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 summarize the clinical manifestations, pathological characteristics, treatment options and prognosis of the world's first case of prostate ductal adenocarcinoma (PDA) complicated with prostate mucinous adenocarcinoma (PMA). [Methods] The clinical and follow-up data of a patient with PDA and PMA treated in Zhongnan Hospital of Wuhan University were retrospectively analyzed, and relevant literature in PubMed and CNKI databases was retrieved. [Results] The patient sought medical attention due to dysuria, frequent urination, urinary urgency and urinary pain for more than half a year, and was admitted to hospital 3 times in total.The initial diagnosis upon the first admission was benign prostatic hyperplasia complicated with prostatic abscess.After 2 months, the patient was readmitted due to worsening symptoms, received transurethral bladder neck incision+ cystoscopy+ transurethral plasma resection of the prostate, and postoperative diagnosis confirmed PDA with local PMA.Three months after surgery, the patient had bleeding.After auxiliary examinations revealed extensive metastasis, he received hormonal therapy.After 9 months, the patient died due to multiple lung metastases. [Conclusion] Early diagnosis has a significant impact on the treatment and prognosis, but there have been no previous reports of PDA combined with PMA, so the lack of specific biomarkers in the early stage has led to missed diagnosis or misdiagnoses.There is no specific treatment for PDA with PMA. Radical prostatectomy was not satisfactory in the treatment of this case.