastName":"Zhou","suffix":""},{"id":176798745,"identity":"17ea0ca9-3bf7-4799-90b2-5824ada5eba5","order_by":9,"name":"Rong Liu","email":"","orcid":"https://orcid.org/0000-0001-5170-6474","institution":"Chinese PLA General Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Rong","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2023-02-12 11:29:41","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2578400/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2578400/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13550-023-00985-4","type":"published","date":"2023-05-25T20:58:58+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":33192579,"identity":"2c0ba962-fbec-4fb1-b0bb-d734c2645589","added_by":"auto","created_at":"2023-02-20 18:21:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":201991,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe workflow of DL model based on PET/CT for pathological grading of patients with pancreatic ductal adenocarcinoma (PDAC)\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-2578400/v1/bb1262761bfbb95fcb87dbbf.png"},{"id":33192580,"identity":"4f9d1eb1-4a4b-4239-a67f-92538c0f8344","added_by":"auto","created_at":"2023-02-20 18:21:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":62103,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSeven-fold cross validation model\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-2578400/v1/b6d7e5939e725a14807a8326.png"},{"id":33192578,"identity":"810f3c92-cedf-4ef8-a514-3a99f2ac007d","added_by":"auto","created_at":"2023-02-20 18:21:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":122968,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCompare the output of different segmentation models (Unet+OL: direct Unet prediction with guidance of organ location; Unet+OLP: Unet+OL prediction with post-processing; GT: Ground Truth)\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-2578400/v1/2d80075305273558de7fd1b0.png"},{"id":33192582,"identity":"f84e7643-ceb3-4a13-a8c5-7415e8f96ff9","added_by":"auto","created_at":"2023-02-20 18:21:52","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":299556,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eReceiver operating characteristic (ROC) curve comparison among different models for predicting the pathological grade of PDAC.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-2578400/v1/44da587d6d35de22b2b622df.png"},{"id":44729850,"identity":"3a4964c0-7fee-4fab-ba53-9d00f559afcb","added_by":"auto","created_at":"2023-10-16 21:21:28","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1260966,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2578400/v1/6d2f9371-0c1b-47da-a1ff-78104dd7765e.pdf"},{"id":33192581,"identity":"2845ca85-6784-4f2d-8a07-27d589f178a5","added_by":"auto","created_at":"2023-02-20 18:21:51","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":404066,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile1.docx","url":"https://assets-eu.researchsquare.com/files/rs-2578400/v1/7ed0db0f072e06bb5167d97b.docx"}],"financialInterests":"","formattedTitle":"18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery","fulltext":[{"header":"Background","content":"\u003cp\u003ePancreatic cancer is a common malignancy and is the fourth most deadly cancer in the world\u003csup\u003e1\u003c/sup\u003e, killing approximately 480,000 people worldwide each year. In the next decade, pancreatic cancer is likely to become the second leading cause of death\u003csup\u003e2\u003c/sup\u003e. Surgery is currently the only treatment that may cure pancreatic cancer. Statistics show\u003csup\u003e3\u003c/sup\u003e that 10% of patients with indications for surgical resection have a 5-year survival rate of 24.6% (and 2.9%) for patients who have (and have not) undergone PDAC resection in Stage I.\u003c/p\u003e\n\u003cp\u003ePredicting pathological grading of pancreatic cancer is an important part of the diagnosis and treatment of pancreatic cancer. Pathological differentiation of pancreatic cancer helps assess the extent, depth, and metastatic status of pancreatic cancer, which is an important basis for determining the best treatment plan and predicting prognosis and plays an important role in guiding surgery and corresponding adjuvant therapy for precise individualized treatment. A study by Golan et al. showed that well-differentiated PDAC was associated with long-term survival after surgery\u003csup\u003e4\u003c/sup\u003e. In contrast, poor differentiation is an independent prognostic factor affecting overall survival\u003csup\u003e5\u003c/sup\u003e. For patients with poorly differentiated PDAC, neoadjuvant therapy may provide longer survival than direct surgery\u003csup\u003e6-8\u003c/sup\u003e. The only method currently available to determine PDAC grading preoperatively is ultrasound or CT-guided puncture biopsy. Tumor tissue columns obtained in this manner do not reliably reflect the structural features of the entire lesion due to their high heterogeneity\u003csup\u003e9\u003c/sup\u003e. A study by Larghi et al. showed that the preoperative grading of EUS-FNB had an accuracy of 56%, sensitivity of 41%, and specificity of 78%\u003csup\u003e10\u003c/sup\u003e. Therefore, a safe and accurate preoperative method for determining the degree of differentiation of PDAC is needed.\u003c/p\u003e\n\u003cp\u003eThe determination of the pathological grading of PDAC relies on pathological slices as pathological examination is the gold standard for diagnosing the disease. However, since pathological tissue is obtained through invasive puncture or surgery, the pathological results have a pronounced lag. Currently, it is common to diagnose PDAC using medical images, such as CT, MRI, or PET, which can be acquired noninvasively; the diagnosis results are obtained in a shorter time. However, due to low image resolution, the information in medical images is not as clear as that in pathological slices. Also, the variations in equipment and operator lead to unstable imaging results, or at least not as stable as pathological examination. Therefore, how to automatically predict the pathological grading of PDAC through imaging data is a challenging task and has not been investigated in the literature, to the best of our knowledge. In this paper, we attempt to bridge this gap.\u003c/p\u003e\n\u003cp\u003eDeep learning is a machine learning method that automatically learns features and classifies through the design and use of multi-layer networks\u003csup\u003e11,12\u003c/sup\u003e. In recent years, deep learning has been widely applied in medical image analysis, including PET/CT image analysis\u003csup\u003e13\u003c/sup\u003e. PET/CT is a whole-body functional imaging examination, which reflects the malignancy or benignity of lesions through the metabolic activity of cells. Wang et al.\u003csup\u003e14\u003c/sup\u003e studied the use of deep learning models to segment lung cancer in PET/CT images and achieved a high accuracy. Chao et al.\u003csup\u003e15\u003c/sup\u003e used a dual-energy CT based deep learning radiomics model to classify PDAC\u0026apos;s lymph node metastasis (LNM) status, and the model\u0026apos;s AUC, sensitivity, and specificity were 0.87, 76% and 83%, respectively.\u003c/p\u003e\n\u003cp\u003eThese studies indicate that deep learning models based on PET/CT have a high accuracy and sensitivity. However, current research on pathological grading of PDAC is still limited. Therefore, our goal is to establish a fully automated deep learning model based on \u003csup\u003e18\u003c/sup\u003eF-FDG-PET/CT for predicting preoperative pathological grading of PDAC.\u003c/p\u003e"},{"header":"Materials And Methods","content":"\u003cp\u003eFig. 1 presents the schematic workflow of the proposed DL model based on PET/CT for pathological grading of patients with PDAC, which consists of multiple processing stages. Below, we elaborate the details related to the workflow, starting with study population, image labeling, to model construction, and finally model testing.\u003c/p\u003e\n\u003ch2\u003eStudy population\u003c/h2\u003e\n\u003cp\u003ePatients who underwent pancreatic surgery at the PLA General Hospital from\u0026nbsp;January 2016 to September 2021\u0026nbsp;and obtained pathological confirmation of PDAC were collected and included in the study according to the inclusion and exclusion criteria, and 370 patients were finally included.\u003c/p\u003e\n\u003cp\u003eInclusion criteria: (i) PDAC was pathologically confirmed by radical pancreatic resection; (ii) PDAC was confirmed by pathological biopsy of non-radical pancreatic surgery; (iii) PET/CT of the pancreas was performed within 1 month before surgery. Exclusion criteria: (i) patients had adjuvant treatment such as radiotherapy, chemotherapy and intervention before surgery; (ii) PET/CT images were of poor quality (tumour and borders could not be distinguished with the naked eye or there were artifacts interfering) and could not be used to analyze patients; (iii) other malignant tumours were combined; (iv) there was no significant FDG uptake at the tumour site (SUVmax \u0026lt; 2.5); (v) pathological findings and images could not correspond. Clinical data such as a patient\u0026rsquo;s age, gender, preoperative CA199 level, tumour location, tumour size (long and short diameter) on PET/CT images, and SUVmax values were also collected.\u003c/p\u003e\n\u003ch2\u003ePET-CT image labeling process\u003c/h2\u003e\n\u003cp\u003eSupplementary Method 1.1 provides detailed information about the PET/CT scanning protocol. The regions of abnormal \u003csup\u003e18\u003c/sup\u003eF-FDG uptake on PET and density abnormality on CT are localized as the lesion region as follows. After the PET/CT image fusion is completed, two experienced PET/CT diagnostic physicians use 3D slicers (version 5.1.0, https://www.slicer.org) software with a threshold of 40% SUVmax to draw out the ROI (Region of Interest) of the target lesion, and all discrepancies are confirmed through discussion. All examination images are interpreted by two senior nuclear medicine specialists (with more than 5 years of PET interpretation experience), including the location of tumor lesion, its relationship with surrounding tissue, the presence of lymph node metastasis, the presence of distant metastasis, and the performance under different sequences.\u003c/p\u003e\n\u003ch2\u003eConstructing the lesion segmentation model\u003c/h2\u003e\n\u003cp\u003eThe whole process of building the deep model for lesion segmentation is shown in Figure 1A. 100 cases of annotated PET/CT images of pancreatic cancer were input into the segmentation model for training.\u003c/p\u003e\n\u003cp\u003eThe PET-CT images were first pre-processed: a. Window width and window level (350, 40) were applied to intercept the gray value; b. Each pair of 3D CT series (512*512*H\u003csub\u003eCT\u003c/sub\u003e) and 3D PET series (of size 96*96*H\u003csub\u003ePET\u003c/sub\u003e, 128*128*H\u003csub\u003ePET\u003c/sub\u003e, 168*168*H\u003csub\u003ePET,\u003c/sub\u003e 170*170*H\u003csub\u003ePET\u003c/sub\u003e) were uniformly resized to 256*256*H\u003csub\u003ePET\u003c/sub\u003e; c. For each slice, the gray scale was normalized [0,1]; d. 3 PET slices and 3 CT slices centered around the corresponding location form an 6-channel input to the model.\u003c/p\u003e\n\u003cp\u003eModel construction: The 6-channel input has a PET part and a CT part, each fed into a 2D-Unet network (Fig.S1.) branch with no shared parameters. The feature vectors of the two 2D-Unets are then concatenated to pass through convolutions, which output the final lesion segmentation mask. The learning rate was set to 1 \u0026times; 10\u003csup\u003e-5\u003c/sup\u003e and the parameters were updated using the Adam optimiser.\u003c/p\u003e\n\u003cp\u003ePost-processing the segmentation result: a. A pre-trained model of organ segmentation was added to nnUnet\u003csup\u003e16\u003c/sup\u003e to provide a coarse segmentation of the abdominal organs. The predicted segmentation of pancreatic organ was then fused with the tumour segmentation results to enhance the lesion segmentation performance; b. Medical image analysis techniques including erosion, expansion, and SUVmax 40% threshold segmentation are further applied to obtain the final lesion segmentation results.\u003c/p\u003e\n\u003ch2\u003eBuilding PDAC pathological grade classification models\u003c/h2\u003e\n\u003cp\u003eDue to the low prevalence of pathological samples with extreme pathological differentiation grades in the clinic, the grades with few samples were merged in this study and all samples were set to two predictive labels: low grade or high grade. Highly, moderate-highly, and moderately differentiated pathologies were defined as low grade; undifferentiated, lowly, and moderate-lowly differentiated were defined as high grade (Fig. S2.). This is similar to the classification method of Wasif and Rochefort et al.\u003csup\u003e17,18\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAccording to the segmentation result, the lesion regions were cropped out of the 3D data of PET, CT, and segmentation Mask, respectively, and three aligned copies of size 64*64*16 (length*width*height) were obtained. The CT data was intercepted with a window width and window level (350, 40) and normalized to [0,1], and the PET data was normalized to [0,1]. The cropped PET, CT, and Mask were concatenated in the channel dimension to obtain a tensor of size 3*64*64*16 (number_of_channels*length*width*height). The tensor was fed into a Unet3D-based Encoder to extract image feature vectors as shown in Figure 1B.\u003c/p\u003e\n\u003cp\u003eCases with clinical data missing ratios greater than 20% were excluded from our study. A total of 21 clinical variables were collected to build predictive models based on clinical experience and literature reports. Subsequently, the individual clinical data was analysed for significance using the Random Forest method (Fig.S3.). Eleven important clinical characteristics including age, BMI, SUVmax, ALT, AST, total bilirubin, direct bilirubin, blood glucose, CEA, CA125 and CA199 were kept. Finally, the clinical data feature vectors were extracted using the MLP through the multilayer perceptron. The part was shown in Figure 1C-D.\u003c/p\u003e\n\u003cp\u003eBoth image features or clinical data features can be used to obtain prediction results for their respective modalities through the fully connected (FC) layer. To obtain better prediction performance, we replaced the last FC layer with a TMC (Trusted Multi-view Classification)\u0026nbsp;\u003csup\u003e19\u003c/sup\u003e to integrate image features and clinical data features and constructed a PET/CT + Clinical data model. TMC is a new multi-view classification algorithm that dynamically integrates different views at an evidence level to promote classification reliability by considering evidence from each view（Supplementary Method 1.2）. The learning rate was set to 1\u0026times;10\u003csup\u003e-5\u003c/sup\u003e and the parameters in the feature extractor were updated using the Adam optimiser.\u003c/p\u003e\n\u003ch2\u003eSeven-fold cross validation for model testing\u003c/h2\u003e\n\u003cp\u003eWe used a 7-fold cross-validation to better evaluate the generalization ability of the model. This is shown in Figure 2. We divided 370 patients into 7 folds, of which 5 folds were the training set for the model in each training round, 1 fold was the internal validation set and 1 fold was used as the test set to test the final performance of the model. The next round was trained by changing the order of the training, validation and test folds. The final model is obtained by averaging the results of the 7 folds.\u003c/p\u003e\n\u003ch2\u003eStatistical analysis\u003c/h2\u003e\n\u003cp\u003eThe clinical data was statistically processed using SPSS 22.0 statistical analysis software: normally distributed measures were expressed as x\u0026plusmn;s and comparisons between groups were made using the Student-t test. Skewed measures were expressed as median (range) and comparisons of count data were made using the 𝑋^2 test or Fisher\u0026apos;s exact probability method.\u003c/p\u003e\n\u003cp\u003eDice score was used to evaluate the pancreatic lesion segmentation model. Accuracy, sensitivity, and specificity of the test dataset results were calculated using receiver operator characteristic curve (ROC) for the classification models. P values less than 0.05 were considered statistically significant.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003ePatient baseline characteristics\u003c/h2\u003e\n\u003cp\u003eFrom\u0026nbsp;January 2016 to September 2021, 613 consecutive patients with PDAC were retrospectively recruited to our cancer center. Of these, 370 patients (164 women and 206 men; mean age 60.08 \u0026plusmn; 9.36 years) were finally screened. These patients were divided into two cohorts based on pathological grading. There were 190 cases in the LG group and 180 cases in the HG group. Table 1 summarizes the baseline characteristics of the patients in the LG and HG groups.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable.1\u0026nbsp;\u003c/strong\u003eBaseline characteristics of patients\u003c/p\u003e\n\u003ctable border=\"0\" cellpadding=\"0\" cellspacing=\"0\" width=\"652\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003eLow grade (n = 190)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003eHigh grade (n =180)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003eP-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003e\u003cstrong\u003eDemographics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eAge (y), mean \u0026plusmn; SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e59.90\u0026plusmn;9.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e60.28\u0026plusmn;9.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.698\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.066\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e93（49%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e71（39%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e97（51%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e109（61%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eBMI\u0026dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e23.4（21.47-25.9）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e23.44（22.03-25.13）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.449\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eSmoke\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e59（31%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e60（33%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.639\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eDrink alcohol\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e45（24%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e60（33%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eAbdominal discomfort\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e88（46%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e99（55%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.095\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eWeight loss（＞5kg）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e39（21%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e57（30%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.015\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePET/CT report\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eTumor size (mm)\u0026dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eMax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e31.5（26.0-45.0）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e31（27.0-42.0）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.728\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eMin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e26（20.0-35.5）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e25（19.5-33.0）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.254\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eTumor SUVmax\u0026dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e5.75（4.4-7.95）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e7.5（5.6-9.6）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eLocation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eHead-Neck\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e103（54%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e121（67%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eBody-Tail\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e87（46%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e59（33%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePathological report\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eNeuroaggression\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e113（59%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e109（61%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.832\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eCancerembolus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e15（8%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e38（21%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eLymph node metastasis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e58（31%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e58（32%）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.834\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003e\u003cstrong\u003eLaboratory findings\u003c/strong\u003e\u0026dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eALT(U/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e22.2（11.5-58.7）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e17.2（13.2-98.5）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.159\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eAST(U/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e17.70（13.5-49.3）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e18.75（13.5-50.8）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.369\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eTotal-bilirubin(umol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e12.4（8.7-18.6）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e15.7（8.7-32.7）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.007\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eDirect-bilirubin(umol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e4.4（2.9-8.8）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e4.9（2.7-20.5）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.037\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eGlucose (mmol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e5.97（5.2-7.0）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e5.48（4.8-6.9）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.162\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eCEA (\u0026mu;g/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e2.77（1.8-5.0）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e3.02（2.0-4.7）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.325\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eCA125(U/mL)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e14.46（8.7-25.4）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e18.96（11.7-34.7）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.901840490797547%\"\u003e\n \u003cp\u003eCA199(U/mL)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.45398773006135%\"\u003e\n \u003cp\u003e170.25（47.8-611.2）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"28.987730061349694%\"\u003e\n \u003cp\u003e229（70.5-738.4）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.656441717791411%\"\u003e\n \u003cp\u003e0.027\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"4\" width=\"100%\"\u003e\n \u003cp\u003e\u0026dagger;\u0026nbsp;Data in parentheses are the interquartile range\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"4\" width=\"100%\"\u003e\n \u003cp\u003eAbbreviation: \u003cem\u003eBMI\u003c/em\u003e, Body Mass Index; \u003cem\u003eALT\u003c/em\u003e, alanine aminotransferase; \u003cem\u003eAST\u003c/em\u003e, aspartate amino transferase; \u003cem\u003eCEA\u003c/em\u003e, Carcinoma Embryonic Antigen; \u003cem\u003eCA 125\u003c/em\u003e, carbohydrate antigen 125; \u003cem\u003eCA 199\u003c/em\u003e, carbohydrate antigen 199\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNo significant differences in clinical characteristics were observed between these two cohorts. Alcohol consumption rates were 24.0% (45/190) and 33.3% (60/180) respectively, with a significant difference between the two cohorts (p=0.04). A total of 21.0% (39 out of 190) of the LG group and 30% (87 out of 180) of the HG group had significant weight loss, with a statistically significant difference between the two groups (p=0.015).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAll PET/CT reports were independently reviewed and assessed by two experienced PET/CT diagnosticians. The median value of SUVmax was 5.75 (2.2-36.0) in the LG group and 7.5 (3.2-31.1) in the HG group, with a statistically significant difference between the two groups (p=0.001). Tumours in the HG group were more likely to be found in the head and neck of the pancreas than in the LG group 67% vs 54% (p=0.01), a statistically significant difference between the two groups. In laboratory tests, total and direct bilirubin levels were lower in the LG group than in the HG group (12.4 vs 15.7, p=0.007; 4.4 vs 4.9, p=0.037), with a statistically significant difference between the two groups. In terms of tumour marker detection, the LG group had lower levels of CA-125 and CA199 than the HG group (14.46 vs 18.96, p=0.004; 170.25 vs 229, p=0.027).\u003c/p\u003e\n\u003ch2\u003ePerformance of lesion segmentation model\u003c/h2\u003e\n\u003cp\u003eThe Dice score for the lesion segmentation Unet model in Table.S1. is 0.72. The Dice score for Unet prediction with guidance of organ location (Unet+OL), that is, with the addition of nnUnet-based organ segmentation, was increased to 0.76. The Dice score for Unet+OL prediction with post-processing (Unet+OLP), which is Unet+OL with the addition of post-processing such as erosion, expansion, and threshold segmentation, was improved to 0.89, which is a significant advantage compared to the Unet and Unet+OL models. As can be seen in Figure 3, in the three validated cases, the region of lesions output by Unet+OLP was closer to the Ground Truth (GT) labeling than that of Unet+OL.\u003c/p\u003e\n\u003ch2\u003ePerformance of PDAC pathological grade classification model\u003c/h2\u003e\n\u003cp\u003eRegarding testing performance, Figure 4 shows that the AUC of the clinical data model was 0.95 in the training cohort, 0.68 in the validation cohort and 0.68 in the test cohort, while the AUC of the PET/CT model was 0.99 in the training cohort, 0.72 in the validation cohort and 0.74 in the test cohort, which was better than the clinical model. In order to improve the efficacy and accuracy of the model, we combined the clinical model with the PET/CT DL model to build a PET/CT + Clinical data model. The AUC of the PET/CT+ Clinical data model reached 0.99, 0.74, and 0.77 in the training, validation, and test cohorts, respectively.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;It can be seen in Table 2 that the accuracy, sensitivity, and specificity of the PET/CT model were 72%, 73%, and 72%, respectively. The accuracy, sensitivity, and specificity of the clinical data model were 66%, 67%, and 66%, respectively. The accuracy, sensitivity, and specificity of PET/CT+ Clinical data model were 75%, 77%, and 73% respectively. The model that integrates both image and clinical data achieved the best performance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable.2 The performance comparison of different models\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellpadding=\"0\" cellspacing=\"0\" width=\"653\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"12.996941896024465%\"\u003e\n \u003cp\u003e\u003cstrong\u003eModels\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCohorts\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.773700305810397%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e\u003cstrong\u003eACC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSENS\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSPEC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPV\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNPV\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" width=\"12.996941896024465%\"\u003e\n \u003cp\u003ePET/CT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003eTrain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.773700305810397%\"\u003e\n \u003cp\u003e0.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.96\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003eVal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.532513181019333%\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003eTest\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.532513181019333%\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" width=\"12.996941896024465%\"\u003e\n \u003cp\u003eClinical\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003eTrain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.773700305810397%\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003eVal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.532513181019333%\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003eTest\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.532513181019333%\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" width=\"12.996941896024465%\"\u003e\n \u003cp\u003ePET/CT+Clinical\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003eTrain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.773700305810397%\"\u003e\n \u003cp\u003e0.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.53822629969419%\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003eVal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.532513181019333%\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003eTest\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.532513181019333%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.77\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.75\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.77\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.73\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.73\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.411247803163445%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.76\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"8\" width=\"100%\"\u003e\n \u003cp\u003eAUC, area under receiver operating characteristic curve; ACC, accuracy; SENS, sensitivity; SPEC, specificity; PPV, positive predictive value; NPV, negative predictive value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Discussion","content":"\u003cp\u003eCurrently, research on deep learning models for PDAC mainly focuses on the disease\u0026apos;s differential diagnosis, preoperative staging, and prognostic analysis. Wei et al.\u003csup\u003e20\u003c/sup\u003e used a combination of machine learning and deep learning algorithms to extract features from PET/CT images to predict the difference between PDAC and autoimmune pancreatitis, developing a multi-domain fusion model with an overall performance of AUC, accuracy, sensitivity, and specificity of 0.96, 0.90, 0.88, and 0.93, respectively. Bian et al.\u003csup\u003e21\u003c/sup\u003e developed and validated an automated preoperative AI algorithm for tumor and lymph node segmentation in CT imaging to predict LN metastasis in PDAC patients. Lee et al.\u003csup\u003e22\u003c/sup\u003e developed a deep learning model based on clinical data to predict postoperative survival in pancreatic cancer patients, and the model\u0026apos;s performance in predicting 2-year OS was comparable to AJCC (AUC, 0.67; P=0.35), and it was better than AJCC in predicting 1-year recurrence free survival (AUC, 0.54; P=0.049). Yao et al.\u003csup\u003e23\u003c/sup\u003e used deep learning to analyze preoperative multi-phase CT and developed imaging-derived biomarkers for predicting overall survival in PDAC, which can be used to predict OS in resectable PDAC patients.\u003c/p\u003e\n\u003cp\u003eResearch has found that the pathological grade of PDAC is largely determined by the fibrous matrix quality in its stroma. Tumors with lower differentiation have more fibrous matrix and occupy more of the contrast agent\u003csup\u003e24\u003c/sup\u003e. This provides a principle for pathological grading of PDAC through imaging studies. For example, Tikhonova et al.\u003csup\u003e25\u003c/sup\u003e used a machine learning algorithm to establish a diagnostic model for image-based PDAC grading based on preoperative CT, using data from 91 patients to establish a diagnostic model. The AUC for pathological grading \u0026ge; 2 (or 3) was 0.75 (or 0.66). Na et al.\u003csup\u003e24\u003c/sup\u003e developed and validated a radiological feature based on contrast-enhanced computed tomography \u0026nbsp;for preoperative prediction of histological grading of PDAC, and the AUC of the final validation set was 0.770.\u003c/p\u003e\n\u003cp\u003eThe above studies are all based on CT. PET/CT, integrating PET and CT into the same device and program, characterizes the lesion from different aspects and provides metabolic information from the former and detailed anatomical information from the latter, which makes the PET/CT image to have both good clarity and a strong ability to distinguish between lesion tissue and normal tissue\u003csup\u003e26\u003c/sup\u003e. Some research has shown that, based on 102 patients with histologically confirmed PDAC, FDG uptake is related to the invasiveness of pancreatic cancer, and SUVmax is significantly related to pathological grading\u003csup\u003e27\u003c/sup\u003e. Therefore, our goal is to unleash the potential of PET/CT in pathological grading of PDAC.\u003c/p\u003e\n\u003cp\u003eIn this study, in order to achieve the prediction of fully automatic PDAC pathological grading and reduce the impact of confounding factors in PET/CT images as much as possible, we first developed a deep learning model for PDAC lesion segmentation. Due to the presence of FDG uptake in organs such as the liver, adrenal gland, small intestine, and bladder in addition to the pancreatic lesion, the model showed abnormal segmentation in the initial training (Figure S4). In order to increase the accuracy of segmentation, nnUnet\u0026apos;s organ segmentation pre-training model was added to provide a rough segmentation of abdominal organs, which are filtered out to enhance the performance of pancreatic lesion segmentation. While nnUnet\u003csup\u003e16\u003c/sup\u003e is an excellent image segmentation model that demonstrated good performance in the field of medical image segmentation especially with a large amount of training images, it is difficult to directly obtain a good segmentation model due to the small number of pancreatic cancer sections in this study. We first trained the nnUnet model of pancreatic organ segmentation using abdominal CT sections of healthy people, preserved the parameters of the model, and applied it to the target cases. The presence of tumors greatly reduced the segmentation results of the nnUnet model, but the preliminary localization of the patient\u0026apos;s pancreas can still be obtained. The addition of nnUnet as a filter increased the Dice score of the model from 0.72 to 0.76. In the post-processing, we added corrosion, expansion, SUVmax 40% threshold segmentation and other post-operations, and finally increased the Dice score to 0.86 (Table.S1.). Through these steps, the segmentation model\u0026apos;s performance was able to achieve an acceptable level.\u003c/p\u003e\n\u003cp\u003eIn building a classification model for PET/CT images, to incorporate more effective information while minimizing the effect of segmentation errors, we cropped out a 3D patch from the raw image centered around the segmented mask. The accuracy of the final PET/CT-based classification model was 72%, sensitivity was 73%, and specificity was 72%.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo further boost the classification performance, we resorted to clinical indicators that are important for revealing pancreatic cancer characteristics. For example, CA199 is closely related to the prognosis of pancreatic cancer patients\u003csup\u003e28,29\u003c/sup\u003e. A study found that CA199 produced by pancreatic cancer cell lines in vitro is associated with histological differentiation in nude mice in vivo\u003csup\u003e30\u003c/sup\u003e. Therefore, we extracted key clinical data indicators such as CA199 and combined them with the PET/CT model to optimize the prediction accuracy. In our early experiment, we used fully connected layers to connect PET/CT with the extracted clinical features, and achieved a minor performance improvement; in some folds in cross validation, the combined model performed worser than the original PET/CT model. Therefore, in order to better integrate the features of PET/CT and clinical data, we used TMC\u003csup\u003e19\u003c/sup\u003e to improve the reliability of classification. This model parameterizes different data and combines them based on Dempster-Shafer theory, which improves the reliability and robustness of the classification model and improves the performance of the model (refer to Supplementary Method 1.2). Finally, to test the generalization ability of the model, we utilized 7-fold cross-validation. The final PET/CT + Clinical data model achieved an accuracy of 75%, sensitivity of 77%, and specificity of 73%. The deep learning model of PET/CT + Clinical data has a significant improvement compared to the traditional EUS-FNB\u003csup\u003e10\u003c/sup\u003e with an accuracy of 56% and sensitivity of 41%.\u003c/p\u003e\n\u003cp\u003eOur research has a few limitations. Firstly, our data is from a single center and lacks external datasets for validation and evaluation. However, because the medical images were acquired using three different PET machines, it increases the generalizability. Additionally, using 7-fold cross-validation ensured the stability of the final model, which partially compensated for the lack of external validation sets. Secondly, all the data included in this study were from surgical patients, so the pathological differentiation was concentrated; however, the data for patients with excessive malignancy was relatively scarce, which did not make the model to show more significant discriminability. Finally, this study only explored the relationship between imaging features and pathological differentiation of PDAC but did not investigate the survival outcome of the patients, which is more concerned by patients and surgeons. Further research is needed to study the survival of patients.\u003c/p\u003e\n\u003cp\u003eOur developed deep learning model can automatically analyze PET/CT images without human intervention, reducing subjective errors and improving the accuracy and reliability of grading. It is important to note that deep learning models cannot completely replace professional knowledge of imaging and pathology. Before using deep learning models to predict pathological grading, professional knowledge needs to be combined for interpretation and evaluation to ensure the accuracy and reliability of the results. In conclusion, deep learning-based PET/CT has a great potential in pathological grading of pancreatic cancer, and more clinical studies are needed to prove its safety and effectiveness before deep learning models can replace traditional pathological staging methods.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eTo the best our knowledge, this is the first report of using a DL model for preoperative prediction of PDAC pathological grading using PET/CT. The model\u0026apos;s predictive performance was improved by combining features of PET/CT and key clinical data.\u003c/p\u003e\n"},{"header":"Abbreviations","content":"\u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003ePDAC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003ePancreatic ductal adenocarcinoma\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eDL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eDeep learning\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eAUC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eArea under curve\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eSUV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eStandardized uptake value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eLNM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eLymph node metastasis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eLG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eLow grade\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eHG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eHigh grade\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eFC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eFully connected\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eTMC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eTrusted Multi-view Classification\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eROC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eReceiver operator characteristic curve\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eGT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eGround truth\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eOL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eOrgan location\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eOLP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eOrgan location with post-processing\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eACC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eAccuracy\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eSENS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eSensitivity\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eSPEC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eSpecificity\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003ePPV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003ePositive predictive value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"17.05426356589147%\"\u003e\n \u003cp\u003eNPV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"82.94573643410853%\"\u003e\n \u003cp\u003eNegative predictive value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eSupplementary Information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdditional file 1: Fig.S1.\u0026nbsp;\u003c/strong\u003eUnet image segmentation network. \u003cstrong\u003eFig.S2.\u003c/strong\u003e Distribution data of pathological differentiation degree in medical records. \u003cstrong\u003eFig.S3.\u0026nbsp;\u003c/strong\u003eRandom forest analysis.\u003cstrong\u003e\u0026nbsp;Fig.S4.\u003c/strong\u003e Examples of segmentation model before and after adding nnUet. \u003cstrong\u003eTable.S1.\u003c/strong\u003e The performance comparison of segmentation process.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors were involved in the study conception and design. GZ, YZL, ZZW, LD, FW, BXX were involved in acquisition of data. GZ, CKB, YZ, S. KZ participated in the development and testing of deep learning models. GZ, CKB, ZZW, YZ were involved in analysis and interpretation of the data. GZ, CKB and S. KZ were involved in drafting of the manuscript. RL is responsible for review, editing and supervision. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Key R\u0026amp;D Program of China (2021ZD0113301)；\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe relevant images and clinical data from this study are not available because they contain private patient information. However, such data can be obtained through agency approval and signed data use agreements and/or signed material transfer agreements.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eApproved by the Ethics Review Committee of the local institution of the PLA General Hospital, and the requirement of written informed consent was waived. All procedures involving human participants in this study comply with institutional and/or national Research Council ethical standards as well as the 1964 Declaration of Helsinki and its later amendments or similar ethical standards.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSiegel RL, Miller KD, Jemal A. Cancer statistics, 2018. \u003cem\u003eCA: a cancer journal for clinicians. \u003c/em\u003e2018;68(1):7-30.\u003c/li\u003e\n\u003cli\u003eHartwig W, Werner J, J\u0026auml;ger D, Debus J, B\u0026uuml;chler MW. Improvement of surgical results for pancreatic cancer. \u003cem\u003eThe Lancet Oncology. \u003c/em\u003e2013;14(11):e476-e485.\u003c/li\u003e\n\u003cli\u003eStrobel O, Neoptolemos J, J\u0026auml;ger D, B\u0026uuml;chler MW. Optimizing the outcomes of pancreatic cancer surgery. \u003cem\u003eNature reviews Clinical oncology. \u003c/em\u003e2019;16(1):11-26.\u003c/li\u003e\n\u003cli\u003eGolan T, Sella T, Margalit O, et al. Short- and Long-Term Survival in Metastatic Pancreatic Adenocarcinoma, 1993-2013. \u003cem\u003eJournal of the National Comprehensive Cancer Network : JNCCN. \u003c/em\u003e2017;15(8):1022-1027.\u003c/li\u003e\n\u003cli\u003eHan SH, Heo JS, Choi SH, et al. Actual long-term outcome of T1 and T2 pancreatic ductal adenocarcinoma after surgical resection. \u003cem\u003eInternational journal of surgery (London, England). \u003c/em\u003e2017;40:68-72.\u003c/li\u003e\n\u003cli\u003ede Geus SW, Eskander MF, Bliss LA, et al. Neoadjuvant therapy versus upfront surgery for resected pancreatic adenocarcinoma: A nationwide propensity score matched analysis. \u003cem\u003eSurgery. \u003c/em\u003e2017;161(3):592-601.\u003c/li\u003e\n\u003cli\u003eShubert CR, Bergquist JR, Groeschl RT, et al. Overall survival is increased among stage III pancreatic adenocarcinoma patients receiving neoadjuvant chemotherapy compared to surgery first and adjuvant chemotherapy: An intention to treat analysis of the National Cancer Database. \u003cem\u003eSurgery. \u003c/em\u003e2016;160(4):1080-1096.\u003c/li\u003e\n\u003cli\u003eVersteijne E, Vogel JA, Besselink MG, et al. Meta-analysis comparing upfront surgery with neoadjuvant treatment in patients with resectable or borderline resectable pancreatic cancer. \u003cem\u003eThe British journal of surgery. \u003c/em\u003e2018;105(8):946-958.\u003c/li\u003e\n\u003cli\u003eGerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. \u003cem\u003eThe New England journal of medicine. \u003c/em\u003e2012;366(10):883-892.\u003c/li\u003e\n\u003cli\u003eLarghi A, Correale L, Ricci R, et al. Interobserver agreement and accuracy of preoperative endoscopic ultrasound-guided biopsy for histological grading of pancreatic cancer. \u003cem\u003eEndoscopy. \u003c/em\u003e2015;47(4):308-314.\u003c/li\u003e\n\u003cli\u003eElemento O, Leslie C, Lundin J, Tourassi G. Artificial intelligence in cancer research, diagnosis and therapy. \u003cem\u003eNature reviews Cancer. \u003c/em\u003e2021;21(12):747-752.\u003c/li\u003e\n\u003cli\u003eKleppe A, Skrede OJ, De Raedt S, Liest\u0026oslash;l K, Kerr DJ, Danielsen HE. Designing deep learning studies in cancer diagnostics. \u003cem\u003eNature reviews Cancer. \u003c/em\u003e2021;21(3):199-211.\u003c/li\u003e\n\u003cli\u003eChoi H. Deep Learning in Nuclear Medicine and Molecular Imaging: Current Perspectives and Future Directions. \u003cem\u003eNuclear medicine and molecular imaging. \u003c/em\u003e2018;52(2):109-118.\u003c/li\u003e\n\u003cli\u003eWang S, Mahon R, Weiss E, et al. Automated Lung Cancer Segmentation Using a PET and CT Dual-Modality Deep Learning Neural Network. \u003cem\u003eInternational journal of radiation oncology, biology, physics. \u003c/em\u003e2023;115(2):529-539.\u003c/li\u003e\n\u003cli\u003eAn C, Li D, Li S, et al. Deep learning radiomics of dual-energy computed tomography for predicting lymph node metastases of pancreatic ductal adenocarcinoma. \u003cem\u003eEuropean journal of nuclear medicine and molecular imaging. \u003c/em\u003e2022;49(4):1187-1199.\u003c/li\u003e\n\u003cli\u003eIsensee F, Petersen J, Klein A, et al. nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation. 2018.\u003c/li\u003e\n\u003cli\u003eWasif N, Ko CY, Farrell J, et al. Impact of tumor grade on prognosis in pancreatic cancer: should we include grade in AJCC staging? \u003cem\u003eAnnals of surgical oncology. \u003c/em\u003e2010;17(9):2312-2320.\u003c/li\u003e\n\u003cli\u003eRochefort MM, Ankeny JS, Kadera BE, et al. Impact of tumor grade on pancreatic cancer prognosis: validation of a novel TNMG staging system. \u003cem\u003eAnnals of surgical oncology. \u003c/em\u003e2013;20(13):4322-4329.\u003c/li\u003e\n\u003cli\u003eHan Z, Zhang C, Fu H, Zhou JT. Trusted Multi-View Classification with Dynamic Evidential Fusion. \u003cem\u003eIEEE transactions on pattern analysis and machine intelligence. \u003c/em\u003e2022;Pp.\u003c/li\u003e\n\u003cli\u003eWei W, Jia G, Wu Z, et al. A multidomain fusion model of radiomics and deep learning to discriminate between PDAC and AIP based on (18)F-FDG PET/CT images. \u003cem\u003eJapanese journal of radiology. \u003c/em\u003e2022:1-11.\u003c/li\u003e\n\u003cli\u003eBian Y, Zheng Z, Fang X, et al. Artificial Intelligence to Predict Lymph Node Metastasis at CT in Pancreatic Ductal Adenocarcinoma. \u003cem\u003eRadiology. \u003c/em\u003e2023;306(1):160-169.\u003c/li\u003e\n\u003cli\u003eLee W, Park HJ, Lee HJ, et al. Preoperative data-based deep learning model for predicting postoperative survival in pancreatic cancer patients. \u003cem\u003eInternational journal of surgery (London, England). \u003c/em\u003e2022;105:106851.\u003c/li\u003e\n\u003cli\u003eYao J, Cao K, Hou Y, et al. Deep Learning for Fully Automated Prediction of Overall Survival in Patients Undergoing Resection for Pancreatic Cancer: A Retrospective Multicenter Study. \u003cem\u003eAnnals of surgery. \u003c/em\u003e2022.\u003c/li\u003e\n\u003cli\u003eChang N, Cui L, Luo Y, Chang Z, Yu B, Liu Z. Development and multicenter validation of a CT-based radiomics signature for discriminating histological grades of pancreatic ductal adenocarcinoma. \u003cem\u003eQuantitative imaging in medicine and surgery. \u003c/em\u003e2020;10(3):692-702.\u003c/li\u003e\n\u003cli\u003eTikhonova VS, Karmazanovsky GG, Kondratyev EV, et al. Radiomics model-based algorithm for preoperative prediction of pancreatic ductal adenocarcinoma grade. \u003cem\u003eEuropean radiology. \u003c/em\u003e2022.\u003c/li\u003e\n\u003cli\u003eBeyer T, Antoch G, M\u0026uuml;ller S, et al. Acquisition protocol considerations for combined PET/CT imaging. \u003cem\u003eJournal of nuclear medicine : official publication, Society of Nuclear Medicine. \u003c/em\u003e2004;45 Suppl 1:25s-35s.\u003c/li\u003e\n\u003cli\u003eWang Z, Chen JQ, Liu JL, Qin XG, Huang Y. FDG-PET in diagnosis, staging and prognosis of pancreatic carcinoma: a meta-analysis. \u003cem\u003eWorld journal of gastroenterology. \u003c/em\u003e2013;19(29):4808-4817.\u003c/li\u003e\n\u003cli\u003eWilliams JL, Kadera BE, Nguyen AH, et al. CA19-9 Normalization During Pre-operative Treatment Predicts Longer Survival for Patients with Locally Progressed Pancreatic Cancer. \u003cem\u003eJournal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract. \u003c/em\u003e2016;20(7):1331-1342.\u003c/li\u003e\n\u003cli\u003eChen Y, Shao Z, Chen W, et al. A varying-coefficient cox model for the effect of CA19-9 kinetics on overall survival in patients with advanced pancreatic cancer. \u003cem\u003eOncotarget. \u003c/em\u003e2017;8(18):29925-29934.\u003c/li\u003e\n\u003cli\u003eIwamura T, Taniguchi S, Kitamura N, et al. Correlation between CA19-9 production in vitro and histological grades of differentiation in vivo in clones isolated from a human pancreatic cancer cell line (SUIT-2). \u003cem\u003eJournal of gastroenterology and hepatology. \u003c/em\u003e1992;7(5):512-519.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"ejnmmi-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ejre","sideBox":"Learn more about [EJNMMI Research](http://ejnmmires.springeropen.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/ejre/default.aspx","title":"EJNMMI Research","twitterHandle":"@officialEANM","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"deep learning, pancreatic cancer, PET/CT, pathological grading, prediction model","lastPublishedDoi":"10.21203/rs.3.rs-2578400/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2578400/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBackground ：The determination of pathological grading has a guiding significance for the treatment of pancreatic ductal adenocarcinoma（PDAC）patients. However, there is a lack of an accurate and safe method to obtain pathological grading before surgery. The aim of this study is to develop a deep learning（DL）model based on 18F-FDG-PET/CT for a fully automatic prediction of preoperative pathological grading of pancreatic cancer. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eResults ：A total of 370 PDAC patients from January 2016 to September 2021 were collected retrospectively. All patients underwent 18F-FDG-PET/CT examination before surgery and obtained pathological results after surgery. A DL model for pancreatic cancer lesion segmentation was first developed using 100 of these cases and applied to the remaining cases to obtain lesion regions. After that, all patients were divided into training set, validation set and test set according to the ratio of 5:1:1. A predictive model of pancreatic cancer pathological grade was developed using the features computed from the lesion regions obtained by the lesion segmentation model and key clinical characteristics of the patients. Finally, the stability of the model was verified by 7-fold cross-validation.The Dice score of the developed PET/CT based tumour segmentation model for PDAC was 0.89. The area under curve (AUC) of the PET/CT-based DL model developed on the basis of the segmentation model was 0.74, with an accuracy, sensitivity, and specificity of 0.72, 0.73, and 0.72, respectively. After integrating key clinical data, the AUC of the model improved to 0.77, with its accuracy, sensitivity, and specificity boosted to 0.75, 0.77, and 0.73, respectively. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConclusion ：To the best of our knowledge, this is the first deep learning model to end-to-end predict the pathological grading of PDAC in a fully automatic manner, which is expected to improve clinical decision making.\u003c/p\u003e","manuscriptTitle":"18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-02-20 18:21:47","doi":"10.21203/rs.3.rs-2578400/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2023-02-17T05:47:41+00:00","index":0,"fulltext":""},{"type":"editorAssigned","content":"","date":"2023-02-14T09:13:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"EJNMMI Research","date":"2023-02-14T02:51:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"ejnmmi-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ejre","sideBox":"Learn more about [EJNMMI Research](http://ejnmmires.springeropen.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/ejre/default.aspx","title":"EJNMMI Research","twitterHandle":"@officialEANM","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3b0512d0-6d27-489e-b701-96b120c0a356","owner":[],"postedDate":"February 20th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2023-10-16T21:07:12+00:00","versionOfRecord":{"articleIdentity":"rs-2578400","link":"https://doi.org/10.1186/s13550-023-00985-4","journal":{"identity":"ejnmmi-research","isVorOnly":false,"title":"EJNMMI Research"},"publishedOn":"2023-05-25 20:58:58","publishedOnDateReadable":"May 25th, 2023"},"versionCreatedAt":"2023-02-20 18:21:47","video":"","vorDoi":"10.1186/s13550-023-00985-4","vorDoiUrl":"https://doi.org/10.1186/s13550-023-00985-4","workflowStages":[]},"version":"v1","identity":"rs-2578400","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2578400","identity":"rs-2578400","version":["v1"]},"buildId":"-HB7Z8yhvgn0wM9Nzuekk","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}