OBJECTIVES: We propose a multi-feature fusion model based on manually extracted features and deep learning features from endoscopic images for grading rebleeding risk of peptic ulcers. METHODS: Based on the endoscopic appearance of peptic ulcers, color features were extracted to distinguish active bleeding (Forrest I) from non-bleeding ulcers (Forrest II and III). The edge and texture features were used to describe the morphology and appearance of the ulcers in different grades. By integrating deep features extracted from a deep learning network with manually extracted visual features, a multi-feature representation of endoscopic images was created to predict the risk of rebleeding of peptic ulcers. RESULTS: In a dataset consisting of 3573 images from 708 patients with Forrest classification, the proposed multi-feature fusion model achieved an accuracy of 74.94% in the 6-level rebleeding risk classification task, outperforming the experienced physicians who had a classification accuracy of 59.9% (P<0.05). The F1 scores of the model for identifying Forrest Ib, IIa, and III ulcers were 90.16%, 75.44%, and 77.13%, respectively, demonstrating particularly good performance of the model for Forrest Ib ulcers. Compared with the first model for peptic ulcer rebleeding classification, the proposed model had improved F1 scores by 5.8%. In the simplified 3-level risk (high-risk, low-risk, and non-endoscopic treatment) classification task, the model achieved F1 scores of 93.74%, 81.30%, and 73.59%, respectively. CONCLUSIONS The proposed multi-feature fusion model integrating deep features from CNNs with manually extracted visual features effectively improves the accuracy of rebleeding risk classification for peptic ulcers, thus providing an efficient diagnostic tool for clinical assessment of rebleeding risks of peptic ulcers.
Humans ; Deep Learning ; Peptic Ulcer ; Risk Assessment ; Peptic Ulcer Hemorrhage ; Recurrence

Objective To explore the predicting value of quantitative CT features of peritumoral nodules for the risk of thymic epi-thelial tumors(TETs).Methods A total of 108 patients with TETs confirmed by pathology were analyzed retrospectively.Accord-ing to the WHO classification of thymic tumors,the patients were divided into high-risk group and low-risk group.The quantitative CT features included the long diameter,volume,mass,and distance between peritumoral nodules and main tumor.The values of these features were assessed using the Mann-Whitney U test,and the thresholds were determined using the receiver operating characteristic(ROC)curve.Results Among 108 TETs patients,there were 45 patients with peritumoral nodules on chest CT,including 30 high-risk group and 15 low-risk group,respectively.Interobserver consistencies were well with intraclass correlation coefficient(ICC)between 0.886 and 0.907.The long diameter,volume,and mass of peritumoral nodules were greater in the high-risk group than in the low-risk group(P＜0.05),and there was no significant difference in the distance to the main tumor(P＞0.05).The mass of peritumoral nodules had the best diagnostic performance in predicting TETs risk,and the thresholds was 27.84 mg with the area under the curve(AUC)of 0.889,sensitivity of 83.3％,specificity of 87.5％,and accuracy of 85.3％,respectively.Conclusion The quantitative CT features of peritumoral nodules especially mass can be useful in predicting the risk of TETs.

BACKGROUND:Most of the studies on combined orthodontic-orthognathic treatment of skeletal class Ⅲ malocclusions have focused on the improvement of the patient's lateral appearance and recovery in the later stages of the treatment,while there are fewer studies observing the microcosmic nature of the alveolar bone remodeling of the lower anterior teeth. OBJECTIVE:To evaluate the therapeutic effect of lower anterior tooth decompensation and alveolar bone remodeling in patients with skeletal class Ⅲ malocclusion before and after orthodontic-orthognathic treatment based on oral X-ray lateral films and oral cone-beam CT. METHODS:From January 2015 to May 2023,15 patients with skeletal class Ⅲ malocclusion who underwent orthodontic-orthognathic surgery at Qingdao Hospital of Rehabilitation University were enrolled.All patients underwent lateral cephalography and cone beam computed tomography before and after treatment.Cephalometric measurement items related to the angle and line distance,lip/lingual bone cracking length(d-La/d-Li)and bone cracking/bone fenestration of the lower anterior teeth before and after treatment were measured. RESULTS AND CONCLUSION:Lateral X-ray films showed that the amount of alveolar bone remodeling after decompensation of the lower anterior teeth showed significant changes compared to before treatment.The root of the tooth moved significantly towards the center of the alveolar bone,and the specific data was closer to normal data,but there were still some differences compared with normal individuals.Based on the cone-beam CT measurement,the bone cracking/bone fenestration length and width of the alveolar bone were improved in almost all the teeth after orthodontic-orthognathic combined treatment,alveolar bone remodeling in some teeth even reached the level of healthy individuals.Before treatment,most patients often experienced bone fenestration/cracking on the lip/lingual side of the lower incisor due to compensatory tooth growth.However,during the preoperative orthodontic stage,decompensation triggered alveolar bone remodeling and significant changes in tooth angle.Preoperative orthodontic treatment caused the upper anterior teeth to retract and the lower anterior teeth to tilt and control the root,but the amount of decompensation before surgery was often insufficient.In the orthognathic surgery stage,the jaw was removed through the positioning guide plate,the maxilla moved forward,and the mandible retreated.During the postoperative orthodontic process,the effect of fine adjustment was better.Although there is a certain degree of recurrence trend in the position of teeth and jawbones,the postoperative orthodontic treatment is closer to the normal value.

Tumors represent one of the primary threats to human life, with the dissemination of malignant tumors being a leading cause of mortality among cancer patients. Early diagnosis of tumors can reliably predict their progression, significantly reducing mortality rates. Tumor markers, including circulating tumor cells, exosomes, proteins, circulating tumor DNA, miRNAs and so on, generated during the tumor development process, have emerged as effective approach for early tumor diagnosis. Several methods are currently employed to detect tumor markers, such as polymerase chain reaction, Northern blotting, next-generation sequencing, flow cytometry, and enzyme-linked immunosorbent assay. However, these methods often suffer from time-consuming process, high costs, low sensitivity, and the requirement for specialized personnel. Therefore, a new rapid, sensitive, and specific tumor detection method is urgently needed.The clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein (Cas) system, originating from the adaptive immune system of bacteria, has found extensive applications in gene editing and nucleic acid detection. Based on the structure and function of Cas proteins, the CRISPR/Cas system can be classified into two classes and six types. Class I systems consist of multiple Cas protein complexes, including types I, III, and IV, while Class II systems comprise single, multi-domain Cas proteins mediated by RNA, including types II (Cas9), V (Cas12), and VI (Cas13). Class II systems have been widely employed in the fields of biotechnology and nucleic acid diagnostics due to their efficient target binding and programmable RNA specificity. Currently, fluorescence method is the most common signal output technique in CRISPR/Cas-based biosensors. However, this method often requires the integration of signal amplification technologies to enhance sensitivity and involves expensive and complex fluorescence detectors. To enhance the detection performance of CRISPR/Cas-based biosensors, the integration of CRISPR/Cas with some alternative techniques can be considered. The CRISPR/Cas integrated electrochemical sensor (E-CRISPR) possesses advantages such as miniaturization, high sensitivity, high specificity, and fast response speed.E-CRISPR can convert the reactions between biomolecules and detecting components into electrical signals, rendering the detection signals more easily readable and reducing the impact of background values. Therefore,E-CRISPR enhances the accuracy of detection results. E-CRISPR has been applied in various fields, including medical and health, environmental monitoring, and food safety. Furthermore, E-CRISPR holds tremendous potential for advancing the detection levels of tumor markers.Among all types of Cas enzymes, the three most widely applied are Cas9, Cas12, and Cas13, along with their respective subtypes. In this work, we provided a brief overview of the principles and characteristics of Class II CRISPR/Cas single-effector proteins. This paper focused on the various detection technologies based on E-CRISPR technique, including electrochemical impedance spectroscopy, voltammetry, photoelectrochemistry, and electrochemiluminescence. We also emphasized the applications of E-CRISPR in the field of tumor diagnosis, which mainly encompasses the detection of three typical tumor markers (ctDNA, miRNA, and proteins). Finally, we discussed the advantages and limitations of E-CRISPR, current challenges, and future development prospects. In summary, althoughE-CRISPR platform has made significant strides in tumor detection, certain challenges still need to be overcome for their widespread clinical application. Continuous optimization of the E-CRISPR platform holds the promise of achieving more accurate tumor subtyping diagnoses in clinical settings, which would be of significant importance for early patient diagnosis and prognosis assessment.

Objective:To investigate the clinical characteristics, treatment and prognosis of newly-treated patients with primary central nervous system lymphoma (PCNSL).Methods:Clinical data of 117 newly-treated PCNSL patients who were admitted to the First Hospital of Jilin University, the Fifth Medical Center of Chinese PLA General Hospital, Harbin Medical University Cancer Hospital, and Cancer Hospital of Chinese Academy of Medical Sciences & Peking Union Medical College from August 2009 to February 2018 were retrospectively analyzed. The patients' age, sex, Eastern Cooperative Oncology Group (ECOG) physical status (PS) score, pathological type, involvement of deep brain tissue, number of lesions, cerebrospinal fluid protein concentration, International Extranodal Lymphoma Study Group (IELSG) score, Memorial Sloan Kettering Cancer Center (MSKCC) score, treatment strategy, and response after the first-line therapy were analyzed using univariate and multivariate Cox proportional hazards models to identify the independent influencing factors for progression-free survival (PFS) and overall survival (OS) of PCNSL patients. Kaplan-Meier method was used for survival analysis.Results:In 117 newly-treated PCNSL patients, 59 cases (50.4%) presented with increased intracranial pressure or focal neurological symptoms at diagnosis; there were 65 cases (55.6%) with single lesions and 52 cases (44.4%) with multiple lesions; 1 patient (0.9%) had lymphoma of T-cell origin, and 116 cases (99.1%) had diffuse large B-cell lymphoma (DLBCL). Among 95 evaluable patients, 41 patients (43.2%) achieved complete remission (CR), 20 patients (21.1%) achieved partial remission (PR), 16 patients (16.8%) achieved stable disease (SD), and 18 patients (18.9%) had progressive disease (PD). In 117 patients with median follow-up of 66.0 months (95% CI 57.9-74.1 months), the median PFS and OS were 17.4 months (95% CI 11.5-23.3 months) and 45.6 months (95% CI 20.1-71.1 months), respectively. The 2-, 3- and 5-year PFS rates were 41.2%, 28.6% and 19.3%, and OS rates were 63.7%, 52.4% and 46.3%, respectively. Univariate Cox regression analysis showed that baseline high-risk MSKCC score group was an adverse prognostic factor for PFS ( P = 0.037), and the first-line chemotherapy with ≥4 cycles of high-dose methotrexate (HDMTX), HDMTX in combination with rituximab, ≥4 cycles of rituximab in combination with HDMTX, and achieving CR or ≥PR after the first-line treatment reduced the risk of disease progression and prolonged the PFS time (all P <0.01); age >60 years old, ECOG-PS score of 2-4 points, elevated cerebrospinal fluid protein concentration, high-risk IELSG score, and high-risk MSKCC score were adverse prognostic factors for OS, and ≥4 cycles of HDMTX and achieving CR or ≥PR after the first-line treatment were favorable factors for OS. Multivariate Cox regression analysis verified that rituximab in combination with HDMTX (yes vs. no: HR = 0.349, 95% CI 0.133-0.912, P = 0.032) and achieving ≥PR after the first-line chemotherapy (yes vs. no: HR = 0.028, 95% CI 0.004-0.195, P < 0.001) were independent favorable factors for PFS; age >60 years old (>60 years old vs. ≤60 years old: HR = 10.878, 95% CI 1.807-65.488, P = 0.009) was independent unfavorable factor for OS, while ≥4 cycles of HDMTX treatment (≥4 cycles vs. <4 cycles: HR = 0.225, 95% CI 0.053-0.947, P = 0.042) was independent favorable factor for OS. Conclusions:The older the PCNSL patients at initial treatment, the worse the prognosis. Intensive and continuous treatment for achieving deeper remission may be the key for improving the outcome of PCNSL patients.

The rise of internet thinking, along with the growing acceptance of the " patient-centered" concept, has played a significant role in driving the development of electronic patient-reported outcomes (ePROs) in the field of clinical evaluation. While several large-scale and well-established ePROs evaluation systems have been widely adopted and implemented worldwide, China has lagged behind in ePROs research and electronic application. Therefore, the purpose of this paper is to comprehensively analyze and summarize the concepts, types, characteristics, state of development, and application of ePROs. Specifically, it focuses on analyzing the structure and content of ePROs evaluation network systems in the United States, France, and the United Kingdom. The ultimate objective is to provide insightful analysis and useful suggestions to aid in the development and evolution of ePROs in China.

Objective:To investigate the application effect of the combined teaching model of digital 3D printed model and Tencent conference in case-based learning (CBL) teaching of oral and maxillofacial surgery.Methods:A total of 80 undergraduates in the classes of 2015 and 2016 were selected from School of Stomatology, Qingdao University. The students in the class of 2015 received traditional teaching, and those in the class of 2016 received the combined CBL teaching model of 3D printed model and Tencent conference. A questionnaire survey was used to evaluate the teaching effect, and theoretical examination was used to assess comprehensive abilities of the two groups. SPSS 24.0 was used to perform the chi-square test and the t-test. Results:There was no significant difference in the degree of satisfaction with teaching between the combined CBL teaching model of 3D printed model and Tencent conference and the traditional teaching model ( P>0.05), and both models were generally recognized and accepted by students. The experimental group had a significantly higher score than the control group (94.05±4.16 vs. 86.10±3.37, P<0.05). Conclusion:The combined teaching model of digital 3D printed model and Tencent conference integrates the advantages of the Internet and digital information and thus provides a certain reference for the teaching methods for other majors in stomatology.

Based on the research results of aspiration in patients with dysphagia after ischemic stroke at home and abroad, this paper reviews the definition, detection methods, and risk factors of aspiration and emphasizes the incidence rate and severity of this disease. The authors conclude that preventing aspiration can decrease the incidence rate of aspiration pneumonia, change the clinical outcome of patients, and thereby save medical resources.