Objective:To develop an artificial intelligence（AI）-based sequential ultrasound-magnetic resonance imaging（US-MRI）diagnostic strategy to optimize the imaging workflow for ovarian masses.Methods:A total of 1 120 patients with pathologically confirmed ovarian masses who underwent both preoperative pelvic ultrasound and MRI between January 2021 and December 2023 at Women's Hospital，Zhejiang University School of Medicine were retrospectively included. Patients were randomly divided into the training（ n=672）and internal test set（ n=448）at a ratio of 6∶4. An external test set（ n=128）was established at the Forth Affiliated Hospital of School of Medicine. Deep learning was used for automated segmentation of MRI lesions，followed by radiomic feature extraction and machine learning classification to construct both a US-MRI multimodal model and sequential US-MRI strategy. Diagnostic performance and potential healthcare cost-saving effects were evaluated across strategies. Results:In the internal test set（ n=448），the AI-based sequential US-MRI strategy achieved a F1 score of 0.863 and a diagnostic accuracy of 82.14%，with no significant difference compared to the US-MRI multi-modal model（ P>0.05）. The sequential strategy identified 82 cases（18.30%，82/448）of patients as low-risk true negatives during initial ultrasound screening，suggesting a potential to reduce the need for MRI examinations in future clinical practice. In the external test set（ n=128），the strategy achieved an F1 score of 0.800 and a confirmed diagnosis rate of 85.94%，with a theoretical reduction of 26.56%（34 cases）in MRI utilization while maintaining a diagnostic accuracy rate higher than that of the multi-modal model（82.18%）. Conclusions:The AI-based US-MRI sequential diagnostic strategy demonstrates favorable diagnostic accuracy while offering the potential to optimize MRI utilization. This approach may enhance the efficiency of imaging resource allocation and reduce healthcare burden in the management of ovarian masses.

Objective To establish a transgenic mouse model of facioscapulohumeral muscular dystrophy(FSHD)using tamoxifen induction and Myf6-CreERT2 and FLExDUX4 mice.Methods Dual transgenic(M6D4/+)mice were generated by crossbreeding Myf6-CreERT2 hemizygous and FLExDUX4 hemizygous mice.Full-length DUX4(DUX4-fl)expression was induced by tamoxifen starting at 3 weeks old.The disease model was evaluated at 9 weeks old by assessing changes in body mass,four-limb strength,inverted screen test,skeletal muscle weight ratio,hematoxylin/eosin,Picrosirius Red,and immunofluorescent staining of skeletal muscle paraffin sections,quantitative real-time polymerase chain reaction(RT-PCR),and RNA-sequencing(RNA-seq)of skeletal muscle.Results Dual transgenic heterozygous mice(M6D4/+)were successfully obtained.These mice exhibited significant physiological and pathological changes at 9 weeks,including delayed weight gain,reduced four-limb strength and endurance,decreased skeletal muscle weight ratio,and increases in centrally nucleated muscle fibers and fibrosis.Expression levels of DUX4 and its targeted genes were significantly up-regulated in skeletal muscle,as demonstrated by RT-PCR.RNA-seq revealed up-regulation of immune regulation-,interleukin-6,and tumor necrosis factor-related genes and down-regulation of skeletal muscle development-and differentiation-related genes.Conclusions M6D4/+mice effectively simulated the skeletal muscle phenotype of FSHD and thus provide a good animal model for research into the pathogenesis,intervention,and treatment of FSHD.

Objective To establish a transgenic mouse model of facioscapulohumeral muscular dystrophy(FSHD)using tamoxifen induction and Myf6-CreERT2 and FLExDUX4 mice.Methods Dual transgenic(M6D4/+)mice were generated by crossbreeding Myf6-CreERT2 hemizygous and FLExDUX4 hemizygous mice.Full-length DUX4(DUX4-fl)expression was induced by tamoxifen starting at 3 weeks old.The disease model was evaluated at 9 weeks old by assessing changes in body mass,four-limb strength,inverted screen test,skeletal muscle weight ratio,hematoxylin/eosin,Picrosirius Red,and immunofluorescent staining of skeletal muscle paraffin sections,quantitative real-time polymerase chain reaction(RT-PCR),and RNA-sequencing(RNA-seq)of skeletal muscle.Results Dual transgenic heterozygous mice(M6D4/+)were successfully obtained.These mice exhibited significant physiological and pathological changes at 9 weeks,including delayed weight gain,reduced four-limb strength and endurance,decreased skeletal muscle weight ratio,and increases in centrally nucleated muscle fibers and fibrosis.Expression levels of DUX4 and its targeted genes were significantly up-regulated in skeletal muscle,as demonstrated by RT-PCR.RNA-seq revealed up-regulation of immune regulation-,interleukin-6,and tumor necrosis factor-related genes and down-regulation of skeletal muscle development-and differentiation-related genes.Conclusions M6D4/+mice effectively simulated the skeletal muscle phenotype of FSHD and thus provide a good animal model for research into the pathogenesis,intervention,and treatment of FSHD.

Objective:To develop an artificial intelligence（AI）-based sequential ultrasound-magnetic resonance imaging（US-MRI）diagnostic strategy to optimize the imaging workflow for ovarian masses.Methods:A total of 1 120 patients with pathologically confirmed ovarian masses who underwent both preoperative pelvic ultrasound and MRI between January 2021 and December 2023 at Women's Hospital，Zhejiang University School of Medicine were retrospectively included. Patients were randomly divided into the training（ n=672）and internal test set（ n=448）at a ratio of 6∶4. An external test set（ n=128）was established at the Forth Affiliated Hospital of School of Medicine. Deep learning was used for automated segmentation of MRI lesions，followed by radiomic feature extraction and machine learning classification to construct both a US-MRI multimodal model and sequential US-MRI strategy. Diagnostic performance and potential healthcare cost-saving effects were evaluated across strategies. Results:In the internal test set（ n=448），the AI-based sequential US-MRI strategy achieved a F1 score of 0.863 and a diagnostic accuracy of 82.14%，with no significant difference compared to the US-MRI multi-modal model（ P>0.05）. The sequential strategy identified 82 cases（18.30%，82/448）of patients as low-risk true negatives during initial ultrasound screening，suggesting a potential to reduce the need for MRI examinations in future clinical practice. In the external test set（ n=128），the strategy achieved an F1 score of 0.800 and a confirmed diagnosis rate of 85.94%，with a theoretical reduction of 26.56%（34 cases）in MRI utilization while maintaining a diagnostic accuracy rate higher than that of the multi-modal model（82.18%）. Conclusions:The AI-based US-MRI sequential diagnostic strategy demonstrates favorable diagnostic accuracy while offering the potential to optimize MRI utilization. This approach may enhance the efficiency of imaging resource allocation and reduce healthcare burden in the management of ovarian masses.

Objective:To investigate the effect of virtual reality (VR) attention training on cognitive function in patients with depressive episode.Methods:64 patients diagnosed as major depressive disorder and bipolar disorder depressive episodes according to the DSM-5 criteria were recruited. They were randomized into virtual reality training (VRT) group ( n=23), computerized cognitive remediation therapy (CCRT) group ( n=21) and blank control group ( n=20). Prior to the intervention, seven cognitive functions were assessed with the MATRICS Consensus Cognitive Battery (MCCB)-B version in all patients, via information processing speed (IPS), attention/alertness (ATT), working memory, word learning, visual learning (VL), reasoning and problem solving, and social cognition. VRT group and CCRT group were trained for four weeks at a frequency of five days a week, and half an hour for each day training. Blank control group did not receive any treatment related to attention training. After the training, three groups were assessed by the MCCB-A. The differences of the cognitive functions among three groups were explored by the repeated analysis of ANOVA and paired sample ttest. Results:(1) Before the intervention, there were no differences in all cognitive functions (all P>0.05) among three groups. (2) After four-week interventions, the cognition of IPS, ATT and VL in VRT group (56.74±9.68, 56.48±10.22, 57.83±4.16), CCRT group (48.90±9.77, 49.48±9.51, 55.95±5.52) and the blank control group (50.35±7.93, 47.55±7.80, 47.95±9.90) had significant groups×time interactions ( F=14.06, 12.88, 9.39, all P<0.01); simple effect analysis showed that IPS and ATT scores in VRT group were higher than both CCRT group and the blank control group (all P<0.05), while the VL scores in VRT group and CCRT group were both higher than the blank control group (all P<0.01).(3) Cognitive functions in VRT group significantly improved in IPS, ATT, VL and overall domains compared with the baseline ( t=-9.33, -6.00, -5.13, -6.26, all P<0.01). Conclusion:VR attention training may be more beneficial than CCRT attention training to improve the attention among depressive patients.

Objective:To investigate the character and prevalence of cognitive impairment of patients with bipolar Ⅱ depression (BD-Ⅱ).Methods:124 patients diagnosed as bipolar Ⅱ depression according to the DSM-5 criteria and 124 demographically matched healthy subjects were recruited. Seven cognitive functions were assessed with the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) Consensus Cognitive Battery(MCCB) in all participants, including speed of processing (SOP), attention vigilance (AV), working memory (WM), verbal learning (VER), visual learning (VIS), reasoning problem solving (RPS), and social cognition (SC), and the composite. Analysis of covariance was used to test the differences in cognitive function. The number and percentage of cognitive domains impairment which was defined as the cognitive domains scored below standard values by 1, 1.5 and 2 standard deviation (SD) were explored.Results:(1) BD-Ⅱ patients were significantly impaired on seven MCCB domains and the composite scores compared with HC (all P<0.01). Correlation analysis showed that the scores of VER, RPS negatively correlated to the number of episodes ( r=-0.212, P=0.018; r=-0.183, P=0.042); (2) Most healthy control participants were not impaired on any 2 cognitive domains at 1.5 SD (79.84%,99/124) and 2 SD (92.74%,115/124) cut-offs, with the 2.42%-6.45% cognitive impairment at the 1.5 SD cut-off, and 0-4.84% at the 2 SD cut-off accordingly. (3) At the 1.5 SD cut-off, 33.06%,41/124 of the BD-Ⅱ patients were cognitively impaired in two or more domains, while at the 2.0 SD cut-off, 14.52%,18/124 of patients were cognitively impaired. Meanwhile, the incidence of impairment in various cognitive domains was 9.68%-24.19% and 3.23%-15.32%, of which the incidence rate of visual learning impairment was 12.90%, and the incidence rate of impairment in working memory and social cognition was 24.19%. Conclusions:Participants with BD-Ⅱ depression were generally impaired on a greater number of cognitive domains with a higher percentage than the healthy controls, especially on the cognitive domains of working memory, visual learning, and social cognition. And the domains of verbal learning and reasoning problem solving were negatively correlated with the number of episodes.

Objective:To investigate the effect of virtual reality (VR) attention training on cognitive function in patients with depressive episode.Methods:64 patients diagnosed as major depressive disorder and bipolar disorder depressive episodes according to the DSM-5 criteria were recruited. They were randomized into virtual reality training (VRT) group ( n=23), computerized cognitive remediation therapy (CCRT) group ( n=21) and blank control group ( n=20). Prior to the intervention, seven cognitive functions were assessed with the MATRICS Consensus Cognitive Battery (MCCB)-B version in all patients, via information processing speed (IPS), attention/alertness (ATT), working memory, word learning, visual learning (VL), reasoning and problem solving, and social cognition. VRT group and CCRT group were trained for four weeks at a frequency of five days a week, and half an hour for each day training. Blank control group did not receive any treatment related to attention training. After the training, three groups were assessed by the MCCB-A. The differences of the cognitive functions among three groups were explored by the repeated analysis of ANOVA and paired sample ttest. Results:(1) Before the intervention, there were no differences in all cognitive functions (all P>0.05) among three groups. (2) After four-week interventions, the cognition of IPS, ATT and VL in VRT group (56.74±9.68, 56.48±10.22, 57.83±4.16), CCRT group (48.90±9.77, 49.48±9.51, 55.95±5.52) and the blank control group (50.35±7.93, 47.55±7.80, 47.95±9.90) had significant groups×time interactions ( F=14.06, 12.88, 9.39, all P<0.01); simple effect analysis showed that IPS and ATT scores in VRT group were higher than both CCRT group and the blank control group (all P<0.05), while the VL scores in VRT group and CCRT group were both higher than the blank control group (all P<0.01).(3) Cognitive functions in VRT group significantly improved in IPS, ATT, VL and overall domains compared with the baseline ( t=-9.33, -6.00, -5.13, -6.26, all P<0.01). Conclusion:VR attention training may be more beneficial than CCRT attention training to improve the attention among depressive patients.

Objective:To investigate the character and prevalence of cognitive impairment of patients with bipolar Ⅱ depression (BD-Ⅱ).Methods:124 patients diagnosed as bipolar Ⅱ depression according to the DSM-5 criteria and 124 demographically matched healthy subjects were recruited. Seven cognitive functions were assessed with the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) Consensus Cognitive Battery(MCCB) in all participants, including speed of processing (SOP), attention vigilance (AV), working memory (WM), verbal learning (VER), visual learning (VIS), reasoning problem solving (RPS), and social cognition (SC), and the composite. Analysis of covariance was used to test the differences in cognitive function. The number and percentage of cognitive domains impairment which was defined as the cognitive domains scored below standard values by 1, 1.5 and 2 standard deviation (SD) were explored.Results:(1) BD-Ⅱ patients were significantly impaired on seven MCCB domains and the composite scores compared with HC (all P<0.01). Correlation analysis showed that the scores of VER, RPS negatively correlated to the number of episodes ( r=-0.212, P=0.018; r=-0.183, P=0.042); (2) Most healthy control participants were not impaired on any 2 cognitive domains at 1.5 SD (79.84%,99/124) and 2 SD (92.74%,115/124) cut-offs, with the 2.42%-6.45% cognitive impairment at the 1.5 SD cut-off, and 0-4.84% at the 2 SD cut-off accordingly. (3) At the 1.5 SD cut-off, 33.06%,41/124 of the BD-Ⅱ patients were cognitively impaired in two or more domains, while at the 2.0 SD cut-off, 14.52%,18/124 of patients were cognitively impaired. Meanwhile, the incidence of impairment in various cognitive domains was 9.68%-24.19% and 3.23%-15.32%, of which the incidence rate of visual learning impairment was 12.90%, and the incidence rate of impairment in working memory and social cognition was 24.19%. Conclusions:Participants with BD-Ⅱ depression were generally impaired on a greater number of cognitive domains with a higher percentage than the healthy controls, especially on the cognitive domains of working memory, visual learning, and social cognition. And the domains of verbal learning and reasoning problem solving were negatively correlated with the number of episodes.

Objective To study preparation of micronized Shuanghuangpo hydrogel patch and its characteristics of transdermal penetration in vitro. Methods Micronized Shuanghuangpo hydrogel patch was prepared with some macromolecular water-soluble materials as gel base.The content of berberine was determined by HPLC method.Its transdermal penetration in vitro was determined according to the method of Chinese Pharmacopoeia 2010 edition. The rat skin penetration test in vitro was performed by modified Franz diffusion cell method. Results The hydrogel patch had constant content of berberine. Its release property in vitro conformed to Higuchi equation. The penetration of berberine in the hydrogel patch through the rat skin followed zero-order dynamics in 12 h.Its release rate was 7.934μg??(cm2)-1??h1/2 and percutaneous rate was 0.571μg??(cm2)-1??h-1. Conclusion The micronized Shuanghuangpo hydrogel patch is a new transdermal agent with sustained release property.