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

Objective:To evaluate the safety, feasibility and operational performance of self-developed medical disposable portable endoscopy (YunSendo) for upper gastrointestinal endoscopy examination in Ba-Ma mini-pigs.Methods:A total of 10 Guangxi Ba-Ma mini-pigs were used in the experiment, and mucosal injury models were established in advance by biopsy forceps in esophagus, stomach, and duodenum. Each experimental animal underwent medical disposable portable endoscopy and Olympus endoscopy (GIF-Q260J) performed by two endoscopists separately. The time when the endoscope reached the duodenum, the number of detected mucosal injuries and endoscopic pictures of different parts with standard image acquisition were recorded. Endoscopic operational performance and endoscopic image quality were evaluated. Different endoscopists recorded experimental results with blind method. The procedures of the two endoscopic examinations were performed by coin-tossing method. The paired t test was used for statistical analysis. Results:There were no statistically significant differences in the insertion time and total operation time between medical disposable portable endoscopy and Olympus endoscopy ( (171.00±9.96) s vs. (164.00±17.84) s, (285.00±33.94) s vs. (273.40±23.46) s; t=1.289 and 1.281, P=0.230 and 0.232). There were no statistically significant differences in the percentage of time of clear visual field during endoscopy insertion and total operation between medical disposable portable endoscopy and Olympus endoscopy ((91.83±1.85)% vs. (91.52±1.51)%, (93.07±3.10)% vs. (92.06±2.57)%; t=0.401 and 0.689, P=0.698 and 0.508). Moreover, there were no statistically significant differences in the score of comprehensive operation performance, score of clear image number, score of image color recognition, score of image illumination, comprehensive score of image quality and number of detected mucosal injuries ((9.66±0.30) points vs. (9.86±0.15) points, (39.50±0.71) points vs. (39.30±1.06) points, (39.70±0.48) points vs. (39.40±0.70) points, (39.40±0.70) points vs. (39.50±0.71) points, (9.88±0.09) points vs. (9.85±0.20) points, 9.80±0.42 vs. 9.90±0.32; t=2.176, 1.000, 1.152, 0.317, 0.629 and 0.557, all P>0.05). There were no adverse events after operation in medical disposable portable endoscopy group and Olympus endoscopy group. Conclusions:The medical disposable portable endoscopy is safe and feasible for endoscopy examination in live animal models. Different parts of upper gastrointestinal tract and mucosal lesions can be clearly detected. The operational performance and the image quality are excellent, which is similar to Olympus endoscopy (GIF-Q260J).

Semen Ziziphi spinosae ( SZS) , as a commonly used traditional Chinese medicine in clinic for insomnia therapy, is rich in pharmacological active ingredients such as saponins, flavonoids, alkaloids, oil and the other chemical compounds. The recent stud-ies indicated that some active ingredients in SZS exhibited a variety of activities including sedative hypnotics, antianxiety and anti-de-pression by regulating particular neurotransmitters such as 5-hydroxytryptamine (5-HT), gamma-amino butyric acid (GABA), norepi-nephrine, dopamine and glutamate. In accordance with the previous studies on pharmacological activities of SZS, this paper summa-rized and reviewed the applications of SZS in the treatment of central nervous system diseases, myocardial diseases and hepatic disea-ses, which might provide solid evidence for the application development of SZS.