A novel deep-learning-based approach for automatic reorientation of 3D cardiac SPECT images.

PURPOSE: Reconstructed transaxial cardiac SPECT images need to be reoriented into standard short-axis slices for subsequent accurate processing and analysis. We proposed a novel deep-learning-based method for fully automatic reorientation of cardiac SPECT images and evaluated its performance on data from two clinical centers. METHODS: We used a convolutional neural network to predict the 6 rigid-body transformation parameters and a spatial transformation network was then implemented to apply these parameters on the input images for image reorientation. A novel compound loss function which balanced the parametric similarity and penalized discrepancy of the prediction and training dataset was utilized in the training stage. Data from a set of 322 patients underwent data augmentation to 6440 groups of images for the network training, and a dataset of 52 patients from the same center and 23 patients from another center were used for evaluation. Similarity of the 6 parameters was analyzed between the proposed and the manual methods. Polar maps were generated from the output images and the averaged count values of the 17 segments were computed from polar maps to evaluate the quantitative accuracy of the proposed method. RESULTS: All the testing patients achieved automatic reorientation successfully. Linear regression results showed the 6 predicted rigid parameters and the average count value of the 17 segments having good agreement with the reference manual method. No significant difference by paired t-test was noticed between the rigid parameters of our method and the manual method (p > 0.05). Average count values of the 17 segments show a smaller difference of the proposed and manual methods than those between the existing and manual methods. CONCLUSION: The results strongly indicate the feasibility of our method in accurate automatic cardiac SPECT reorientation. This deep-learning-based reorientation method has great promise for clinical application and warrants further investigation.

68Ga-FAPI and 18F-PET/CT Images in Intestinal Tuberculosis.

ABSTRACT: A 28-year-old woman presented with abdominal pain, bowel dysfunction, and weight loss for 3 months. 18F-FDG PET/CT revealed multiple hypermetabolic lesions in the intestines and peritoneal thickening/caking with moderate FDG activity. 68Ga-FAPI PET/CT showed intense FAPI uptake in the aforementioned FDG-avid lesions and a larger number of abnormal foci with intense FAPI uptake in the peritoneum than that shown in 18F-FDG images. Endoscopy-guided biopsy from the colonic mucosa was consistent with tuberculosis. The positive findings of 68Ga-FAPI in the current case highlighted that 68Ga-FAPI may have value in the evaluation of intestinal tuberculosis.

[68Ga]Ga-FAPI PET/CT Improves the T Staging of Patients with Newly Diagnosed Nasopharyngeal Carcinoma: A Comparison with [18F]F-FDG.

PURPOSE: This study aimed to explore the value of [68Ga]Ga-labelled fibroblast activation protein inhibitor ([68Ga]Ga-FAPI) positron emission tomography/computed tomography (PET/CT) in the initial staging of patients with newly diagnosed nasopharyngeal carcinoma (NPC), compared with 2-deoxy-2[18F]fluoro-D-glucose ([18F]F-FDG) PET/CT. MATERIALS AND METHODS: Forty-seven treatment-naïve patients with newly diagnosed NPC underwent magnetic resonance imaging (MRI), [68Ga]Ga-DOTA-FAPI-04 PET/CT and [18F]F-FDG PET/CT within 1 week. The diagnostic efficiency of all imaging modalities for evaluating primary tumour extension was compared from the two aspects of soft tissue and bony structure involvement. The accuracy of two PET/CT methods for diagnosing cervical lymph node (CLN) metastases was compared, and MRI served as the standard reference. T and N stages were assessed by MRI, [68Ga]Ga-FAPI PET/CT and [18F]F-FDG PET/CT. Immunohistochemical (IHC) staining for FAP was conducted in 22 of the patients. RESULTS: [68Ga]Ga-FAPI PET/CT outperformed [18F]F-FDG PET/CT in the assessment of primary tumour invasion in the cavernous sinus (10 vs. 1, p < 0.001) and bony structures (207 vs. 177, p < 0.001). Compared with MRI, [68Ga]Ga-FAPI PET/CT upgraded and underestimated T stage in 13 and 2 patients, while [18F]F-FDG PET/CT upgraded and underestimated T stage in 5 and 13 patients. However, [68Ga]Ga-FAPI PET/CT was inferior to [18F]F-FDG PET/CT in diagnosing positive CLNs based on the analyses of patients, neck sides, neck levels and individual nodes. [68Ga]Ga-FAPI PET/CT changed therapeutic schedules in 8 patients because of stage group changes. The presence of FAP with high quantity and intensity in cancer-associated fibroblasts (CAFs) was confirmed in all tumour specimens. CONCLUSION: [68Ga]Ga-FAPI PET/CT outperformed [18F]F-FDG PET/CT in detecting the cavernous sinus and bony structure involvement of primary NPC tumours, suggesting its value in improving T staging and therapeutic regimen selection. However, the performance of [68Ga]Ga-FAPI PET/CT is less promising for N staging because it detected fewer positive CLNs than [18F]F-FDG PET/CT.

Characterization of the benign lesions with increased 68Ga-FAPI-04 uptake in PET/CT.

OBJECTIVES: The objective of the study was to characterize benign lesions showing increased 68Ga-FAPI-04 uptake on FAPI PET/CT. METHODS: We retrospectively reviewed 182 patients with suspected various cancers who were performed 68Ga-FAPI-04 PET/CT imaging from August 2020 to December 2020. The diagnoses of the benign lesions were made by the CT findings (CT), other imaging information (OII) (contrast enhance CT, FDG PET, ultrasound, MRI or others), clinical information (CI) (medical history, laboratory examination, symptom, physical sign and follow-up information) or histological biopsy (HB). RESULTS: A total of 185 primary malignant tumors were detected by FAPI PET/CT with the median SUVmax of 9.0 (range from 0.97 to 25.71). There were 360 benign lesions with increased FAPI uptake were detected in 146 (146/182, 80.2%) patients with the median SUVmax of 3.64 (range from 1.39 to 21.56), including inflammatory processes (n = 231, 64.2%), exostosis (n = 54, 15%), hemorrhoid (n = 47, 13.1%), fracture (n = 17, 4.7%), hepatic fibrosis (n = 4, 1.1%), and others (n = 7, 1.9%). CONCLUSION: Benign lesions with increased 68Ga-FAPI-04 uptake are common. The overall SUVmax of benign lesions is lower than that of malignant tumors, however there is a large overlap of SUVmax range. Similar to FDG PET, some benign lesions can be easily diagnosed by combining CT findings, special location and clinical data, but there are still some lesions that may be confused with malignant lesions, which need to be paid more attention. TRAIL REGISTRATION: NIH ClinicalTrials.gov (NCT04499365).

Widespread Skin Infiltration of Leukemia Cutis on 18F-FDG PET/CT.

A 67-year-old man experienced multiple cutaneous nodules for 3 weeks. His initial skin biopsy only showed local inflammation. However, after a 2-week anti-inflammation therapy, the skin manifestations were not improved while the leukocyte count gradually increased. FDG PET/CT revealed widespread cutaneous and subcutaneous hypermetabolic lesions in many parts of the body, and heterogeneously increased uptakes in the bone marrow. Bone marrow biopsy and second skin biopsy were performed. Pathological examination from both specimen demonstrated acute monocytic leukemia.