Objective:To develop a multi-task artificial intelligence-assisted system for small bowel capsule endoscopy (SBCE) based on various Transformer neural network architectures. The system integrates lesion recognition, cumulative time statistics, and progress bar marking functions to enhance the efficiency and accuracy of endoscopic image interpretation while effectively reducing missed diagnoses. Methods:A dataset comprising 12 annotated categories of images captured by three different brands of capsule endoscopy devices was collected. Transfer learning and fine-tuning were conducted on five pre-trained Transformer models. Performance metrics, including accuracy, sensitivity, specificity, and recognition speed, were evaluated to select the best-performing model. The optimal model was converted from PyTorch to Open Neural Network Exchange (ONNX) format. Using OpenCV and MMCV tools, a multi-task SBCE-assisted reading system was developed. Results:A total of 34,799 images were included in the study. The best-performing model, FocalNet, achieved a weighted average sensitivity of 85.69%, specificity of 98.58%, accuracy of 85.69%, and an AUC of 0.98 across all categories. Its diagnostic accuracy outperformed junior physicians (χ²=17.26, p<0.05) and showed no statistical difference compared to senior physicians (χ²=0.0716, p>0.05). The multi-task AI-assisted reading system, "FocalCE-Master", developed based on FocalNet, achieved a diagnostic speed of 592.40 frames per second, significantly faster than endoscopists. By integrating cumulative time bar charts with progress bar marking functionality, the system enables rapid localization and review of lesions, effectively streamlining the diagnostic workflow of SBCE. Conclusion:The multi-task SBCE-assisted reading system developed using Transformer networks demonstrated rapid and accurate classification of various small bowel lesions. It holds significant potential in enhancing diagnostic efficiency and image review speed for endoscopists. However, the AI system has not yet been validated in prospective clinical trials, and further real-world studies are needed to confirm its clinical applicability.
Background: There are many risk factors related to early death after surgery among patients with acute Stanford type A aortic dissection (ATAAD) that have been analyzed in previous studies, but no evidence-based study has been conducted to confirm these risk factors.Aims: The aims of this study were to investigate risk factors for early death after surgery in patients with ATAAD via systematic review and meta-analysis and assess evidence-based strategies for preventing adverse events.Methods: The protocol for this study was prospectively registered with PROSPERO (CRD 42022332772). The authors systematically searched PubMed, Ovid, Scopus, Web of Science and Cochrane Library following the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines from database construction to May 2021. Studies that met the selection criteria were determined by two independent researchers, and the odds ratios (ORs) and 95% confidence intervals (95% CIs) were reported for the risk factors and were pooled using Stata 15.0.Results: A total of 23 studies including 5510 patients met the inclusion criteria, and 10 risk factors were analyzed in this meta-analysis. The preoperative risk factors for early death after surgery in patients with ATAAD were age [(OR: 1.03, 95% CI (1.01, 1.06)], male sex [(OR: 1.43, 95% CI (1.06, 1.92)], shock [(OR: 1.91, 95% CI (1.06, 3.45)], malperfusion [(OR: 3.45, 95% CI (2.24, 5.31)] and cardiac tamponade [(OR: 3.89, 95% CI (1.17, 12.98)]. Conclusion: Patients with ATAAD who have an older age, male sex, shock, malperfusion and cardiac tamponade have a higher risk for early death after surgery. However, more highly homogenous studies are needed to demonstrate these results. Clinical staff should pay more attention to these factors and take individual actions to reduce mortality after surgery in patients with ATAAD.
Objective:To study and discuss the formation mechanism of Rh null blood group in a case from the perspective of genetics, and to study the Rh blood group genes of her family members. Methods:The Rh blood type phenotype of the proband seeking medical treatment at Department of Internal Medicine in Taizhou People’s Hospital was detected by blood type serology. RhCE genotyping as performed by fluorescence PCR. RhD, RhCE and RhAG exon sequencing were performed using Sanger sequencing. And then the Rh null formation mechanism of the proband was analyzed. For comparison, the Rh blood type phenotype of the proband’s sister and son was detected by serology, and RhCE genotyping, RhD, RhCE and RhAG exon sequencing were performed. Results:The genotype of the proband was CcDEe, and the RhAG exon sequencing showed a homozygous frameshift mutation, with the mutation site at Exon5, nucleotide changed as c.732delC, and amino acid change as p.Phe245Serfs*16. The serological results, genotyping, RhAG exon sequencing and mutation location of the sister of the proband were the same as those of the proband; The serological result of her son was CCDee, the genotyping result was CCDee, and the RhAG exon sequencing result was heterozygous frameshift mutation, consistent with the proband.Conclusion:A novel mutation site in the RhAG gene was discovered after analysis. This site causes incomplete expression of RhAG protein in patients, which in turn affects the expression of other Rh antigens on the cell membrane, resulting in a serological result of Rh null.
