Search

Background: Type 2 diabetes mellitus (T2DM) is a crucial risk factor for cognitive impairment. Accurate assessment of patients’ cognitive function and early intervention is helpful to improve patient’s quality of life. At present, neuropsychiatric screening tests is often used to perform this task in clinical practice. However, it may have poor repeatability. Moreover, several studies revealed that machine learning (ML) models can effectively assess cognitive impairment in Alzheimer’s disease (AD) patients. We investigated whether we could develop an MRI-based ML model to evaluate the cognitive state of patients with T2DM.Objective: To propose MRI-based ML models and assess their performance to predict cognitive dysfunction in patients with type 2 diabetes mellitus (T2DM).Methods: Fluid Attenuated Inversion Recovery (FLAIR) of magnetic resonance images (MRI) were derived from 122 patients with T2DM. Cognitive function was assessed using the Chinese version of the Montréal Cognitive Assessment Scale-B (MoCA-B). Patients with T2DM were separated into the Dementia (DM) group (n = 40), MCI group (n = 52), and normal cognitive state (N) group (n = 30), according to the MoCA scores. Radiomics features were extracted from MR images with the Radcloud platform. The variance threshold, SelectKBest, and least absolute shrinkage and selection operator (LASSO) were used for the feature selection. Based on the selected features, the ML models were constructed with three classifiers, k-NearestNeighbor (KNN), Support Vector Machine (SVM), and Logistic Regression (LR), and the validation method was used to improve the effectiveness of the model. The area under the receiver operating characteristic curve (ROC) determined the appearance of the classification. The optimal classifier was determined by the principle of maximizing the Youden index.Results: 1,409 features were extracted and reduced to 13 features as the optimal discriminators to build the radiomics model. In the validation set, ROC curves revealed that the LR classifier had the best predictive performance, with an area under the curve (AUC) of 0.831 in DM, 0.883 in MIC, and 0.904 in the N group, compared with the SVM and KNN classifiers.Conclusion: MRI-based ML models have the potential to predict cognitive dysfunction in patients with T2DM. Compared with the SVM and KNN, the LR algorithm showed the best performance.

ObjectiveTo investigate the correlation between transcription factor 7-like 2 (TCF7L2) gene polymorphisms and gestational diabetes mellitus (GDM) risk in the central Chinese population.MethodsThis case–control study examined the association of seven TCF7L2 gene single-nucleotide polymorphisms (SNPs) (rs11196218, rs4506565, rs7895340, rs7901695, rs11196205, rs12243326, and rs290487) with GDM risk in the central Chinese population (843 GDM and 877 controls). The clinical information and blood samples were collected by trained interviewers and nurses. Genotyping of SNPs was conducted on the Sequenom MassARRAY platform. Statistical analyses including t-test, ANOVA, chi-square test, Fisher’s exact test, and logistic regression were performed.ResultsDifferences in age, pre-pregnant body mass index (BMI), and family history of type 2 diabetes mellitus (T2DM) between the case and control groups were significant (p < 0.05). Compared with the wild-type genotype, pregnant women with genotypes of rs4506565-AT (OR = 1.89, 95%CI: 1.18–3.02), rs7895340 GA (OR = 1.93, 95%CI: 1.06–3.54), rs7901695-TC (OR = 1.79, 95%CI: 1.11–2.88), and rs11196205-GC (OR = 2.15, 95%CI: 1.16–3.98) had a significantly higher risk of GDM, adjusted by age, pre-pregnant BMI, and family history of T2DM. Functional annotation showed that all these four SNPs fell in the functional elements of human pancreatic islets. Further cumulative effects analysis concluded that when participants carried all these four risk genotypes, the risk of GDM was 3.51 times (OR = 3.51, 95%CI: 1.38–8.90) than that of those without any risk genotypes.ConclusionsThe findings of this study suggested that rs4506565, rs7895340, rs7901695, and rs11196205 were the genetic susceptibility SNPs of GDM in the central Chinese population. Further studies are needed to validate our findings and clarify the underlying mechanisms.

Esophageal squamous cell carcinoma (ESCC) is a malignant tumor developing from the esophageal squamous epithelium, and is the most common histological subtype of esophageal cancer (EC). EC ranks 10th in morbidity and sixth in mortality worldwide. The morbidity and mortality rates in China are both higher than the world average. Current treatments of ESCC are surgical treatment, radiotherapy, and chemotherapy. Neoadjuvant chemoradiotherapy plus surgical resection is recommended for advanced patients. However, it does not work in the significant promotion of overall survival (OS) after such therapy. Research on targeted therapy in ESCC mainly focus on EGFR and PD-1, but neither of the targeted drugs can significantly improve the 3-year and 5-year survival rates of disease. Phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) pathway is an important survival pathway in tumor cells, associated with its aggressive growth and malignant progression. Specifically, proliferation, apoptosis, autophagy, and so on. Related genetic alterations of this pathway have been investigated in ESCC, such as PI3K, AKT and mTOR-rpS6K. Therefore, the PI3K/AKT/mTOR pathway seems to have the capability to serve as research hotspot in the future. Currently, various inhibitors are being tested in cells, animals, and clinical trials, which targeting at different parts of this pathway. In this work, we reviewed the research progress on the PI3K/AKT/mTOR pathway how to influence biological behaviors in ESCC, and discussed the interaction between signals downstream of this pathway, especially eukaryotic translation initiation factors (eIFs) and the development and progression of ESCC, to provide reference for the identification of new therapeutic targets in ESCC.