Development of a Malignancy Potential Binary Prediction Model Based on Deep Learning for the Mitotic Count of Local Primary Gastrointestinal Stromal Tumors

Objective@#The mitotic count of gastrointestinal stromal tumors (GIST) is closely associated with the risk of planting and metastasis. The purpose of this study was to develop a predictive model for the mitotic index of local primary GIST, based on deep learning algorithm. @*Materials and Methods@#Abdominal contrast-enhanced CT images of 148 pathologically confirmed GIST cases were retrospectively collected for the development of a deep learning classification algorithm. The areas of GIST masses on the CT images were retrospectively labelled by an experienced radiologist. The postoperative pathological mitotic count was considered as the gold standard (high mitotic count, > 5/50 high-power fields [HPFs]; low mitotic count, ≤ 5/50 HPFs). A binary classification model was trained on the basis of the VGG16 convolutional neural network, using the CT images with the training set (n = 108), validation set (n = 20), and the test set (n = 20). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated at both, the image level and the patient level. The receiver operating characteristic curves were generated on the basis of the model prediction results and the area under curves (AUCs) were calculated. The risk categories of the tumors were predicted according to the Armed Forces Institute of Pathology criteria. @*Results@#At the image level, the classification prediction results of the mitotic counts in the test cohort were as follows:sensitivity 85.7% (95% confidence interval [CI]: 0.834–0.877), specificity 67.5% (95% CI: 0.636–0.712), PPV 82.1% (95% CI: 0.797–0.843), NPV 73.0% (95% CI: 0.691–0.766), and AUC 0.771 (95% CI: 0.750–0.791). At the patient level, the classification prediction results in the test cohort were as follows: sensitivity 90.0% (95% CI: 0.541–0.995), specificity 70.0% (95% CI: 0.354–0.919), PPV 75.0% (95% CI: 0.428–0.933), NPV 87.5% (95% CI: 0.467–0.993), and AUC 0.800 (95% CI: 0.563–0.943). @*Conclusion@#We developed and preliminarily verified the GIST mitotic count binary prediction model, based on the VGG convolutional neural network. The model displayed a good predictive performance.

Influences of Oral Undenatured Collagen Type Ⅱ on Knee Joint Range of Motion in Patients with Knee Osteoarthritis / 医用生物力学

Objective To evaluate the effects of oral undenatured collagen type Ⅱ on rehabilitation of knee osteoarthritis. Methods Sixteen patients with levelⅠand Ⅱ knee osteoarthritis (KOA) were recruited in this study. The WOMAC questionnaire was used to measure the symptoms of knee joint in patients. The KneeKG 3D evaluation system was used to measure the kinematics characteristics of the knee joints. The pair t-test was applied to analyze the differences in knee kinematics after taking undenatured collagen type Ⅱ for 3 months. Results After taking undenatured collagen type Ⅱ, the patient’s feeling of knee joint pain, stiffness, and the difficulty of daily life was reduced. In addition, the knee flexion angle increased from 51.8°±15.2° to 58.4°±10.7° (P<0.05), and the knee varus angle decreased to 2.1°±4.8°. The improvement rate of knee varus was 82%. Conclusions Taking undenatured collagen type Ⅱ for 3 months could play a role in rehabilitation for patients with early KOA. The knee range of motion was close to the level of healthy elderly people, thus improving the symptoms of knee varus.

Research progress on relationship between bile acids and intestinal mucosal mechanical barrier function / 中华消化外科杂志

Intestinal mucosal mechanical barrier is one of the most important structure to maintain the body homeostasis.The occurrences of inflammatory bowel disease,necrotizing enterocolitis and poor prognosis of patients with obstructive jaundice are closely related to the damage of the mucosal barrier function.Long-term high fat diet and obstructive jaundice can cause the abnormality of bile acids metabolism.These pathological conditions are often associated with the destruction of intestinal mucosal barrier function.So the correlations between abnormal bile metabolism and intestinal mucosal mechanical barrier function have aroused interests of many researchers.They found that bile acids the important component of bile are closely related to the intestinal barrier function.The paper reviewed the recent articles and summarized the mechanisms of the deficiency of bile acids,excessive bile acids and abnormal bile acids composition damaging the intestinal mucosal barrier function.It will provide reference for the new fields of study,prevention of the toxic effects of bile acids and the improvement of the prognosis of patients.

Progress of molecular subtypes of colorectal cancer / 中华肿瘤杂志

Colorectal cancer is one of the leading causes of morbidity and mortality around the world. Although the staging and classification systems, such as tumor/node/metastasis (TNM) staging, are widely used in clinic, there are some limitations. For example, the patients with the same pathological type, TNM stage and treatment regimen show a completely different prognosis and outcome. In the present, molecular subtyping was concerned for a more precise and accurate staging of colorectal cancer. Herein, we reviewed the literature of the molecular subtypes of colorectal cancer in the past decades. The clinical significance of various molecular subtyping systems was evaluated and compared. It will provide reference for the precision medicine of colorectal cancer in the future.