Search

Background: Rising prostate-specific antigen (PSA) levels following radical prostatectomy are indicative of a poor prognosis, which may associate with periprostatic adipose tissue (PPAT). Accordingly, we aimed to construct a dynamic online nomogram to predict tumor short-term prognosis based on 18F-PSMA-1007 PET/CT of PPAT. Methods: Data from 268 prostate cancer (PCa) patients who underwent 18F-PSMA-1007 PET/CT before prostatectomy were analyzed retrospectively for model construction and validation (training cohort: n = 156; internal validation cohort: n = 65; external validation cohort: n = 47). Radiomics features (RFs) from PET and CT were extracted. Then, the Rad-score was constructed using logistic regression analysis based on the 25 optimal RFs selected through maximal relevance and minimal redundancy, as well as the least absolute shrinkage and selection operator. A nomogram was constructed to predict short-term prognosis which determined by persistent PSA. Results: The Rad-score consisting of 25 RFs showed good discrimination for classifying persistent PSA in all cohorts (all P < 0.05). Based on the logistic analysis, the radiomics-clinical combined model, which contained the optimal RFs and the predictive clinical variables, demonstrated optimal performance at an AUC of 0.85 (95% CI: 0.78—0.91), 0.77 (95% CI: 0.62—0.91) and 0.84 (95% CI: 0.70—0.93) in the training, internal validation and external validation cohorts. In all cohorts, the calibration curve was well-calibrated. Analysis of decision curves revealed greater clinical utility for the radiomics-clinical combined nomogram. Conclusion: The radiomics-clinical combined nomogram serves as a novel tool for preoperative individualized prediction of short-term prognosis among PCa patients. [ABSTRACT FROM AUTHOR]

Purpose: Biochemical recurrence (BCR) following radical prostatectomy (RP) is a significant concern for patients with prostate cancer. Reliable prediction models are needed to identify patients at risk for BCR and facilitate appropriate management. This study aimed to develop and validate a clinical-radiomics model based on preoperative [18 F]PSMA-1007 PET for predicting BCR-free survival (BRFS) in patients who underwent RP for prostate cancer. Materials and methods: A total of 236 patients with histologically confirmed prostate cancer who underwent RP were retrospectively analyzed. All patients had a preoperative [18 F]PSMA-1007 PET/CT scan. Radiomics features were extracted from the primary tumor region on PET images. A radiomics signature was developed using the least absolute shrinkage and selection operator (LASSO) Cox regression model. The performance of the radiomics signature in predicting BRFS was assessed using Harrell's concordance index (C-index). The clinical-radiomics nomogram was constructed using the radiomics signature and clinical features. The model was externally validated in an independent cohort of 98 patients. Results: The radiomics signature comprised three features and demonstrated a C-index of 0.76 (95% CI: 0.60–0.91) in the training cohort and 0.71 (95% CI: 0.63–0.79) in the validation cohort. The radiomics signature remained an independent predictor of BRFS in multivariable analysis (HR: 2.48, 95% CI: 1.47–4.17, p < 0.001). The clinical-radiomics nomogram significantly improved the prediction performance (C-index: 0.81, 95% CI: 0.66–0.95, p = 0.007) in the training cohort and (C-index: 0.78 95% CI: 0.63–0.89, p < 0.001) in the validation cohort. Conclusion: We developed and validated a novel [18 F]PSMA-1007 PET-based clinical-radiomics model that can predict BRFS following RP in prostate cancer patients. This model may be useful in identifying patients with a higher risk of BCR, thus enabling personalized risk stratification and tailored management strategies. [ABSTRACT FROM AUTHOR]

This document is a correction notice for an article titled "Development and validation of [18 F]-PSMA-1007 PET-based radiomics model to predict biochemical recurrence-free survival following radical prostatectomy" published in the European Journal of Nuclear Medicine & Molecular Imaging. The correction addresses an error in Figure 7 of the original article. The incorrect version of the figure is shown, followed by the correct version. The correction has been made by the authors. The publisher, Springer Nature, remains neutral in terms of jurisdictional claims and institutional affiliations. The authors of the article are Tiancheng Li, Mimi Xu, Shuye Yang, Guolin Wang, Yinuo Liu, Kaifeng Liu, Kui Zhao, and Xinhui Su. [Extracted from the article]

Objectives: Gallium-68 (68Ga)–labeled somatostatin analog (SSA) PET imaging has been widely used in clinical practice of neuroendocrine neoplasms (NENs). Compared with 68Ga, 18F has a great practical and economic advantage. Although a few studies have shown the characteristics of [18F] AlF-NOTA-octreotide ([18F]-OC) in healthy volunteers and small NEN patient groups, its clinical value needs further investigation. Herein, this retrospective study aimed to evaluate the diagnostic accuracy of [18F]-OC PET/CT in detecting NENs, as well as to compare it with contrast-enhanced CT/MRI. Methods: We retrospectively reviewed the data of 93 patients who had undergone [18F]-OC PET/CT and CT or MRI scans. Of these patients, there were 45 patients with suspected NENs for diagnostic evaluation, and 48 patients with pathologically confirmed NENs for detecting metastasis or recurrence. [18F]-OC PET/CT images were evaluated visually and semi-quantitatively by measuring maximum standardized uptake value of tumor (SUVmax), tumor-to-background SUVmax ratio (TBR), and SUVmax of hypophysis (SUVhypophysis). A total of 276 suspected NEN lesions were found in these 93 patients. The results of histopathology or radiographic follow-up served as the reference standard for the final diagnosis. Results: Forty-five patients with suspected NENs were confirmed by histopathological examination via resection or biopsy. [18F]-OC PET/CT showed high radiotracer uptake in the lesions of G1-G3 NENs. [18F]-OC PET/CT showed superior performance with 96.3% of sensitivity, 77.8% of specificity, and 88.9% of accuracy in diagnosing NENs compared to CT/MRI. When cutoffs of SUVmax, TBR, and SUVhypophysis were 8.3, 3.1, and 15.4, [18F]-OC PET/CT had the best equilibrium between sensitivity and specificity for differentiating NEN from non-NEN lesions. For a total of 276 suspected NEN lesions, the sensitivity, specificity, and accuracy of [18F]-OC PET/CT for diagnosis of NENs were 90.5%, 82.1%, and 88.8%, respectively, and were higher than those of CT and MRI. G1 and G2 NENs had higher TBR and lower CT enhancement intensity than G3. The SUVmax and TBR had a positive correlation with CT enhancement intensity in G2 rather than in G1 or G3. Conclusions: [18F]-OC PET/CT is a promising imaging modality for initial diagnosis and detecting metastasis or postoperative recurrence in NENs. [ABSTRACT FROM AUTHOR]

Background: Idiopathic inflammatory myopathy (IIM) is a group of autoimmune diseases that frequently leads to unfavorable outcome. This study aimed at identifying the clinical value of PET/CT scan in predicting the outcome of adult IIM patients. Methods: Adult IIM patients who were admitted to the four divisions of the First Affiliated Hospital, Zhejiang University School of Medicine (FAHZJU), from January 1, 2017, to December 31, 2020, were retrospectively reviewed. PET/CT scan and other factors of IIM patients who met the inclusion and exclusion criteria were collected and analyzed. Results: A total of 69 adult IIM patients were finally enrolled into this study. Thirty cases (43.5%) of all the patients enrolled died in follow-up, and the medium follow-up time was 11.90 (4.00, 23.80) months. In particular, 14 patients died within 3 months. The univariate Cox proportional hazards regression analyses revealed pulmonary bacterial infection (P < 0.001), rapidly progressive interstitial lung disease (RP-ILD, P = 0.018), maximum standard uptake value of spleen (spleen SUVmax, P = 0.002), and positivity of anti-MDA5 antibody (P = 0.041) were significantly related to survival in follow-up. The following multivariate Cox proportional hazards regression analysis identified pulmonary bacterial infection (P = 0.003) and spleen SUVmax (P = 0.032) as factors significantly associated with survival of IIM-ILD patients. The subsequent receiver operating characteristic curve analysis showed SUVmax was comparably effective in predicting death within 3 months. Conclusion: Spleen SUVmax and complication of pulmonary bacterial infection were significantly associated with survival of IIM patients. In addition, elevated spleen SUVmax was efficient in predicting unfavorable outcome of adult IIM patients. Key Points • IIM is a group of autoimmune diseases that frequently leads to unfavorable outcome • Complications of splenic SUVmax and pulmonary bacterial infection were significantly associated with survival in IIM patients [ABSTRACT FROM AUTHOR]

Diabetes mellitus (DM) and neurosyphilis (NS) may both damage the blood-brain barrier (BBB). It seems that non-neurosyphilis (non-NS) patients with high HbA1c levels are likely to develop into NS. However, the correlation of HbA1c level with BBB disruption in syphilis (non-NS) patients is unclear. In this study, we used dynamic contrast-enhanced (DCE) MRI to quantify regional BBB permeability in syphilis (non-NS) patients and detected several molecular biomarkers of cerebrospinal fluid (CSF). We found that BBB permeability values in the hippocampus, white matter, and cortex inferior temporal gyrus were correlated with albumin quotient (Qalb), CSF concentrations of interleukin IL-6 and IL-10. Moreover, BBB breakdown in white matter was correlated with CSF concentrations of sICAM-1 and sVCAM-1. In conclusion, our data suggest that BBB integrity may be liable to be disrupted in syphilis (non-NS) patients, patients with high HbA1c levels, as well as syphilis (non-NS) patients with high HbA1c levels, and it is particularly important to control blood glucose in these patients. [ABSTRACT FROM AUTHOR]

Vulnerable atherosclerotic plaques are responsible for most cardiovascular diseases (CVDs). Folate receptor (FR) positive activated macrophages were thought to be a prominent component in the development of vulnerable plaque. The objective of this study is to develop folate conjugated two-dimensional (2D) Pd@Au nanomaterials (Pd@Au-PEG-FA) for targeted multimodal imaging of the FRs in advanced atherosclerotic plaques. Pharmacokinetic and imaging studies (single photon emission computed tomography (SPECT), computed tomography (CT) and photoacoustic (PA) imaging) were performed to confirm the prolonged blood half-life and enrichment of radioactivity in atherosclerotic plaques. Strong signals were detected in vivo with SPECT, CT and PA imaging in heavy atherosclerotic plaques, which were significantly higher than those of the normal aortas after injection of Pd@Au-PEG-FA. Blocking studies with preinjection of excess FA could effectively reduce the targeting ability of Pd@Au-PEG-FA in atherosclerotic plaques, further demonstrating the specific binding of Pd@Au-PEG-FA for plaque lesions. Histopathological characterization revealed that the signal of probe was in accordance with the high-risk plaques. In summary, the Pd@Au-PEG-FA has favorable pharmacokinetic properties and provides a valuable approach for detecting high-risk plaques in the presence of FRs in atherosclerotic plaques. [ABSTRACT FROM AUTHOR]

Nanoparticles with effective tumor accumulation and efficient renal clearance have attracted significant interests for clinical applications. We prepared 2.5 nm tyrosine based carbon dots (TCDs) with phenolic hydroxyl groups on the surface for directly 125I labeling. The 125I labeled polyethylene glycol (PEG) functionalized TCDs (125I-TCDPEGs) showed excellent radiochemical stability both in vitro and in vivo. Due to the enhanced permeability and retention effect, these 125I-TCDPEGs demonstrated a tumor accumulation around 4%–5% of the injected dose per gram (ID/g) for U87MG, 4T1, HepG2 and MCF7 tumor-bearing mice at 1 h post-injection. Meanwhile, the 125I-TCDPEGs also could be fast renally excreted, with less than 0.6% ID/g left in the liver and spleen within 24 h. These radioactive carbon dots not only can be used for cellular fluorescence imaging due to their intrinsic optical property, but are also effective single photon emission computed tomography (SPECT) imaging agents for tumor. Together with their excellent biocompatibility and stability, we anticipate these 125I-TCDPEGs of great potential for early tumor diagnosis in clinic. What's more, our TCDPEGs are also proved to be feasible carriers for other iodine isotopes such as 127I and 131I for different biomedical application. [ABSTRACT FROM AUTHOR]

Cancer is one of the major causes of human death. There are many types of cancer treatment including surgery, chemotherapy, radiotherapy, and photothermal therapy. Combining different therapies can synergistically enhance the therapeutic effect. We developed porous hollow palladium nanoparticles (PHPdNPs) to co-deliver 131I (a radioisotope that is commonly used in radiotherapy) and doxorubicin (DOX; a chemotherapy drug). Compared with other mesoporous nanocarriers, our PHPdNPs exhibited impressive photothermal conversion efficiency and stability. Drug loading is high and drug release is controllable by repeated laser irradiation and acidic pH in tumor microenvironments. Owing to the specific interaction between palladium and iodine, the PHPdNPs serve as effective 131I delivery vehicles with excellent radiochemical stability. A single dose of [131I]PHPdNPs-DOX has superior antitumor efficacy because it enables a combination of chemo-, photothermal-, and radio-therapy. Moreover, the nanocomplex has no obvious side-effects in mice. Therefore, we believe that PHPdNPs are excellent candidates for multimodal imaging-guided therapy. [ABSTRACT FROM AUTHOR]

The insulin-like growth factor 1 receptor (IGF-1R) serves as an attractive target for cancer molecular imaging and therapy. Previous single photon emission computerized tomography (SPECT) studies showed that the IGF-1R-targeting Affibody molecules Tc-Z-GGGC, [Tc(CO)]-(HE)-Z and In-DOTA-Z can discriminate between high and low IGF-1R-expression tumors and have the potential for patient selection for IGF-1R-targeted therapy. Compared with SPECT, positron emission tomography (PET) may improve imaging of IGF-1R-expression, because of its high sensitivity, high spatial resolution, strong quantification ability. The aim of the present study was to develop the Cu-labeled NOTA-conjugated Affibody molecule Z as a PET probe for imaging of IGF-1R-positive tumor. An Affibody analogue (Ac-Cys-Z) binding to IGF-1R was site-specifically conjugated with NOTA and labeled with Cu. Binding affinity and specificity of Cu-NOTA-Z to IGF-1R were evaluated using human glioblastoma U87MG cells. Small-animal PET, biodistribution, and metabolic stability studies were conducted on mice bearing U87MG xenografts after the injection of Cu-NOTA-Z with or without co-injection of unlabeled Affibody proteins. The radiosynthesis of Cu-NOTA-Z was completed successfully within 60 min with a decay-corrected yield of 75 %. Cu-NOTA-Z bound to IGF-1R with low nanomolar affinity ( K = 28.55 ± 3.95 nM) in U87MG cells. Cu-NOTA-Z also displayed excellent in vitro and in vivo stability. In vivo biodistribution and PET studies demonstrated targeting of U87MG gliomas xenografts was IGF-1R specific. The tumor uptake was 5.08 ± 1.07 %ID/g, and the tumor to muscle ratio was 11.89 ± 2.16 at 24 h after injection. Small animal PET imaging studies revealed that Cu-NOTA-Z could clearly identify U87MG tumors with good contrast at 1-24 h after injection. This study demonstrates that Cu-NOTA-Z is a promising PET probe for imaging IGF-1R positive tumor. [ABSTRACT FROM AUTHOR]

Objectives: To develop an end-to-end radiomics-based pipeline for the prediction of International Society of Urological Pathology grade group (ISUP GG) in prostate cancer (PCa).This retrospective study includes 356 patients (241 in training set and 115 in independent test set) with histopathologically confirmed PCa who underwent [18F]PSMA-1007 PET/CT scan. Patients were classified into two groups according to their ISUP GG (1–3 vs. 4–5). Radiomics features were extracted from the whole, automatically segmented prostate on PET/CT images, 30 models were constructed by combining 6 feature selection algorithms and 5 machine learning classifiers. The clinical model incorporated age, total prostate-specific antigen (tPSA), maximum standardized uptake value (SUVmax), and prostate volume. The predictive performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC), balanced accuracy (bAcc), and decision curve analysis (DCA).The best-performing radiomics model significantly outperformed clinical model (AUC 0.879 ± 0.041 vs. 0.799 ± 0.051, bAcc 0.745 ± 0.074 vs. 0.629 ± 0.045). On an external independent test set, best-performing radiomics model perform better than clinical model, with an AUC of 0.861 vs. 0.750, p = 0.002 (Delong), and bAcc of 0.764 vs. 0.582, p = 0.043 (McNemar). The learning curve, calibration curve and DCA demonstrated goodness-of-fit and improved benefits in clinical practice.The end-to-end radiomics-based pipeline is an effective non-invasive tool to predict ISUP GG in PCa.Methods: To develop an end-to-end radiomics-based pipeline for the prediction of International Society of Urological Pathology grade group (ISUP GG) in prostate cancer (PCa).This retrospective study includes 356 patients (241 in training set and 115 in independent test set) with histopathologically confirmed PCa who underwent [18F]PSMA-1007 PET/CT scan. Patients were classified into two groups according to their ISUP GG (1–3 vs. 4–5). Radiomics features were extracted from the whole, automatically segmented prostate on PET/CT images, 30 models were constructed by combining 6 feature selection algorithms and 5 machine learning classifiers. The clinical model incorporated age, total prostate-specific antigen (tPSA), maximum standardized uptake value (SUVmax), and prostate volume. The predictive performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC), balanced accuracy (bAcc), and decision curve analysis (DCA).The best-performing radiomics model significantly outperformed clinical model (AUC 0.879 ± 0.041 vs. 0.799 ± 0.051, bAcc 0.745 ± 0.074 vs. 0.629 ± 0.045). On an external independent test set, best-performing radiomics model perform better than clinical model, with an AUC of 0.861 vs. 0.750, p = 0.002 (Delong), and bAcc of 0.764 vs. 0.582, p = 0.043 (McNemar). The learning curve, calibration curve and DCA demonstrated goodness-of-fit and improved benefits in clinical practice.The end-to-end radiomics-based pipeline is an effective non-invasive tool to predict ISUP GG in PCa.Results: To develop an end-to-end radiomics-based pipeline for the prediction of International Society of Urological Pathology grade group (ISUP GG) in prostate cancer (PCa).This retrospective study includes 356 patients (241 in training set and 115 in independent test set) with histopathologically confirmed PCa who underwent [18F]PSMA-1007 PET/CT scan. Patients were classified into two groups according to their ISUP GG (1–3 vs. 4–5). Radiomics features were extracted from the whole, automatically segmented prostate on PET/CT images, 30 models were constructed by combining 6 feature selection algorithms and 5 machine learning classifiers. The clinical model incorporated age, total prostate-specific antigen (tPSA), maximum standardized uptake value (SUVmax), and prostate volume. The predictive performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC), balanced accuracy (bAcc), and decision curve analysis (DCA).The best-performing radiomics model significantly outperformed clinical model (AUC 0.879 ± 0.041 vs. 0.799 ± 0.051, bAcc 0.745 ± 0.074 vs. 0.629 ± 0.045). On an external independent test set, best-performing radiomics model perform better than clinical model, with an AUC of 0.861 vs. 0.750, p = 0.002 (Delong), and bAcc of 0.764 vs. 0.582, p = 0.043 (McNemar). The learning curve, calibration curve and DCA demonstrated goodness-of-fit and improved benefits in clinical practice.The end-to-end radiomics-based pipeline is an effective non-invasive tool to predict ISUP GG in PCa.Conclusion: To develop an end-to-end radiomics-based pipeline for the prediction of International Society of Urological Pathology grade group (ISUP GG) in prostate cancer (PCa).This retrospective study includes 356 patients (241 in training set and 115 in independent test set) with histopathologically confirmed PCa who underwent [18F]PSMA-1007 PET/CT scan. Patients were classified into two groups according to their ISUP GG (1–3 vs. 4–5). Radiomics features were extracted from the whole, automatically segmented prostate on PET/CT images, 30 models were constructed by combining 6 feature selection algorithms and 5 machine learning classifiers. The clinical model incorporated age, total prostate-specific antigen (tPSA), maximum standardized uptake value (SUVmax), and prostate volume. The predictive performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC), balanced accuracy (bAcc), and decision curve analysis (DCA).The best-performing radiomics model significantly outperformed clinical model (AUC 0.879 ± 0.041 vs. 0.799 ± 0.051, bAcc 0.745 ± 0.074 vs. 0.629 ± 0.045). On an external independent test set, best-performing radiomics model perform better than clinical model, with an AUC of 0.861 vs. 0.750, p = 0.002 (Delong), and bAcc of 0.764 vs. 0.582, p = 0.043 (McNemar). The learning curve, calibration curve and DCA demonstrated goodness-of-fit and improved benefits in clinical practice.The end-to-end radiomics-based pipeline is an effective non-invasive tool to predict ISUP GG in PCa. [ABSTRACT FROM AUTHOR]

Purpose: Technological advances in instruments have greatly promoted the development of positron emission tomography (PET) scanners. State-of-the-art PET scanners such as uEXPLORER can collect PET images of significantly higher quality. However, these scanners are not currently available in most local hospitals due to the high cost of manufacturing and maintenance. Our study aims to convert low-quality PET images acquired by common PET scanners into images of comparable quality to those obtained by state-of-the-art scanners without the need for paired low- and high-quality PET images.In this paper, we proposed an improved CycleGAN (IE-CycleGAN) model for unpaired PET image enhancement. The proposed method is based on CycleGAN, and the correlation coefficient loss and patient-specific prior loss were added to constrain the structure of the generated images. Furthermore, we defined a normalX-to-advanced training strategy to enhance the generalization ability of the network. The proposed method was validated on unpaired uEXPLORER datasets and Biograph Vision local hospital datasets.For the uEXPLORER dataset, the proposed method achieved better results than non-local mean filtering (NLM), block-matching and 3D filtering (BM3D), and deep image prior (DIP), which are comparable to Unet (supervised) and CycleGAN (supervised). For the Biograph Vision local hospital datasets, the proposed method achieved higher contrast-to-noise ratios (CNR) and tumor-to-background SUVmax ratios (TBR) than NLM, BM3D, and DIP. In addition, the proposed method showed higher contrast, SUVmax, and TBR than Unet (supervised) and CycleGAN (supervised) when applied to images from different scanners.The proposed unpaired PET image enhancement method outperforms NLM, BM3D, and DIP. Moreover, it performs better than the Unet (supervised) and CycleGAN (supervised) when implemented on local hospital datasets, which demonstrates its excellent generalization ability.Methods: Technological advances in instruments have greatly promoted the development of positron emission tomography (PET) scanners. State-of-the-art PET scanners such as uEXPLORER can collect PET images of significantly higher quality. However, these scanners are not currently available in most local hospitals due to the high cost of manufacturing and maintenance. Our study aims to convert low-quality PET images acquired by common PET scanners into images of comparable quality to those obtained by state-of-the-art scanners without the need for paired low- and high-quality PET images.In this paper, we proposed an improved CycleGAN (IE-CycleGAN) model for unpaired PET image enhancement. The proposed method is based on CycleGAN, and the correlation coefficient loss and patient-specific prior loss were added to constrain the structure of the generated images. Furthermore, we defined a normalX-to-advanced training strategy to enhance the generalization ability of the network. The proposed method was validated on unpaired uEXPLORER datasets and Biograph Vision local hospital datasets.For the uEXPLORER dataset, the proposed method achieved better results than non-local mean filtering (NLM), block-matching and 3D filtering (BM3D), and deep image prior (DIP), which are comparable to Unet (supervised) and CycleGAN (supervised). For the Biograph Vision local hospital datasets, the proposed method achieved higher contrast-to-noise ratios (CNR) and tumor-to-background SUVmax ratios (TBR) than NLM, BM3D, and DIP. In addition, the proposed method showed higher contrast, SUVmax, and TBR than Unet (supervised) and CycleGAN (supervised) when applied to images from different scanners.The proposed unpaired PET image enhancement method outperforms NLM, BM3D, and DIP. Moreover, it performs better than the Unet (supervised) and CycleGAN (supervised) when implemented on local hospital datasets, which demonstrates its excellent generalization ability.Results: Technological advances in instruments have greatly promoted the development of positron emission tomography (PET) scanners. State-of-the-art PET scanners such as uEXPLORER can collect PET images of significantly higher quality. However, these scanners are not currently available in most local hospitals due to the high cost of manufacturing and maintenance. Our study aims to convert low-quality PET images acquired by common PET scanners into images of comparable quality to those obtained by state-of-the-art scanners without the need for paired low- and high-quality PET images.In this paper, we proposed an improved CycleGAN (IE-CycleGAN) model for unpaired PET image enhancement. The proposed method is based on CycleGAN, and the correlation coefficient loss and patient-specific prior loss were added to constrain the structure of the generated images. Furthermore, we defined a normalX-to-advanced training strategy to enhance the generalization ability of the network. The proposed method was validated on unpaired uEXPLORER datasets and Biograph Vision local hospital datasets.For the uEXPLORER dataset, the proposed method achieved better results than non-local mean filtering (NLM), block-matching and 3D filtering (BM3D), and deep image prior (DIP), which are comparable to Unet (supervised) and CycleGAN (supervised). For the Biograph Vision local hospital datasets, the proposed method achieved higher contrast-to-noise ratios (CNR) and tumor-to-background SUVmax ratios (TBR) than NLM, BM3D, and DIP. In addition, the proposed method showed higher contrast, SUVmax, and TBR than Unet (supervised) and CycleGAN (supervised) when applied to images from different scanners.The proposed unpaired PET image enhancement method outperforms NLM, BM3D, and DIP. Moreover, it performs better than the Unet (supervised) and CycleGAN (supervised) when implemented on local hospital datasets, which demonstrates its excellent generalization ability.Conclusion: Technological advances in instruments have greatly promoted the development of positron emission tomography (PET) scanners. State-of-the-art PET scanners such as uEXPLORER can collect PET images of significantly higher quality. However, these scanners are not currently available in most local hospitals due to the high cost of manufacturing and maintenance. Our study aims to convert low-quality PET images acquired by common PET scanners into images of comparable quality to those obtained by state-of-the-art scanners without the need for paired low- and high-quality PET images.In this paper, we proposed an improved CycleGAN (IE-CycleGAN) model for unpaired PET image enhancement. The proposed method is based on CycleGAN, and the correlation coefficient loss and patient-specific prior loss were added to constrain the structure of the generated images. Furthermore, we defined a normalX-to-advanced training strategy to enhance the generalization ability of the network. The proposed method was validated on unpaired uEXPLORER datasets and Biograph Vision local hospital datasets.For the uEXPLORER dataset, the proposed method achieved better results than non-local mean filtering (NLM), block-matching and 3D filtering (BM3D), and deep image prior (DIP), which are comparable to Unet (supervised) and CycleGAN (supervised). For the Biograph Vision local hospital datasets, the proposed method achieved higher contrast-to-noise ratios (CNR) and tumor-to-background SUVmax ratios (TBR) than NLM, BM3D, and DIP. In addition, the proposed method showed higher contrast, SUVmax, and TBR than Unet (supervised) and CycleGAN (supervised) when applied to images from different scanners.The proposed unpaired PET image enhancement method outperforms NLM, BM3D, and DIP. Moreover, it performs better than the Unet (supervised) and CycleGAN (supervised) when implemented on local hospital datasets, which demonstrates its excellent generalization ability. [ABSTRACT FROM AUTHOR]