OBJECTIVE: The objective of this study was to preliminarily assess the ability of metabolic parameters and radiomics derived from 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) to distinguish mass-forming pancreatic lymphoma from pancreatic carcinoma using machine learning.
METHODS: A total of 88 lesions from 86 patients diagnosed as mass-forming pancreatic lymphoma or pancreatic carcinoma were included and randomly divided into a training set and a validation set at a 4-to-1 ratio. The segmentation of regions of interest was performed using ITK-SNAP software, PET metabolic parameters and radiomics features were extracted using 3Dslicer and PYTHON. Following the selection of optimal metabolic parameters and radiomics features, Logistic regression (LR), support vector machine (SVM), and random forest (RF) models were constructed for PET metabolic parameters, CT radiomics, PET radiomics, and PET/CT radiomics. Model performance was assessed in terms of area under the curve (AUC), accuracy, sensitivity, and specificity in both the training and validation sets.
RESULTS: Strong discriminative ability observed in all models, with AUC values ranging from 0.727 to 0.978. The highest performance exhibited by the combined PET and CT radiomics features. AUC values for PET/CT radiomics models in the training set were LR 0.994, SVM 0.994, RF 0.989. In the validation set, AUC values were LR 0.909, SVM 0.883, RF 0.844.
CONCLUSIONS: Machine learning models utilizing the metabolic parameters and radiomics of 18F-FDG PET/CT show promise in distinguishing between pancreatic carcinoma and mass-forming pancreatic lymphoma. Further validation on a larger cohort is necessary before practical implementation in clinical settings.

OBJECTIVE: To improve the diagnostic efficacy of 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) for Chinese patients with fever of unknown origin (FUO) and inflammation of unknown origin (IUO), with combined clinical parameters.
METHODS: FUO/IUO patients who underwent a standard diagnostic work-up and 18F-FDG PET/CT scanning were enrolled and divided into a local uptake lesion subgroup and a non-specific abnormal uptake subgroup. Beneficial clinical parameters for improving the diagnostic efficacy of PET/CT were identified.
RESULTS: From January 2014 to January 2019, 253 FUO/IUO patients were studied. In total, 147 patients had local uptake lesions and 106 patients had non-specific abnormal uptake. In the local uptake lesion group, the positioning accuracy of PET/CT was 37.2% in grades 1 and 2, and 66.3% in grades 3 and 4. With the following combination of clinical parameters, the positioning accuracy increased to 75.0% and 90.0%, respectively: time from admission to performing PET/CT scanning <6.5 days and C-reactive protein level >95 mg/l. In the non-specific abnormal uptake group, the combination of sex (male), bicytopenia, and lactic dehydrogenase improved the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for diagnosing malignancy from 64.3%, 69%, 60%, and 72.7%, respectively, to 83.3%, 81%, 81.4%, and 82.9%, respectively. With the combination of sex (male), white blood count, serum ferritin level, and hepatosplenomegaly, the infection prediction model had a sensitivity, specificity, PPV, and NPV of 78%, 76.2%, 76.6%, and 77.6%, respectively.
CONCLUSIONS: Combined clinical parameters improved the localization diagnostic value of 18F-FDG PET/CT in the local uptake lesion subgroup and the etiological diagnostic value in the non-specific abnormal uptake subgroup.