BACKGROUND: Over 20 million colonoscopies are performed in China annually. An automatic clinical decision support system (CDSS) with accurate semantic recognition of colonoscopy reports and guideline-based is helpful to relieve the increasing medical burden and standardize the healthcare. In this study, the CDSS was built under a hierarchical-label interpretable classification framework, trained by a state-of-the-art transformer-based model, and validated in a multi-center style. METHODS: We conducted stratified sampling on a previously established dataset containing 302,965 electronic colonoscopy reports with pathology, identified 2041 patients' records representative of overall features, and randomly divided into the training and testing sets (7:3). A total of five main labels and 22 sublabels were applied to annotate each record on a network platform, and the data were trained respectively by three pre-training models on Chinese corpus website, including bidirectional encoder representations from transformers (BERT)-base-Chinese (BC), the BERT-wwm-ext-Chinese (BWEC), and ernie-3.0-base-zh (E3BZ). The performance of trained models was subsequently compared with a randomly initialized model, and the preferred model was selected. Model fine-tuning was applied to further enhance the capacity. The system was validated in five other hospitals with 3177 consecutive colonoscopy cases. RESULTS: The E3BZ pre-trained model exhibited the best performance, with a 90.18% accuracy and a 69.14% Macro-F1 score overall. The model achieved 100% accuracy in identifying cancer cases and 99.16% for normal cases. In external validation, the model exhibited favorable consistency and good performance among five hospitals. CONCLUSIONS The novel CDSS possesses high-level semantic recognition of colonoscopy reports, provides appropriate recommendations, and holds the potential to be a powerful tool for physicians and patients. The hierarchical multi-label strategy and pre-training method should be amendable to manage more medical text in the future.
Humans ; Colonoscopy/methods* ; Deep Learning ; Decision Support Systems, Clinical ; Female ; Male

To address the problem of data silos in dental specialties caused by equipment heterogeneity, this study developed an Intelligent Internet of Things (IoT)-enabled dental chair platform (hereinafter referred to as the intelligent platform) based on the concept of medical-engineering integration. The platform adopts a three-tier chair-domain interconnection architecture: the bottom tier integrates multi-source sensors and standardized interfaces for automated data acquisition and linkage with hospital information systems; the middle tier provides clinic-level management and remote teaching collaboration; and the top tier employs a blockchain-based secure cloud database for resource allocation and data management. Clinical validation at The Affiliated Stomatological Hospital of Nanjing Medical University demonstrated that, compared with a control group from the same period in 2023, the trial group achieved a 38.0% increase in average daily patient visits (80.6±6.8 vs. 58.4±5.2, t=15.16, P<0.001), a 24.6% reduction in average treatment time [(36.1±6.3) min vs. (47.9±8.5) min, t=7.72, P<0.001], a 39.2% reduction in waiting time [23.3 (16.5, 30.1) min vs. 38.3 (28.3, 48.3) min, U=32.00, P<0.001], a 30.4% reduction in equipment idle rate [8.7% (5.1%, 12.3%) vs. 12.5% (7.4%, 17.6%), U=251.00, P=0.003], and an increase in patient satisfaction from 88.2% (1 519/1 723) to 94.3% (2 186/2 318) ( t=7.26, P<0.001). User research confirmed that the functions most favored by clinicians and patients were "dental chair parameter updating and clinical data integration" [74.7% (80/107)] and "chairside display of diagnostic images" [76.8% (119/155)], respectively. Looking forward, the intelligent platform has the potential to integrate artificial intelligence-assisted diagnosis and 5G-enabled multicenter collaboration to further expand its clinical applications and accelerate the digital transformation of dental healthcare.

Objective To explore the differences in neural activity between patients with anorexia nervosa(AN)and healthy controls(HC),as well as the association between these differences and symptoms in AN patients,using activation likelihood estimation(ALE)meta-analysis.Methods The literature search covered the period from 2000 to March 2025.From a pool of 588 identified studies,4 studies focusing on the neural activity differences between AN patients and HC were selected for inclusion.These studies comprised 106 participants and 21 sets of coordinates.The ALE meta-analysis method was employed,and the GingerALE software was used to systematically analyze the reported brain region changes and their peak coordinates,aiming to investigate the differences in brain functional activity between AN patients and HC.Results Compared to the HC group,AN group showed significantly enhanced activation in the left parahippocampal gyrus/amygdala(ALE value=0.39×10-2),right parahippocampal gyrus/amygdala(ALE value=0.39×10-2)and suboccipital gyrus(ALE value=0.39×10-2),along with a significant reduction in activation in Brodmann area 17(ALE value=0.61×10-2)(P<0.01,FWE corrected).Conclusion Key brain regions in AN patients including the parahippocampal gyrus,inferior occipital gyrus,and amygdala demonstrate significant functional activation abnormalities.

To address the problem of data silos in dental specialties caused by equipment heterogeneity, this study developed an Intelligent Internet of Things (IoT)-enabled dental chair platform (hereinafter referred to as the intelligent platform) based on the concept of medical-engineering integration. The platform adopts a three-tier chair-domain interconnection architecture: the bottom tier integrates multi-source sensors and standardized interfaces for automated data acquisition and linkage with hospital information systems; the middle tier provides clinic-level management and remote teaching collaboration; and the top tier employs a blockchain-based secure cloud database for resource allocation and data management. Clinical validation at The Affiliated Stomatological Hospital of Nanjing Medical University demonstrated that, compared with a control group from the same period in 2023, the trial group achieved a 38.0% increase in average daily patient visits (80.6±6.8 vs. 58.4±5.2, t=15.16, P<0.001), a 24.6% reduction in average treatment time [(36.1±6.3) min vs. (47.9±8.5) min, t=7.72, P<0.001], a 39.2% reduction in waiting time [23.3 (16.5, 30.1) min vs. 38.3 (28.3, 48.3) min, U=32.00, P<0.001], a 30.4% reduction in equipment idle rate [8.7% (5.1%, 12.3%) vs. 12.5% (7.4%, 17.6%), U=251.00, P=0.003], and an increase in patient satisfaction from 88.2% (1 519/1 723) to 94.3% (2 186/2 318) ( t=7.26, P<0.001). User research confirmed that the functions most favored by clinicians and patients were "dental chair parameter updating and clinical data integration" [74.7% (80/107)] and "chairside display of diagnostic images" [76.8% (119/155)], respectively. Looking forward, the intelligent platform has the potential to integrate artificial intelligence-assisted diagnosis and 5G-enabled multicenter collaboration to further expand its clinical applications and accelerate the digital transformation of dental healthcare.

Objective To explore the differences in neural activity between patients with anorexia nervosa(AN)and healthy controls(HC),as well as the association between these differences and symptoms in AN patients,using activation likelihood estimation(ALE)meta-analysis.Methods The literature search covered the period from 2000 to March 2025.From a pool of 588 identified studies,4 studies focusing on the neural activity differences between AN patients and HC were selected for inclusion.These studies comprised 106 participants and 21 sets of coordinates.The ALE meta-analysis method was employed,and the GingerALE software was used to systematically analyze the reported brain region changes and their peak coordinates,aiming to investigate the differences in brain functional activity between AN patients and HC.Results Compared to the HC group,AN group showed significantly enhanced activation in the left parahippocampal gyrus/amygdala(ALE value=0.39×10-2),right parahippocampal gyrus/amygdala(ALE value=0.39×10-2)and suboccipital gyrus(ALE value=0.39×10-2),along with a significant reduction in activation in Brodmann area 17(ALE value=0.61×10-2)(P<0.01,FWE corrected).Conclusion Key brain regions in AN patients including the parahippocampal gyrus,inferior occipital gyrus,and amygdala demonstrate significant functional activation abnormalities.

The prefrontal cortex and hippocampus may support sequential working memory beyond episodic memory and spatial navigation. This stereoelectroencephalography (SEEG) study investigated how the dorsolateral prefrontal cortex (DLPFC) interacts with the hippocampus in the online processing of sequential information. Twenty patients with epilepsy (eight women, age 27.6 ± 8.2 years) completed a line ordering task with SEEG recordings over the DLPFC and the hippocampus. Participants showed longer thinking times and more recall errors when asked to arrange random lines clockwise (random trials) than to maintain ordered lines (ordered trials) before recalling the orientation of a particular line. First, the ordering-related increase in thinking time and recall error was associated with a transient theta power increase in the hippocampus and a sustained theta power increase in the DLPFC (3-10 Hz). In particular, the hippocampal theta power increase correlated with the memory precision of line orientation. Second, theta phase coherences between the DLPFC and hippocampus were enhanced for ordering, especially for more precisely memorized lines. Third, the theta band DLPFC → hippocampus influence was selectively enhanced for ordering, especially for more precisely memorized lines. This study suggests that theta oscillations may support DLPFC-hippocampal interactions in the online processing of sequential information.
Adult ; Female ; Humans ; Young Adult ; Epilepsy ; Hippocampus ; Memory, Short-Term ; Mental Recall ; Prefrontal Cortex ; Theta Rhythm ; Male

Objective:To explore the protective mechanism of tea polyphenols (TP) on mouse oral cancer.Methods:A total of 50 mice were divided into control group, model group, TP group, Selisistat group, TP+ Selisistat group, with 10 mice in each group. The control group was gavaged with physiological saline, while the model group, TP group, Selisistat group, and TP+ Selisistat group were gavaged with 300 mg/L 4-NQO to establish a mouse oral cancer model. Physiological saline, 200 mg/kg TP, 0.01 mg/kg Selisistat, and 200 mg/kg TP+ 0.01 mg/kg Selisistat were gavaged respectively. The weight changes of each group of mice were compared; HE staining was used to observe the morphology of mouse oral tumor tissue; Enzyme linked immunosorbent assay was used to detect the levels of malondialdehyde (MDA) and superoxide dismutase (SOD) in serum; Immunoblotting and immunohistochemistry were used to detect the expression of silencing information regulatory factor (Sirt1) and nuclear factor E2 related factor 2 (Nrf2) proteins in mouse oral tissues.Results:Compared with the control group, the model group mice had a decrease in body weight [(23.19±1.36)g], a decrease in serum SOD level [(91.64±8.75)U/ml], an increase in MDA level [(5.18±0.46)nmol/ml], a decrease in Sirt1 (0.38±0.05) and Nrf2 (0.36±0.05) protein expression in oral tissue, and an increase in Nrf2 acetylation level (0.84±0.11) (all P<0.05). Compared with the model group, the TP group mice had an increase in body weight [(25.28±1.25)g], elevated serum SOD levels [(121.24±10.68)U/ml], decreased MDA levels [(3.89±0.42)nmol/ml], increased expression of Sirt1 (0.61±0.09) and Nrf2 (0.58±0.06) proteins in oral tissue, and decreased Nrf2 protein acetylation levels (0.39±0.05); The Selisistat group mice showed a decrease in body weight [(21.41±1.07)g], a decrease in serum SOD levels [(72.16±7.43)U/ml], an increase in MDA levels [(5.87±0.41)nmol/ml], a decrease in Sirt1 (0.23±0.04) and Nrf2 protein (0.24±0.03) expression in oral tissue, and an increase in Nrf2 acetylation levels (1.12±0.14) ( P<0.05). The body weight [(23.32±1.27)g], serum SOD levels [(92.58±8.13)U/ml], and oral Sirt1 (0.41±0.06) and Nrf2 (0.38±0.05) protein expression in the TP+ Selisistat group mice were higher than those in the Selisistat group, while MDA [(5.11±0.38)nmol/ml] and Nrf2 acetylation levels (0.82±0.09) were lower than those in the Selisistat group (all P<0.05). Conclusions:Tea polyphenols can alleviate oral tissue damage and alleviate oxidative stress in mice with oral cancer, and their mechanism may be related to the upregulation of the Sirt1/Nrf2 pathway.

Objective::This study aimed to determine the likelihood of gestational diabetes mellitus (GDM) in subsequent pregnancy among women without GDM history and to identify risk factors for GDM in subsequent pregnancy.Methods::This retrospective cohort study involved participants who delivered twice in same hospital of 18 research centers when delivered the second baby from January 2018 to December 2018. Finally 6204 women were enrolled and 5180 women without GDM history were analyzed further. Women were categorized as non-GDM or GDM based on the blood glucose values of the subsequent pregnancy, and the characteristics and GDM risk of these groups were compared. A univariate analysis of potential risk factors was performed using the Chi-squared test and/or t-test for qualitative or quantitative variables, respectively. Associations with P values <0.1 were chosen to be included in the multivariate binary logistic regression model. Results::In primary analysis of 6204 women, the incidence of GDM in subsequent pregnancy is 48.9% (490/1002) in women with GDM history and 16.1% (835/5202) in women without GDM history. In a further analysis for 5180 women without GDM at index pregnancy, compared with the non-GDM group, the GDM group had a significantly higher age, prepregnancy body mass index, and blood glucose value at each oral glucose tolerance test (OGTT) timepoint (fasting, 1 h and 2 h) during the index and subsequent pregnancies, as well as higher weight retention during the interval between the two pregnancies ( P<0.001). Age above 35 years in subsequent pregnancy (odds ratio ( OR)=1.540, 95% confidence interval ( CI) = 1.257-1.886, P<0.001), macrosomia in index pregnancy ( OR=1.749, 95% CI=1.277-2.395, P=0.001), OGTT blood glucose values in index pregnancy (fasting, OR=2.487, 95% CI=1.883-3.285, P<0.001; 1 h, OR=1.142, 95% CI=1.051-1.241, P=0.002; 2 h, OR=1.290, 95% CI=1.162-1.432, P<0.001) and weight retention ( OR=1.052, 95% CI=1.035-1.068, P<0.001) were independent risk factors for GDM in subsequent pregnancy. Conclusion::For women without GDM history, GDM risk factors including age, macrosomia history, OGTT value, and weight retention, these can be evaluated before a subsequent pregnancy. Early warning and interventions are needed for women at high risk.

Objective::This study aimed to determine the likelihood of gestational diabetes mellitus (GDM) in subsequent pregnancy among women without GDM history and to identify risk factors for GDM in subsequent pregnancy.Methods::This retrospective cohort study involved participants who delivered twice in same hospital of 18 research centers when delivered the second baby from January 2018 to December 2018. Finally 6204 women were enrolled and 5180 women without GDM history were analyzed further. Women were categorized as non-GDM or GDM based on the blood glucose values of the subsequent pregnancy, and the characteristics and GDM risk of these groups were compared. A univariate analysis of potential risk factors was performed using the Chi-squared test and/or t-test for qualitative or quantitative variables, respectively. Associations with P values <0.1 were chosen to be included in the multivariate binary logistic regression model. Results::In primary analysis of 6204 women, the incidence of GDM in subsequent pregnancy is 48.9% (490/1002) in women with GDM history and 16.1% (835/5202) in women without GDM history. In a further analysis for 5180 women without GDM at index pregnancy, compared with the non-GDM group, the GDM group had a significantly higher age, prepregnancy body mass index, and blood glucose value at each oral glucose tolerance test (OGTT) timepoint (fasting, 1 h and 2 h) during the index and subsequent pregnancies, as well as higher weight retention during the interval between the two pregnancies ( P<0.001). Age above 35 years in subsequent pregnancy (odds ratio ( OR)=1.540, 95% confidence interval ( CI) = 1.257-1.886, P<0.001), macrosomia in index pregnancy ( OR=1.749, 95% CI=1.277-2.395, P=0.001), OGTT blood glucose values in index pregnancy (fasting, OR=2.487, 95% CI=1.883-3.285, P<0.001; 1 h, OR=1.142, 95% CI=1.051-1.241, P=0.002; 2 h, OR=1.290, 95% CI=1.162-1.432, P<0.001) and weight retention ( OR=1.052, 95% CI=1.035-1.068, P<0.001) were independent risk factors for GDM in subsequent pregnancy. Conclusion::For women without GDM history, GDM risk factors including age, macrosomia history, OGTT value, and weight retention, these can be evaluated before a subsequent pregnancy. Early warning and interventions are needed for women at high risk.

OBJECTIVE: To investigate the effect of environmental estrogen bisphenol A (BPA) exposure on apoptosis of mouse ovarian preantral follicular granulosa cells and ovarian development and explore the underlying mechanism. METHODS: Mouse ovarian preantral follicular granulosa cells were isolated from female ICR mice at postnatal day (PND) 10 and cultured RESULTS: Compared with the control cells group, the isolated cells exposed to a low concentration of BPA (50 μmol/L) showed a significantly lowered apoptosis rate, increased mitochondrial membrane potential, and enhanced cellular proliferation ( CONCLUSIONS BPA can concentration-dependently regulate the function of ovarian preantral follicular granulosa cells in mice and potentially affects both the pregnant mice and the offspring female mice in light of early ovarian development.
Animals ; Apoptosis ; Benzhydryl Compounds ; Female ; Granulosa Cells ; Mice ; Mice, Inbred ICR ; Ovarian Follicle ; Phenols ; Pregnancy