ObjectiveTo explore the clinical application value of CIE 1976 L*a*b*（Lab） color space in the differential diagnosis of early esophageal cancer and non-cancerous lesions. MethodsWe selected the endoscopic images of patients with esophageal lesions confirmed by pathology who underwent white light imaging endoscopy （WLI） and narrow band imaging endoscopy （NBI）. Five regions of interest （ROI） were selected respectively from the mucosa of the lesion area and the mucosa of the surrounding normal area for labeling. The Lab color space parameters were extracted and counted， and the color difference values（ΔE*）were calculated. ResultsA total of 213 eligible patients were included for analysis in the study. In WLI and NBI modes， there were differences in mucosal color between the early esophageal cancer group and the non-cancer group （P<0.05）. Compared with WLI mode， NBI mode could significantly increase the color difference between early esophageal cancer and non-cancerous lesions （P<0.05）. The lightness component value （L* value） of the early esophageal cancer lesion area was lower than that of the non-cancerous lesion area， and this color difference was more significant in the NBI mode （P<0.05）. In WLI mode， there was no significant difference in yellow-blue component value （b* value） between the mucosa of early esophageal cancer and non-cancerous lesions. However， in the NBI mode， the b* value of the mucosa in the non-cancerous lesion area was higher than that in the early esophageal cancer lesion area （P<0.05）. On the red-green axis， the mucosa of the early esophageal cancer and non-cancerous lesions was red in WLI mode and green in NBI mode. There was no significant difference in red-green component value （a* value） between the two groups. ConclusionThere are color differences between early esophageal cancer and non-cancerous lesions under WLI and NBI. The color of early esophageal cancer is darker under WLI， and the color of non-cancerous lesions is yellower under non-magnified NBI mode. Lab color space is helpful to identify early esophageal cancer and non-cancerous lesions.

ObjectiveTo explore the latent profiles of sleep patterns and their association with gastrointestinal heat accumulation syndrome in preschool children in Beijing. MethodsA cross-sectional study design was adopted， and preschool children were recruited from 57 kindergartens in Beijing using convenience sampling. Sleep characteristics and gastrointestinal heat accumulation syndrome was assessed using the Chinese version of the Sleep Disturbance Scale for Children （SDSC） and Self-rating Diagnostic Scale for Gastrointestinal Heat Retention Syndrome in children （GHRS）. Latent profile analysis （LPA） was applied to identify distinct sleep-related subgroups， and multivariate logistic regression was used to analyze the association between different sleep profiles and gastrointestinal heat accumulation syndrome. ResultsA total of 6878 preschool children were included in the study. Among them， 3.46% of the children （238/6878） had sleep disorders， while 54.42% （3743/6878） had gastrointestinal heat accumulation. Compared to children without gastrointestinal heat accumulation syndrome， those with this syndrome exhibited a higher prevalence of sleep disorders， higher SDSC scores， higher T-transformed total scores， and higher scores across all dimensions， including disorders of initiating sleep， disorders of maintaining sleep， sleep hyperhidrosis， sleep breathing disorders， parasomnias， nonrestorative sleep and excessive somnolence （P＜0.01）. Latent profile analysis identified four distinct sleep characteristic profiles， which were the mixed high-symptom group （32.39%， 2228/6878）， sleep initiation and maintenance difficulty symptom group （49.07%， 3375/6878）， low-symptom group （14.19%， 976/6878）， and the group with no significant symptoms （4.35%， 299/6878）. Multivariate logistic regression analysis demonstrated that the mixed high-symptom group， sleep initiation and maintenance difficulty symptom group， and low-symptom group were significantly associated with gastrointestinal heat accumulation syndrome. Among them， the mixed high-symptom group had the most significant association with gastrointestinal heat accumulation （OR=12.657， P＜0.001）， followed by the sleep initiation and maintenance difficulty symptom group （OR=4.964， P＜0.001）. ConclusionSleep characteristics in preschool children mainly present as composite patterns with different symptom combinations. Moreover， the sleep latent profiles of the mixed high-symptom group， the sleep initiation and maintenance difficulty symptom group， and the low-symptom group were significantly associated with gastrointestinal heat accumulation syndrome.

OBJECTIVES: To investigate the structural changes of rat thoracic aorta and changes in expression levels of Bmal1 and cyclins in thoracic aorta endothelial cells following heat stress. METHODS: Twenty male SD rats were randomized equally into control group and heat stress group. After exposure to 32 ℃ for 2 weeks in the latter group, the rats were examined for histopathological changes and Bmal1 expression in the thoracic aorta using HE staining and immunohistochemistry. In the cell experiments, cultured rat thoracic aortic endothelial cells (RTAECs) were incubated at 40 ℃ for 12 h with or without prior transfection with a Bmal1-specific small interfering RNA (si-Bmal1) or a negative sequence. In both rat thoracic aorta and RTAECs, the expressions of Bmal1, the cell cycle proteins CDK1, CDK4, CDK6, and cyclin B1, and apoptosis-related proteins Bax and Bcl-2 were detected using Western blotting. TUNEL staining was used to detect cell apoptosis in rat thoracic aorta, and the changes in cell cycle distribution and apoptosis in RTAECs were analyzed with flow cytometry. RESULTS: Compared with the control rats, the rats exposed to heat stress showed significantly increased blood pressures and lowered heart rate with elastic fiber disruption and increased expressions of Bmal1, cyclin B1 and CDK1 in the thoracic aorta (P<0.05). In cultured RTAECs, heat stress caused significant increase of Bmal1, cyclin B1 and CDK1 protein expression levels, which were obviously lowered in cells with prior si-Bmal1 transfection. Bmal1 knockdown also inhibited heat stress-induced increase of apoptosis in RTAECs as evidenced by decreased expression of Bax and increased expression of Bcl-2. CONCLUSIONS Heat stress upregulates Bmal1 expression and causes alterations in expressions of cyclins to trigger apoptosis of rat thoracic aorta endothelial cells, which can be partly alleviated by suppressing Bmal1 expression.
Animals ; ARNTL Transcription Factors/genetics* ; Male ; Aorta, Thoracic/metabolism* ; Rats ; Rats, Sprague-Dawley ; Endothelial Cells/metabolism* ; Apoptosis ; Cells, Cultured ; Heat-Shock Response ; Cyclin B1/metabolism* ; CDC2 Protein Kinase/metabolism* ; Cyclins/metabolism* ; RNA, Small Interfering ; bcl-2-Associated X Protein/metabolism*

Accurately obtaining the crop quantity and density is not only crucial for the demand-based input of water and fertilizer in the field but also vital for ensuring the yield and quality of crops. Aerial photography by unmanned aerial vehicles (UAVs) can quickly acquire the distribution image information of crops over a large area. However, the accurate recognition of a single type of dense targets is a huge challenge for most recognition algorithms. Taking banana seedlings as an example in this study, we captured the images of banana plantations by UAVs from high altitudes to explore an efficient recognition method for dense targets. We proposed a strategy of "cut-recognition-stitch" and constructed a counting method based on the improved Faster R-CNN algorithm. First, the images containing highly dense targets were cropped into a large number of image tiles according to different sizes (simulating different flight altitudes), and the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm was adopted to improve the image quality. A banana seedling dataset containing 36 000 image tiles was constructed. Then, the Faster R-CNN network with optimized parameters was used to train the banana seedling recognition model. Finally, the recognition results were reversely stitched together, and a boundary deduplication algorithm was designed to correct the final counting results to reduce the repeated recognition caused by image cropping. The results show that the recognition accuracy of the Faster R-CNN with optimized parameters for banana image datasets of different sizes can reach up to 0.99 at most. The deduplication algorithm can reduce the average counting error for the original aerial images from 1.60% to 0.60%, and the average counting accuracy of banana seedlings reaches 99.4%. The proposed method effectively addresses the challenge of recognizing dense small objects in high-resolution aerial images, providing an efficient and reliable technical solution for intelligent crop density monitoring in precision agriculture.
Musa/growth & development* ; Crops, Agricultural/growth & development* ; Algorithms ; Neural Networks, Computer ; Unmanned Aerial Devices ; Seedlings/growth & development* ; Image Processing, Computer-Assisted/methods* ; Photography ; Agriculture/methods*

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

In the clinical stage, suspected hemolytic plasma may cause hemolysis illness, manifesting as symptoms such as heart failure, severe anemia, etc. Applying a deep learning method to plasma images significantly improves recognition accuracy, so that this paper proposes a plasma quality detection model based on improved "You Only Look Once" 5th version (YOLOv5). Then the model presented in this paper and the evaluation system ‌were introduced‌ into the plasma datasets, and ‌the average accuracy of the final classification reached 98.7%‌. The results of this paper's experiment were obtained through the combination of several key algorithm modules including‌ omni-dimensional dynamic convolution, pooling with separable kernel attention, residual bi-fusion feature pyramid network, ‌and‌ re-parameterization convolution. The method of this paper‌ obtains the feature information of spatial mapping efficiently, and enhances the average recognition accuracy of plasma quality detection. This paper presents a high-efficiency detection method for plasma images, aiming to provide a practical approach to prevent hemolysis illnesses caused by external factors.
Algorithms ; Humans ; Hemolysis ; Plasma ; Deep Learning ; Image Processing, Computer-Assisted/methods*

To verify whether the two B-cell epitopes Pep1 and Pep4 in the FAdV-4 WZ fiber can be used as candidate epitopes for multivalent epitope vaccines,epitopes Pep1 and Pep4 were tandemly linked with the chicken infectious bronchitis virus strain M41 S1 protein gene in different patterns,and a recombinant fusion plasmid was constructed and expressed in E.coli BL21(DE3).It was confirmed by Western blot and ELISA tests that all four expressed fusion proteins reacted specific-ally with anti-M41 whole virus serum and WZ strain anti-Fiber 2-knob protein serum.After purifi-cation and immunization of BALB/c mice,specific antibodies against the peptide epitopes were de-tected in mouse sera.The results showed that the Pep4 epitope induced a stronger immune re-sponse than the Pep1 epitope.When Pep1 was connected with the amino and carboxyl termini of the fusion protein,respectively,both resulted in the production of the same level of anti-Pep1 anti-bodies in the immunized animals,whereas when Pep4 was connected with the carboxyl terminus of the fusion protein,the immunized animals produced a higher level of anti-Pep4-specific antibodies.This research indicates that the B cell epitopes Pep1 and Pep4 of the reactive WZ strain Fiber 2,when conjugated with proteins to form fusion proteins,can enhance the immunogenicity of Pep1 and Pep4 without affecting the antigenicity of the carrier protein.This study provides technical support and serves as a reference for the design and development of a multivalent epitope vaccine for FAdV-4.

The number of patients with eye diseases in China is enormous,and the negative effects of these conditions,such as impaired visual function,psychological burdens,and restricted social participation,are becoming increasingly severe.Due to the limited and unevenly distributed ophthalmic resources,and the significant limitations of traditional diagnostic and therapeutic approaches in terms of accuracy and efficiency,there is an urgent need for more sensitive and efficient modalities.With the rapid advancement of artificial intelligence technology,ophthalmic diagnosis has entered a new stage of intelligent transformation.Facial photos,as a noninvasive and convenient medium,show unique advantages in eye disease diagnosis.Artificial intelligence systems based on facial photo analysis have been applied to the screening and diagnosis of conditions such as myopia,strabismus,ptosis,and thyroid eye disease,showing promising results.This review introduces the workflow of intelligent diagnosis for ocular diseases based on facial photographs,with a focus on recapitulating relevant research findings both domestically and internationally in recent years.It summarizes the innovative features and application advantages of intelligent diagnosis systems for eye diseases based on facial photos,analyzes the current technical bottlenecks and challenges in application,proposes corresponding countermeasures,and discusses future development directions,aiming to provide references and new insights for the intelligent screening and diagnosis of eye diseases.

Objective To explore the correlation between serum insulin-like growth factor1(IGF-1),insulin-like growth factor binding protein3(IGFBP-3),bone-specific alkaline phosphatase(BAP)and abundance of intestinal flora in children with idiopathic short stature(ISS).Methods A total of 86 children with ISS who were treated with short-acting recombinant growth hormone for the first time in Department of Endocrinology of Baoding Hospital,Beijing Children's Hospital,Capital Medical University,were enrolled from July 2021 to June 2023.Before treatment and at 6 months after treatment,physical indexes(height,weight)and bone age indexes[bone age index(BAI),bone age difference(BAD)]were detected with the same instrument at the same place.The fasting venous blood was collected to detect levels of serum IGF-1,IGFBP-3 and BAP by enzyme-linked immunosorbent assay.The abundances of intestinal floras were monitored by 16s rDNA high-throughput sequencing technology combined with bioinformation analysis technology.A total of 40 healthy children during the same period were enrolled as control group,and differences in different indexes within the same groups or between the two groups were compared by independent sample t test or paired t test.The correlation between abundances of intestinal floras and levels of serum IGF-1,IGFBP-3 and BAP was analyzed by Pearson correlation analysis.Results Compared with the control group.the height,weight,BAD,BAI,levels of serum IGF,IGFBP-3 and BAP,Chaol,Shannon and Pielou e of ISS patients decreased before treatment,and the differences were statistically significant(t=7.119～44.143,all P<0.05).After treatment,height,weight,BAD,BAI,levels of serum IGF,IGFBP-3 and BAP,Chaol,Shannon and Pielou e were increased in ISS children(t=3.145～39.188),the above indexes were still lower than those in control group(t=2.389～69.893),the differences were statistically significant(all P<0.05),respectively.In intestinal floras,dominant bacterial groups at phylum level were Firmicutes,Bacteroidetes,Actinobacteria and Proteobacteria,while dominant bacterial groups at genus level were Blautia,Bifidobacterium,Faecalibacterium and Bacteroides.Compared with the control group,before treatment,the abundances of Actinobacteria and Proteobacteria increased in the ISS group,while the abundances of Bacteroidetes,Bifidobacteria and Bacteroides decreased,and the differences were statistically significant(t=25.224～46.345,all P<0.05).After treatment,abundances of Bacteroidetes,Bifidobacterium and Bacteroides in the ISS group increased compared to before treatment,while abundance of Proteobacteria decreased,and the differences were statistically significant(t=10.055～39.245,all P<0.05).The abundances of Bacteroidetes,Bifidobacterium and Bacteroides in ISS group were lower than those in control group,while abundance of Proteobacteria was higher than that in control group,and the differences were statistically significant(t=6.290～20.540,all P<0.05).The abundances of Bacteroidetes,Bifidobacterium and Bacteroides were positively correlated with levels of serum IGF-1,IGFBP-3 and BAP(r=0.300～0.759,all P<0.05).Conclusion The abundances of intestinal floras are decreased,and the structure is abnormal in children with ISS.The abundances of beneficial bacteria in intestinal floras are related to levels of serum IGF-1,IGFBP-3 and BAP.In clinical treatment of ISS,combined treatment with intestinal flora disorders may be conducted.

Objective To explore curative effect of two thrombolytic regimens in the treatment of ultra-early acute cerebral infarction(ACI)of different age groups and different thrombolytic time windows.Methods A total of 166 patients with ACI were enrolled,with interval from onset to treatment≤4.5 h.According to different thrombolytic regiments,patients were divided into the recombinant tissue plasminogen activator(rt-PA)group(104 cases,intravenous thrombolysis with 0.9 mg/kg rt-PA)and the tenecteplase(TNK)group(62 cases,intravenous thrombolysis with 0.25 mg/kg TNK).According to different age groups and different thrombolytic time windows,patients were divided into the high age group(≥70 years),the low age group(＜70 years),the short time window group(＜3 h)and the long time window group(3-4.5 h).After thrombolysis,laboratory indexes[neutrophil(NEU),lymphocyte(LYM)],scores of National Institutes of Health Stroke Scale(NIHSS),modified Rankin scale(mRS)and incidence of adverse events were compared in different groups by factorial analysis method.Results There was no significant difference in any index between the rt-PA group and the TNK group(P＞0.05).Compared with the high age group at 7 d after thrombolysis,NIHSS score,NEU,neutrophil to lymphocyte ratio(NLR)and high sensitive C-reactive protein(hs-CRP)were lower,while LYM and albumin(ALB)were higher in the low age group(P＜0.05).Compared with the long time window group,NEU and NLR were lower,while LYM was higher in the short time window group(P＜0.05).There was no significant difference in total incidence of adverse events within 14 d of thrombolysis between the different age groups and the different thrombolytic time window groups(P＞0.05).Conclusion For patients with ultra-early ACI,0.9 mg/kg rt-PA and 0.25 mg/kg TNK have comparable efficacy in intravenous thrombolysis,and both have better effects when patient age is＜70 years old and the thrombolysis time window is＜3 hours.