Objective:To evaluate the efficacy of 3D printing technology combined with computer navigation-assisted screw implantation in the treatment of atlantoaxial instability (AAI) complicated by vertebral artery anomalies.Methods:A retrospective case series study was conducted to analyze the clinical data of 23 patients with AAI complicated by vertebral artery anomalies who were admitted to Honghui Hospital of Xi′an Jiaotong University between January 2019 and January 2023, including 10 males and 13 females, aged 19-70 years [(51.0±13.3)years]. Vertebral artery anomalies were categorized into unilateral high-riding vertebral artery with unilateral dominance ( n=14), bilateral high-riding vertebral arteries with unilateral dominance ( n=6), and ponticulus posticus ( n=3). All the patients underwent preoperative planning using a 3D-printed model of the atlantoaxial complex with the vertebral artery, followed by posterior atlantoaxial pedicle screw fixation with computer-assisted navigation. Operative duration and intraoperative blood loss were recorded. The accuracy of pedicle screw placement was assessed at 3 days postoperatively using the Gertzbein-Robbins classification. Visual analogue scale (VAS) scores and Japanese Orthopedic Association (JOA) scores were evaluated preoperatively, at 3 days, 3 months postoperatively, and at the last follow-up. Bony fusion was assessed using cervical CT with 3D reconstruction at the last follow-up. Complications were also observed. Results:All the patients were followed up for 12-19 months [(15.1±1.9)months]. The operative duration was 125-167 minutes [(140.6±10.9)minutes] and intraoperative blood loss was 200-600 ml [(295.7±77.8)ml]. At 3 days postoperatively, all the 66 pedicle screws were safely placed, with 60 screws (91%) rated as Gertzbein-Robbins Grade 0 and 6 screws (9%) as Grade 1. At 3 days and 3 months postoperatively, and at the last follow-up, the VAS scores were (4.0±1.0)points, (2.0±0.6)points, and (1.3±0.5)points, and the JOA scores were (14.2±1.2)points, (16.0±0.8)points, and (16.6±0.5)points, both of which were not only significantly improved compared with preoperative (5.6±1.3)points and (12.8±1.5)points, but also further improved over time ( P<0.05). At the last follow-up, 22 patients (96%) achieved satisfactory atlantoaxial bony fusion. No vertebral artery injury, spinal cord or nerve injury, cerebrospinal fluid leakage, or screw loosening were observed in any patients. Conclusion:For patients with AAI complicated by vertebral artery anomalies, 3D printing combined with computer navigation-assisted navigation for atlantoaxial pedicle screw implantation offers multiple advantages, including minimal surgical trauma, high screw placement accuracy, pain relief, neurological function improvement, high fusion rate, and lowered incidence of complications.

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 efficacy of 3D printing technology combined with computer navigation-assisted screw implantation in the treatment of atlantoaxial instability (AAI) complicated by vertebral artery anomalies.Methods:A retrospective case series study was conducted to analyze the clinical data of 23 patients with AAI complicated by vertebral artery anomalies who were admitted to Honghui Hospital of Xi′an Jiaotong University between January 2019 and January 2023, including 10 males and 13 females, aged 19-70 years [(51.0±13.3)years]. Vertebral artery anomalies were categorized into unilateral high-riding vertebral artery with unilateral dominance ( n=14), bilateral high-riding vertebral arteries with unilateral dominance ( n=6), and ponticulus posticus ( n=3). All the patients underwent preoperative planning using a 3D-printed model of the atlantoaxial complex with the vertebral artery, followed by posterior atlantoaxial pedicle screw fixation with computer-assisted navigation. Operative duration and intraoperative blood loss were recorded. The accuracy of pedicle screw placement was assessed at 3 days postoperatively using the Gertzbein-Robbins classification. Visual analogue scale (VAS) scores and Japanese Orthopedic Association (JOA) scores were evaluated preoperatively, at 3 days, 3 months postoperatively, and at the last follow-up. Bony fusion was assessed using cervical CT with 3D reconstruction at the last follow-up. Complications were also observed. Results:All the patients were followed up for 12-19 months [(15.1±1.9)months]. The operative duration was 125-167 minutes [(140.6±10.9)minutes] and intraoperative blood loss was 200-600 ml [(295.7±77.8)ml]. At 3 days postoperatively, all the 66 pedicle screws were safely placed, with 60 screws (91%) rated as Gertzbein-Robbins Grade 0 and 6 screws (9%) as Grade 1. At 3 days and 3 months postoperatively, and at the last follow-up, the VAS scores were (4.0±1.0)points, (2.0±0.6)points, and (1.3±0.5)points, and the JOA scores were (14.2±1.2)points, (16.0±0.8)points, and (16.6±0.5)points, both of which were not only significantly improved compared with preoperative (5.6±1.3)points and (12.8±1.5)points, but also further improved over time ( P<0.05). At the last follow-up, 22 patients (96%) achieved satisfactory atlantoaxial bony fusion. No vertebral artery injury, spinal cord or nerve injury, cerebrospinal fluid leakage, or screw loosening were observed in any patients. Conclusion:For patients with AAI complicated by vertebral artery anomalies, 3D printing combined with computer navigation-assisted navigation for atlantoaxial pedicle screw implantation offers multiple advantages, including minimal surgical trauma, high screw placement accuracy, pain relief, neurological function improvement, high fusion rate, and lowered incidence of complications.

Objective To train an automatic recognition and classification model of ulcerative colitis colonoscopy image based on ResNet,and to test its accuracy,in order to help doctors improve the clinical detection rate and classification accuracy of ulcerative colitis.Methods A total of 4000 colonoscopy images were retrospectively collected from the Colonoscopy Center of Hebei Hospital of Traditional Chinese Medicine from January 2018 to October 2023,and were divided into normal group,mild group,moderate group and severe group according to Mayo endoscopic scoring criteria,with 1000 images for each group.After pre-processing such as brightness adjustment and Angle rotation,the number of images was expanded to 20,000,and the data set was randomly divided into training set,verification set and test set according to the ratio of 7∶2∶1.The training set and verification set are input into the ResNet model to learn and test its stability.After all training is completed,the accuracy of the model is recorded through the test set,and the accurate regression curve is made to evaluate the classification effect of the model.Results In the test set,the accuracy of classification of ulcerative colitis was 99.8％in normal group,98.8％in mild group,95.6％in moderate group and 97.8％in severe group.Conclusion ResNet has good performance in image recognition and classification of ulcerative colitis,can improve the clinical accuracy of ulcerative colitis,and can assist doctors to identify and classify the disease,which has a more reliable clinical application value.

Objective:To explore the clinical and genetic characteristics of four children with Kabuki syndrome (KS) due to variants of KMT2D gene. Methods:Four children with KS diagnosed at the Children′s Hospital of Shanxi Province between January 2020 and December 2022 were selected as the study subjects. Whole exome sequencing was carried out for the children and their family members. Candidate variants were verified by Sanger sequencing and pathogenicity analysis.Results:The KS phenotype scores for the four children were 7, 8, 6, and 6, respectively. Child 2 also presented with a rare solitary kidney malformation. Genetic testing revealed that all children had harbored novel de novo variants of the KMT2D gene, including c. 16472_16473del, c. 858dup, c. 11899C>T, and c. 12844C>T, respectively. Based on the guidelines from the American College of Medical Genetics and Genomics (ACMG), all of the variants were classified as pathogenic. Conclusion:For children showing phenotypes such as distinctive facial features, intellectual disability, developmental delay, cardiac abnormalities, and urinary system anomalies, KS should be considered. Early diagnosis and intervention can be achieved through genetic testing, especially in the presence of KMT2D gene mutations.

Objective:To investigate the learning curve for a five-step procedure, namely, a transthoracic single-port assisted laparoscopic transabdominal diaphragmatic approach, for Siewert type II adenocarcinoma of the esophagogastric junction.Methods:In this retrospective cohort study, we analyzed relevant clinical data of 66 patients with Siewert type II adenocarcinoma of the esophagogastric junction who had undergone the five-step procedure performed by the same surgeon in the Gastrointestinal Surgery Department of Guangdong Provincial Hospital of Chinese Medicine from May 2017 to April 2023. The learning curve were plotted using cumulative summation analysis and selected indicators, including intraoperative blood loss, duration of surgery, time to first flatus, time to first tolerance of liquid food, length of hospital stay, and incidence of perioperative complications at different stages were compared. The data were analyzed using SPSS 24.0 statistical software. Numerical data are presented as cases (%) and data were analyzed using the χ 2 test or Fisher's exact test. Normally distributed measurement data are presented as x±s, and independent sample t-testing was performed for inter group comparison. Non-normally distributed measurement data are presented as M( Q1, Q3) and the Mann–Whitney U test was used for inter group comparison. Results:The five-step procedure had been successfully completed without switching to open surgery in all 66 study patients. There were no perioperative deaths, blood loss was 100 (50, 200) mL and duration of surgery 329.4±87.3 minutes. The equation of optimal fit for the duration of surgery was y=0.031x 3-4.4757x 2+164.97x-264.4 ( P<0.001, R2=0.9797). The cumulative summation learning curve reached a vertex when 25 surgical procedures had accumulated. Using 25 cases as the cut-off, we divided the learning curves into learning and proficiency periods and patients into learning (25) and proficiency period groups (41). There were no statistically significant differences between the two groups of patients in sex, age, body mass index, American Society of Anesthesiologists score, history of abdominal surgery, comorbidities, preoperative neoadjuvant therapy, maximum tumor diameter, surgical procedure, or T and N stage of tumor ( P>0.05). The following factors differed significantly (all P<0.05) between the learning and proficiency stages: in the latter there was less intraoperative blood loss (100 [50, 100] ml vs. 200 [100, 200] ml, U=-3.940, P<0.001), shorter duration of surgery ([289.8±50.7] minutes vs. [394.4±96.0] minutes, t=5.034, P<0.001), more mediastinal lymph nodes removed (5 [2, 8] vs. 2 [1, 5], U=-2.518, P=0.012), earlier time to first flatus (2 [2, 3] days vs. 4 [3, 6] days, U=-4.016, P<0.001), earlier time to first tolerance of liquid food (5 [4, 6] days vs. 7 [6, 8] days, U=-2.922, P=0.003), shorter duration of hospital stay (8 [8, 10] vs. 10 [9, 12] days, U=-2.028, P=0.043). The incidence of surgical complications did not differ significantly between the two groups ( P=0.238). Conclusion:Satisfactory results can be achieved with the five-step procedure for patients with Siewert type II adenocarcinoma of the esophagogastric junction once 25 procedures have been performed.

Objective:To explore the effectiveness and safety of programmed death 1(PD-1) inhibitory combined with chemotherapy as a neoadjuvant therapy for locally advanced resectable oral squamous cell carcinoma.Methods:This study was a randomized controlled phase Ⅱ trial. Patients recruited from Tianjin Medical University Cancer Institute and Hospital from July 2021 to February 2023 were randomly divided into two groups in a 1∶1 ratio: the experimental group (Toripalimab combined with albumin paclitaxel and cisplatin) and the control group (albumin paclitaxel and cisplatin); patients in both groups underwent three cycles of neoadjuvant therapy. After completion of neoadjuvant therapy, patients were evaluated and subsequent surgical treatment was performed. According to the completion of treatment, the analysis was conducted on both the full analysis set and the protocol set. The effectiveness and safety of treatments were evaluated. SPSS 20.0 software was used for statistical analysis.Results:A total of 41 cases with oral cancer were enrolled, including 26 males and 15 females, aged between 34 and 74 years old. There were 23 cases in the experimental group and 18 cases in the control group. A total of 23 cases completed neoadjuvant therapy and surgery according to the protocol. Experimental group and control group showed respectively the complete response rates of 1/19 and 0/17, the partial response rates of 13/19 and 8/17, the stage-down rates of 4/19 and 3/17, the pathologic complete response rate of 8/14 and 2/9, with no statistically significant differences in individual rates between two groups ( P>0.05). The major pathological response rate of 13/14 in experimental group was higher than that of 2/9 in control group ( P<0.05). The incidence of grade 3-4 adverse reactions related to treatment was low in both groups (4/23 vs. 3/18, χ 2=0.13, P=0.72), and the most common serious adverse reactions in the experimental group were granulocyte deficiency and electrolyte disorder. There were no adverse reactions that affected subsequent surgical treatment or caused death, and the safety and tolerability were good. The median follow-up time was 15 months, and the one-year disease-free survival rate of the experimental group was higher than that of control group (92.86% vs. 77.78%, χ 2=0.62, P=0.42), with a relative decrease of 87% in the risk of disease progression or death ( P=0.029). For patients with programmed death-ligand 1(PD-L1) protein expression combined positive score≥20, the experimental group showed higher major pathological response rate than control group (5/5 vs. 0/4, P=0.03). Conclusion:The neoadjuvant therapy of immunotherapy combined with chemotherapy can improve the pathological remission of oral squamous cell carcinoma and the long-term survival benefits and the prognosis of patients.

Objective:To investigate the learning curve for a five-step procedure, namely, a transthoracic single-port assisted laparoscopic transabdominal diaphragmatic approach, for Siewert type II adenocarcinoma of the esophagogastric junction.Methods:In this retrospective cohort study, we analyzed relevant clinical data of 66 patients with Siewert type II adenocarcinoma of the esophagogastric junction who had undergone the five-step procedure performed by the same surgeon in the Gastrointestinal Surgery Department of Guangdong Provincial Hospital of Chinese Medicine from May 2017 to April 2023. The learning curve were plotted using cumulative summation analysis and selected indicators, including intraoperative blood loss, duration of surgery, time to first flatus, time to first tolerance of liquid food, length of hospital stay, and incidence of perioperative complications at different stages were compared. The data were analyzed using SPSS 24.0 statistical software. Numerical data are presented as cases (%) and data were analyzed using the χ 2 test or Fisher's exact test. Normally distributed measurement data are presented as x±s, and independent sample t-testing was performed for inter group comparison. Non-normally distributed measurement data are presented as M( Q1, Q3) and the Mann–Whitney U test was used for inter group comparison. Results:The five-step procedure had been successfully completed without switching to open surgery in all 66 study patients. There were no perioperative deaths, blood loss was 100 (50, 200) mL and duration of surgery 329.4±87.3 minutes. The equation of optimal fit for the duration of surgery was y=0.031x 3-4.4757x 2+164.97x-264.4 ( P<0.001, R2=0.9797). The cumulative summation learning curve reached a vertex when 25 surgical procedures had accumulated. Using 25 cases as the cut-off, we divided the learning curves into learning and proficiency periods and patients into learning (25) and proficiency period groups (41). There were no statistically significant differences between the two groups of patients in sex, age, body mass index, American Society of Anesthesiologists score, history of abdominal surgery, comorbidities, preoperative neoadjuvant therapy, maximum tumor diameter, surgical procedure, or T and N stage of tumor ( P>0.05). The following factors differed significantly (all P<0.05) between the learning and proficiency stages: in the latter there was less intraoperative blood loss (100 [50, 100] ml vs. 200 [100, 200] ml, U=-3.940, P<0.001), shorter duration of surgery ([289.8±50.7] minutes vs. [394.4±96.0] minutes, t=5.034, P<0.001), more mediastinal lymph nodes removed (5 [2, 8] vs. 2 [1, 5], U=-2.518, P=0.012), earlier time to first flatus (2 [2, 3] days vs. 4 [3, 6] days, U=-4.016, P<0.001), earlier time to first tolerance of liquid food (5 [4, 6] days vs. 7 [6, 8] days, U=-2.922, P=0.003), shorter duration of hospital stay (8 [8, 10] vs. 10 [9, 12] days, U=-2.028, P=0.043). The incidence of surgical complications did not differ significantly between the two groups ( P=0.238). Conclusion:Satisfactory results can be achieved with the five-step procedure for patients with Siewert type II adenocarcinoma of the esophagogastric junction once 25 procedures have been performed.

Objective To train an automatic recognition and classification model of ulcerative colitis colonoscopy image based on ResNet,and to test its accuracy,in order to help doctors improve the clinical detection rate and classification accuracy of ulcerative colitis.Methods A total of 4000 colonoscopy images were retrospectively collected from the Colonoscopy Center of Hebei Hospital of Traditional Chinese Medicine from January 2018 to October 2023,and were divided into normal group,mild group,moderate group and severe group according to Mayo endoscopic scoring criteria,with 1000 images for each group.After pre-processing such as brightness adjustment and Angle rotation,the number of images was expanded to 20,000,and the data set was randomly divided into training set,verification set and test set according to the ratio of 7∶2∶1.The training set and verification set are input into the ResNet model to learn and test its stability.After all training is completed,the accuracy of the model is recorded through the test set,and the accurate regression curve is made to evaluate the classification effect of the model.Results In the test set,the accuracy of classification of ulcerative colitis was 99.8％in normal group,98.8％in mild group,95.6％in moderate group and 97.8％in severe group.Conclusion ResNet has good performance in image recognition and classification of ulcerative colitis,can improve the clinical accuracy of ulcerative colitis,and can assist doctors to identify and classify the disease,which has a more reliable clinical application value.