Background. ¹⁸F-FDG PET/CT is widely used in the prognosis evaluation of tumor patients. The radiomics features can provide additional information for clinical prognostic assessment. Purpose. Purpose is to explore the prognostic value of radiomics features from dual-time ¹⁸F-FDG PET/CT images for locally advanced pancreatic cancer (LAPC) patients treated with stereotactic body radiation therapy (SBRT). Materials and Methods. This retrospective study included 70 LAPC patients who received early and delayed ¹⁸F-FDG PET/CT scans before SBRT treatment. A total of 1188 quantitative imaging features were extracted from dual-time PET/CT images. To avoid overfitting, the univariate analysis and elastic net were used to obtain a sparse set of image features that were applied to develop a radiomics score (Rad-score). Then, the Harrell consistency index (C-index) was used to evaluate the prognosis model. Results. The Rad-score from dual-time images contains six features, including intensity histogram, morphological, and texture features. In the validation cohort, the univariate analysis showed that the Rad-score was the independent prognostic factor ( , hazard ratio [HR]: 3.2). And in the multivariate analysis, the Rad-score was the only prognostic factor ( , HR: 4.1) that was significantly associated with the overall survival (OS) of patients. In addition, according to cross-validation, the C-index of the prognosis model based on the Rad-score from dual-time images is better than the early and delayed images (0.720 vs. 0.683 vs. 0.583). Conclusion. The Rad-score based on dual-time ¹⁸F-FDG PET/CT images is a promising noninvasive method with better prognostic value.

Accurate, automated medical image detection is critical in clinical diagnosis and analysis. Since the pancreatic lesions in CT images are similar to the surrounding tissues, they are difficult to be detected. The lesions in PET images have the disadvantage of blurred edges that the precise localization is slightly insufficient. PET/CT integrates functional and anatomical imaging, combining the advantages of the high contrast of PET images and the high spatial resolution of CT images to assist doctors in detecting lesions. Therefore, it is significant for us to study the object detection of lesions based on PET/CT. At the same time, the context information extraction ability of the basic framework Faster R-CNN is insufficient. Therefore, we propose a multi-scale adaptive attention feature fusion network (MAFF) based on PET/CT to realize the automatic detection of pancreatic lesions. First, we fuse multi-scale features through a feature pyramid module to obtain richer contextual information and add an attention module to achieve preliminary screening of input features. Second, we design an adaptive attention feature fusion network to make feature semantic information selection more focused by recalibrating multimodal feature maps. Finally, we adopt a pooling module, which not only solves the problem of different sizes of region proposals but also avoids the localization error caused by quantization. Experimental results show that our proposed multimodal algorithm outperforms other algorithms on two challenging tasks, pancreatic cancer lesion detection, and head and neck cancer lesion detection.

Background: 18F-FDG PET/CT is widely used in the prognosis evaluation of tumor patients. The radiomics features can provide additional information for clinical prognostic assessment. Purpose: To explore the prognostic value of radiomics features from dual time ¹⁸F-FDG PET/CT images for locally advanced pancreatic cancer (LAPC) patients treated with stereotactic body radiation therapy (SBRT). Materials and Methods: This retrospective study included 70 LAPC patients who received early and delayed ¹⁸F-FDG PET/CT scans before SBRT treatment. A total of 1188 quantitative imaging features were extracted from dual time PET/CT images. To avoid over-fitting, the univariate analysis and elastic net were used to obtain a sparse set of image features, that were applied to develop a radiomics score (Rad-score). Then the Harrell consistency index (C-index) was used to evaluate the prognosis model. Results: The Rad-score from dual time images contains six features, including intensity histogram, morphological, and texture features. In the validation cohort, the univariate analysis showed that the Rad-score was the independent prognostic factor (p<0.001, hazard ratio [HR]: 3.2). And in the multivariate analysis, the Rad-score was the only prognostic factor (p < 0.01, HR: 4.1) that was significantly associated with the OS of patients. In addition, according to cross-validation, the C-index of the prognosis model based on the Rad-score from dual-time images is better than the early and delay images (0.720 vs 0.683 vs 0.583). Conclusion: The Rad-score based on dual-time 18F-FDG PET/CT images is a promising non-invasive method with better prognostic value.