Objective:To investigate the denoising performance of different deep learning (DL) strategies on low-dose brain 18F-FDG PET images. Methods:This retrospective methodological study was conducted on brain PET/CT images of 50 patients (35 males, 15 females, age 20-87 years) who received 3.7MBq/kg 18F-FDG at the Fifth Affiliated Hospital of Sun Yat-sen University between May 2023 and January 2024. Full-dose PET data were acquired with 2min scan. CT scans were acquired before PET scanning. Low-dose PET sinograms were generated by down-sampling the full-dose list mode data to 1/2, 1/4, and 1/20 of full-dose count level. Both full-dose and low-dose sinograms were reconstructed with random, CT-based attenuation and scatter corrections using the three-dimensional (3D) ordered-subsets expectation maximization (OSEM) algorithm (2 iterations, 20 subsets). A total of 4 DL denoising methods were established: (1) 3D conditional generative adversarial networks (GAN) using only low-dose PET as input (GAN-1); (2) 3D attention-based GAN (AttGAN) with low-dose PET input (AttGAN-1); (3) 3D AttGAN with low-dose PET and CT inputs (AttGAN-2); (4) 3D AttGAN with frequency-separation using low-dose PET and CT inputs (AttGAN-FS-2). For AttGAN-FS-2, during the frequency division process, high- and low-frequency components were extracted from the PET reconstructed images via Fourier transform, then inversed Fourier transform, denoised separately, and finally combined to produce the final denoised images. The dataset was separated into training (70%), validation (10%) and testing (20%) sets using simple random sampling without replacement with a fixed random seed. A 5-fold cross-validation scheme was then applied to test all 50 patients. Performance was evaluated against full-dose PET using normalized mean square error (NMSE), structural similarity (SSIM), peak signal-to-noise ratio (PSNR), contrast-to-noise ratio (CNR), SUV mean and SUV max bias of selected brain ROIs. Wilcoxon signed rank test was used to analyze the differences between the denoising methods. Results:AttGAN-FS-2 showed the best performance among all dose levels, with statistical difference as compared by low-dose PET and GAN-1 denoised images for NMSE, SSIM, PSNR, and CNR ( Z values: 2.92-6.15, all P<0.005). NMSE, SSIM quantitative evaluation results (median) of each model at 1/20 dose were: GAN-1: 0.08, 0.87, AttGAN-1: 0.08, 0.88, AttGAN-2: 0.07, 0.89, AttGAN-FS-2: 0.06, 0.91, respectively ( Z values: 3.24-5.77, all P<0.005). Conclusion:The DL-based method combined with multiple strategies AttGAN-FS-2 shows improved denoising performance for low-dose brain PET images.

Objective: This study investigated the value of multiparametric MRI (mpMRI) in predicting Gleason score (GS) upgrading and downgrading in radical prostatectomy (RP) compared with presurgical biopsy. Materials and Methods: Clinical and mpMRI data were retrospectively collected from 219 patients with prostate disease between January 2015 and December 2021. All patients underwent systematic prostate biopsy followed by RP. MpMRI included conventional diffusion-weighted and dynamic contrast-enhanced imaging. Multivariable logistic regression analysis was performed to analyze the factors associated with GS upgrading and downgrading after RP. Receiver operating characteristic curve analysis was used to estimate the area under the curve (AUC) to indicate the performance of the multivariable logistic regression models in predicting GS upgrade and downgrade after RP. Results: The GS after RP was upgraded, downgraded, and unchanged in 92, 43, and 84 patients, respectively. The AUCs of the clinical (percentage of positive biopsy cores [PBCs], time from biopsy to RP) and mpMRI models (prostate cancer [PCa] location, Prostate Imaging Reporting and Data System [PI-RADS] v2.1 score) for predicting GS upgrading after RP were 0.714 and 0.749, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, tPSA, PCa location, and PIRADS v2.1 score) was 0.816, which was larger than that of the clinical factors alone (P < 0.001). The AUCs of the clinical (age, percentage of PBCs, ratio of free/total PSA [F/T]) and mpMRI models (PCa diameter, PCa location, and PI-RADS v2.1 score) for predicting GS downgrading after RP were 0.749 and 0.835, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, F/T, PCa diameter, PCa location, and PI-RADS v2.1 score) was 0.883, which was larger than that of the clinical factors alone (P < 0.001). Conclusion Combining clinical factors and mpMRI findings can predict GS upgrade and downgrade after RP more accurately than using clinical factors alone.

Objective: This study investigated the value of multiparametric MRI (mpMRI) in predicting Gleason score (GS) upgrading and downgrading in radical prostatectomy (RP) compared with presurgical biopsy. Materials and Methods: Clinical and mpMRI data were retrospectively collected from 219 patients with prostate disease between January 2015 and December 2021. All patients underwent systematic prostate biopsy followed by RP. MpMRI included conventional diffusion-weighted and dynamic contrast-enhanced imaging. Multivariable logistic regression analysis was performed to analyze the factors associated with GS upgrading and downgrading after RP. Receiver operating characteristic curve analysis was used to estimate the area under the curve (AUC) to indicate the performance of the multivariable logistic regression models in predicting GS upgrade and downgrade after RP. Results: The GS after RP was upgraded, downgraded, and unchanged in 92, 43, and 84 patients, respectively. The AUCs of the clinical (percentage of positive biopsy cores [PBCs], time from biopsy to RP) and mpMRI models (prostate cancer [PCa] location, Prostate Imaging Reporting and Data System [PI-RADS] v2.1 score) for predicting GS upgrading after RP were 0.714 and 0.749, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, tPSA, PCa location, and PIRADS v2.1 score) was 0.816, which was larger than that of the clinical factors alone (P < 0.001). The AUCs of the clinical (age, percentage of PBCs, ratio of free/total PSA [F/T]) and mpMRI models (PCa diameter, PCa location, and PI-RADS v2.1 score) for predicting GS downgrading after RP were 0.749 and 0.835, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, F/T, PCa diameter, PCa location, and PI-RADS v2.1 score) was 0.883, which was larger than that of the clinical factors alone (P < 0.001). Conclusion Combining clinical factors and mpMRI findings can predict GS upgrade and downgrade after RP more accurately than using clinical factors alone.

Objective: This study investigated the value of multiparametric MRI (mpMRI) in predicting Gleason score (GS) upgrading and downgrading in radical prostatectomy (RP) compared with presurgical biopsy. Materials and Methods: Clinical and mpMRI data were retrospectively collected from 219 patients with prostate disease between January 2015 and December 2021. All patients underwent systematic prostate biopsy followed by RP. MpMRI included conventional diffusion-weighted and dynamic contrast-enhanced imaging. Multivariable logistic regression analysis was performed to analyze the factors associated with GS upgrading and downgrading after RP. Receiver operating characteristic curve analysis was used to estimate the area under the curve (AUC) to indicate the performance of the multivariable logistic regression models in predicting GS upgrade and downgrade after RP. Results: The GS after RP was upgraded, downgraded, and unchanged in 92, 43, and 84 patients, respectively. The AUCs of the clinical (percentage of positive biopsy cores [PBCs], time from biopsy to RP) and mpMRI models (prostate cancer [PCa] location, Prostate Imaging Reporting and Data System [PI-RADS] v2.1 score) for predicting GS upgrading after RP were 0.714 and 0.749, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, tPSA, PCa location, and PIRADS v2.1 score) was 0.816, which was larger than that of the clinical factors alone (P < 0.001). The AUCs of the clinical (age, percentage of PBCs, ratio of free/total PSA [F/T]) and mpMRI models (PCa diameter, PCa location, and PI-RADS v2.1 score) for predicting GS downgrading after RP were 0.749 and 0.835, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, F/T, PCa diameter, PCa location, and PI-RADS v2.1 score) was 0.883, which was larger than that of the clinical factors alone (P < 0.001). Conclusion Combining clinical factors and mpMRI findings can predict GS upgrade and downgrade after RP more accurately than using clinical factors alone.

Objective: This study investigated the value of multiparametric MRI (mpMRI) in predicting Gleason score (GS) upgrading and downgrading in radical prostatectomy (RP) compared with presurgical biopsy. Materials and Methods: Clinical and mpMRI data were retrospectively collected from 219 patients with prostate disease between January 2015 and December 2021. All patients underwent systematic prostate biopsy followed by RP. MpMRI included conventional diffusion-weighted and dynamic contrast-enhanced imaging. Multivariable logistic regression analysis was performed to analyze the factors associated with GS upgrading and downgrading after RP. Receiver operating characteristic curve analysis was used to estimate the area under the curve (AUC) to indicate the performance of the multivariable logistic regression models in predicting GS upgrade and downgrade after RP. Results: The GS after RP was upgraded, downgraded, and unchanged in 92, 43, and 84 patients, respectively. The AUCs of the clinical (percentage of positive biopsy cores [PBCs], time from biopsy to RP) and mpMRI models (prostate cancer [PCa] location, Prostate Imaging Reporting and Data System [PI-RADS] v2.1 score) for predicting GS upgrading after RP were 0.714 and 0.749, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, tPSA, PCa location, and PIRADS v2.1 score) was 0.816, which was larger than that of the clinical factors alone (P < 0.001). The AUCs of the clinical (age, percentage of PBCs, ratio of free/total PSA [F/T]) and mpMRI models (PCa diameter, PCa location, and PI-RADS v2.1 score) for predicting GS downgrading after RP were 0.749 and 0.835, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, F/T, PCa diameter, PCa location, and PI-RADS v2.1 score) was 0.883, which was larger than that of the clinical factors alone (P < 0.001). Conclusion Combining clinical factors and mpMRI findings can predict GS upgrade and downgrade after RP more accurately than using clinical factors alone.

Objective:To explore the effect of social cognition and interaction training (SCIT) combined with transcranial magnetic stimulation (TMS) on intrinsic motivation and social cognition in patients with schizophrenia.Methods:Forty-two stable schizophrenia patients were randomly divided into the SCIT + TMS group( n=22) and the SCIT group( n=20). All the subjects received 20 sessions of SCIT treatment, and the SCIT+ TMS group simultaneously received 15 sessions of intermittent theta burst stimulation(iTBS) over the left dorsolateral prefrontal cortex(DLPFC). All the subjects were assessed by intrinsic motivation inventory for schizophrenia research(IMI-SR), Chinese version of the ambiguous intentions hostility questionnaire(AIHQ-C), theory of mind-picture sequencing task(ToM-PST), mentalization scale (MentS), Chinese version of interpersonal reactivity index (IRI-C) and positive and negative syndrome scale (PANSS) before and after intervention. SPSS 26.0 software was used for data analysis.Wilcoxon signed-rank test was used for intra-group comparison before and after treatment, while Mann-Whitney U test and covariance analysis were used for inter-group comparison.Spearman correlation analysis and Logistic regression analyses were used to explore the association between the intrinsic motivation and social cognition. Results:There were no significant differences on IMI-SR scores before and after treatment between the two groups(all P>0.05). In the SCIT+ TMS group, the total score of hostility bias (HB), HB scores in ambiguous scenes, HB scores in intentional scenes, and aggressive bias (AB) scores in ambiguous scenes of AIHQ-C scale after treatment were lower than those befor treatment( Z=-2.044--3.112, all P<0.05), while the total score of ToM-PST(18.50(16.00, 21.00) vs 15.50(11.75, 18.00), Z=-2.598, P=0.009) and IRI-C imagination score (12.18±3.79, 14.41±4.73, t=-2.694, P=0.014) were higher than those before treatment.In the SCIT group, the total score of ToM-PST after treatment was higher than that before treatment(21.00(20.00, 22.00) vs 17.00(14.50, 20.75), Z=-2.518, P=0.012).There was no significant statistical difference in MentS scores between after treatment and before treatment ( P>0.05). The difference in AIHQ-C intentional scenario AB score before and after treatment was higher in the SCIT+ TMS group than in the SCIT group ( Z=-1.996, P=0.046), while there was no statistically significant difference in the difference before and after treatment in social cognitive scores between the two groups (all P>0.05).In the combined two samples, Spearman correlation analysis showed that the total score of IMI-SR before treatment was positively correlated with the primary belief score of ToM-PST understanding, reciprocity score, MentS total score, other person mentalization score, motivation mentalization score, IRI-C total score, viewpoint taking score, and empathy concern score after treatment( r=0.341-0.509, all P<0.05), while negatively correlated with AIHQ-C total score and factor scores ( r=-0.434--0.645, P<0.05).Logistic regression analysis showed that the total score of IMI-SR had negative impact on AIHQ-C total HB score( B=-0.047, OR=0.954, 95% CI=0.917-0.993).The value score had a positive impact on the total score of MentS ( B=0.143, OR=1.154, 95% CI=1.043-1.277), other person mentalization score( B=0.166, OR=1.181, 95% CI=1.058-1.318), motivation mentalization score( B=0.111, OR=1.117, 95% CI=1.021-1.223), IRI-C total score ( B=0.138, OR=1.148, 95% CI=1.038-1.270), and viewpoint taking score( B=0.194, OR=1.214, 95% CI=1.076-1.369). Interest score had a positive impact on IRI-C empathy concern score ( B=0.098, OR=1.103, 95% CI=0.998-1.218) and ToM-PST understanding primary belief score( B=0.130, OR=1.138, 95% CI=1.010-1.283) and reciprocity score( B=0.189, OR=1.208, 95% CI=1.057-1.380). Conclusion:The research results did not confirm the effect of TMS over the DLPFC on enhancing intrinsic motivation, as well as the synergistic effect of SCIT treatment on social cognition. But the correlation results indicates that improving schizophrenia patients' intrinsic motivation level in cognitive training is meaningful for promoting social cognition.

Objective: This study investigated the value of multiparametric MRI (mpMRI) in predicting Gleason score (GS) upgrading and downgrading in radical prostatectomy (RP) compared with presurgical biopsy. Materials and Methods: Clinical and mpMRI data were retrospectively collected from 219 patients with prostate disease between January 2015 and December 2021. All patients underwent systematic prostate biopsy followed by RP. MpMRI included conventional diffusion-weighted and dynamic contrast-enhanced imaging. Multivariable logistic regression analysis was performed to analyze the factors associated with GS upgrading and downgrading after RP. Receiver operating characteristic curve analysis was used to estimate the area under the curve (AUC) to indicate the performance of the multivariable logistic regression models in predicting GS upgrade and downgrade after RP. Results: The GS after RP was upgraded, downgraded, and unchanged in 92, 43, and 84 patients, respectively. The AUCs of the clinical (percentage of positive biopsy cores [PBCs], time from biopsy to RP) and mpMRI models (prostate cancer [PCa] location, Prostate Imaging Reporting and Data System [PI-RADS] v2.1 score) for predicting GS upgrading after RP were 0.714 and 0.749, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, tPSA, PCa location, and PIRADS v2.1 score) was 0.816, which was larger than that of the clinical factors alone (P < 0.001). The AUCs of the clinical (age, percentage of PBCs, ratio of free/total PSA [F/T]) and mpMRI models (PCa diameter, PCa location, and PI-RADS v2.1 score) for predicting GS downgrading after RP were 0.749 and 0.835, respectively. The AUC of the combined diagnostic model (age, percentage of PBCs, F/T, PCa diameter, PCa location, and PI-RADS v2.1 score) was 0.883, which was larger than that of the clinical factors alone (P < 0.001). Conclusion Combining clinical factors and mpMRI findings can predict GS upgrade and downgrade after RP more accurately than using clinical factors alone.

Objective:To evaluate the imaging stability of a portable boom-type ophthalmic ultra-high-resolution optical coherence tomography (OCT) device in multiple application scenarios.Methods:The boom-type mode and handheld mode of the portable boom-type OCT and the desktop OCT were used to perform three-dimensional imaging tests on three healthy adults undergoing physical examinations at the Zhongshan Ophthalmic Center, Sun Yat-sen University as well as on OEMI-7 model eyes in a sitting position.The same two modes of the portable boom-type OCT were used to perform three-dimensional imaging on four awake non-sedated infants, two sedated infants and four healthy adults in the supine position.The obtained 3D imaging data were processed using a correlation analysis method between adjacent B-scans, and the offset of B-scan in the axial (z-axis) and the fast axis transverse (x-axis) were calculated.The procedures for in vivo human eye experiments followed the Declaration of Helsinki and were approved by the Ethics Committee of Zhongshan Ophthalmic Center, Sun Yat-sen University (No.2020 KYPJ154).All subjects and infant guardians signed the informed consent form. Results:Compared with the handheld imaging mode, the axial and fast axis lateral motion offsets of the model eye were significantly reduced in the boom-type imaging mode from (124.00±12.49)μm to (48.00±15.87)μm and from (24.00±1.00)μm to (2.67±0.57)μm, respectively ( t=2.932, 4.337; both P<0.001).In both human and model eyes, the axial and fast axis lateral motion offsets of the boom-type mode were significantly lower than in the traditional handheld operation mode (both P<0.001).The axial and lateral motion offsets between the boom-type mode and desk-top OCT imaging were comparable, without significant differences (both P>0.05).In both sedated and awake, non-sedated infants in the supine position, the axial offset of the portable boom-type OCT system was similar to that of the healthy adults, without significant difference in the overall comparison ( P=0.385), and the lateral offsets were higher than those of healthy adults, with statistically significant differences (both P=0.013).There was no significant difference in axial deviation between sedated and non-sedated infants ( P>0.05).The lateral deviation of non-sedated infants was higher than that of sedated infants, though the difference was not statistically significant ( P=0.247). Conclusions:The portable boom-type OCT system can maintain high-speed, high-resolution imaging performance while achieving imaging stability comparable to traditional desktop OCT systems.It is more suitable for bedside imaging of supine subjects, especially uncooperative infants, and has good clinical application prospects.

Objective:To explore the effect of social cognition and interaction training (SCIT) combined with transcranial magnetic stimulation (TMS) on intrinsic motivation and social cognition in patients with schizophrenia.Methods:Forty-two stable schizophrenia patients were randomly divided into the SCIT + TMS group( n=22) and the SCIT group( n=20). All the subjects received 20 sessions of SCIT treatment, and the SCIT+ TMS group simultaneously received 15 sessions of intermittent theta burst stimulation(iTBS) over the left dorsolateral prefrontal cortex(DLPFC). All the subjects were assessed by intrinsic motivation inventory for schizophrenia research(IMI-SR), Chinese version of the ambiguous intentions hostility questionnaire(AIHQ-C), theory of mind-picture sequencing task(ToM-PST), mentalization scale (MentS), Chinese version of interpersonal reactivity index (IRI-C) and positive and negative syndrome scale (PANSS) before and after intervention. SPSS 26.0 software was used for data analysis.Wilcoxon signed-rank test was used for intra-group comparison before and after treatment, while Mann-Whitney U test and covariance analysis were used for inter-group comparison.Spearman correlation analysis and Logistic regression analyses were used to explore the association between the intrinsic motivation and social cognition. Results:There were no significant differences on IMI-SR scores before and after treatment between the two groups(all P>0.05). In the SCIT+ TMS group, the total score of hostility bias (HB), HB scores in ambiguous scenes, HB scores in intentional scenes, and aggressive bias (AB) scores in ambiguous scenes of AIHQ-C scale after treatment were lower than those befor treatment( Z=-2.044--3.112, all P<0.05), while the total score of ToM-PST(18.50(16.00, 21.00) vs 15.50(11.75, 18.00), Z=-2.598, P=0.009) and IRI-C imagination score (12.18±3.79, 14.41±4.73, t=-2.694, P=0.014) were higher than those before treatment.In the SCIT group, the total score of ToM-PST after treatment was higher than that before treatment(21.00(20.00, 22.00) vs 17.00(14.50, 20.75), Z=-2.518, P=0.012).There was no significant statistical difference in MentS scores between after treatment and before treatment ( P>0.05). The difference in AIHQ-C intentional scenario AB score before and after treatment was higher in the SCIT+ TMS group than in the SCIT group ( Z=-1.996, P=0.046), while there was no statistically significant difference in the difference before and after treatment in social cognitive scores between the two groups (all P>0.05).In the combined two samples, Spearman correlation analysis showed that the total score of IMI-SR before treatment was positively correlated with the primary belief score of ToM-PST understanding, reciprocity score, MentS total score, other person mentalization score, motivation mentalization score, IRI-C total score, viewpoint taking score, and empathy concern score after treatment( r=0.341-0.509, all P<0.05), while negatively correlated with AIHQ-C total score and factor scores ( r=-0.434--0.645, P<0.05).Logistic regression analysis showed that the total score of IMI-SR had negative impact on AIHQ-C total HB score( B=-0.047, OR=0.954, 95% CI=0.917-0.993).The value score had a positive impact on the total score of MentS ( B=0.143, OR=1.154, 95% CI=1.043-1.277), other person mentalization score( B=0.166, OR=1.181, 95% CI=1.058-1.318), motivation mentalization score( B=0.111, OR=1.117, 95% CI=1.021-1.223), IRI-C total score ( B=0.138, OR=1.148, 95% CI=1.038-1.270), and viewpoint taking score( B=0.194, OR=1.214, 95% CI=1.076-1.369). Interest score had a positive impact on IRI-C empathy concern score ( B=0.098, OR=1.103, 95% CI=0.998-1.218) and ToM-PST understanding primary belief score( B=0.130, OR=1.138, 95% CI=1.010-1.283) and reciprocity score( B=0.189, OR=1.208, 95% CI=1.057-1.380). Conclusion:The research results did not confirm the effect of TMS over the DLPFC on enhancing intrinsic motivation, as well as the synergistic effect of SCIT treatment on social cognition. But the correlation results indicates that improving schizophrenia patients' intrinsic motivation level in cognitive training is meaningful for promoting social cognition.

Objective:To evaluate the imaging stability of a portable boom-type ophthalmic ultra-high-resolution optical coherence tomography (OCT) device in multiple application scenarios.Methods:The boom-type mode and handheld mode of the portable boom-type OCT and the desktop OCT were used to perform three-dimensional imaging tests on three healthy adults undergoing physical examinations at the Zhongshan Ophthalmic Center, Sun Yat-sen University as well as on OEMI-7 model eyes in a sitting position.The same two modes of the portable boom-type OCT were used to perform three-dimensional imaging on four awake non-sedated infants, two sedated infants and four healthy adults in the supine position.The obtained 3D imaging data were processed using a correlation analysis method between adjacent B-scans, and the offset of B-scan in the axial (z-axis) and the fast axis transverse (x-axis) were calculated.The procedures for in vivo human eye experiments followed the Declaration of Helsinki and were approved by the Ethics Committee of Zhongshan Ophthalmic Center, Sun Yat-sen University (No.2020 KYPJ154).All subjects and infant guardians signed the informed consent form. Results:Compared with the handheld imaging mode, the axial and fast axis lateral motion offsets of the model eye were significantly reduced in the boom-type imaging mode from (124.00±12.49)μm to (48.00±15.87)μm and from (24.00±1.00)μm to (2.67±0.57)μm, respectively ( t=2.932, 4.337; both P<0.001).In both human and model eyes, the axial and fast axis lateral motion offsets of the boom-type mode were significantly lower than in the traditional handheld operation mode (both P<0.001).The axial and lateral motion offsets between the boom-type mode and desk-top OCT imaging were comparable, without significant differences (both P>0.05).In both sedated and awake, non-sedated infants in the supine position, the axial offset of the portable boom-type OCT system was similar to that of the healthy adults, without significant difference in the overall comparison ( P=0.385), and the lateral offsets were higher than those of healthy adults, with statistically significant differences (both P=0.013).There was no significant difference in axial deviation between sedated and non-sedated infants ( P>0.05).The lateral deviation of non-sedated infants was higher than that of sedated infants, though the difference was not statistically significant ( P=0.247). Conclusions:The portable boom-type OCT system can maintain high-speed, high-resolution imaging performance while achieving imaging stability comparable to traditional desktop OCT systems.It is more suitable for bedside imaging of supine subjects, especially uncooperative infants, and has good clinical application prospects.