Kirsten rat sarcoma viral oncogene homolog (KRAS) protein inhibitors are a promising class of therapeutics, but research on molecules that effectively penetrate the blood-brain barrier (BBB) remains limited, which is crucial for treating central nervous system (CNS) malignancies. Although molecular generation models have recently advanced drug discovery, they often overlook the complexity of biological and chemical factors, leaving room for improvement. In this study, we present a structure-constrained molecular generation workflow designed to optimize lead compounds for both drug efficacy and drug absorption properties. Our approach utilizes a variational autoencoder (VAE) generative model integrated with reinforcement learning for multi-objective optimization. This method specifically aims to enhance BBB permeability (BBBp) while maintaining high-affinity substructures of KRAS inhibitors. To support this, we incorporate a specialized KRAS BBB predictor based on active learning and an affinity predictor employing comparative learning models. Additionally, we introduce two novel metrics, the knowledge-integrated reproduction score (KIRS) and the composite diversity score (CDS), to assess structural performance and biological relevance. Retrospective validation with KRAS inhibitors, AMG510 and MRTX849, demonstrates the framework's effectiveness in optimizing BBBp and highlights its potential for real-world drug development applications. This study provides a robust framework for accelerating the structural enhancement of lead compounds, advancing the drug development process across diverse targets.

Objective:To explore the effect of intelligent voice follow-up system in elderly patients with diabetes.Methods:From March 2021 to June 2022, 182 patients with type 2 diabetes hospitalized in the Healthcare Center of Beijing Friendship Hospital Affiliated to Capital Medical University were selected by convenience sampling. The patients were randomly divided into a control group and an observation group, with 91 cases in each group. The control group implemented routine follow-up management, while the observation group implemented follow-up management based on an intelligent voice follow-up system. The medication compliance, diet behavior compliance, fasting blood glucose and glycated hemoglobin were compared between the two groups.Results:After intervention, 49 cases (53.85%) in the observation group fully complied with medication, 32 cases (35.16%) partially complied, and 10 cases (10.99%) did not comply. In the control group, 35 cases (38.46%) fully complied with medication, 34 cases (37.36%) partially complied, and 22 cases (24.18%) did not comply. The medication compliance of the observation group was better than that of the control group, with a statistically significant difference ( Z=2.502, P<0.05). The medication compliance of the observation group after intervention was better than before intervention, with a statistically significant difference ( Z=3.293, P<0.05). After intervention, the scores of all dimensions and total scores of the Dietary Behavior Compliance Scale in the observation group were higher than those in the control group and before intervention, with statistically significant differences ( P<0.05). After the intervention, the fasting blood glucose and glycated hemoglobin in the observation group were lower than those in the control group and before the intervention, and the difference was statistically significant ( P<0.05) . Conclusions:The intelligent voice follow-up system can effectively improve the medication compliance and diet compliance of elderly diabetes patients, and improve the blood glucose control, which is worthy of clinical practice.

Objective:To evaluate the correlation between the timed up and go(TUG)test and fall risks in elderly frail patients.Methods:From July to September 2019, elderly frail patients who were treated at the cardiovascular department of our hospital were enrolled.Basic clinical data and fall-related information of patients were collected.Patients were divided into the fall group and the non-fall group.Results on the body mass index(BMI), TUG, 4-meter maximum walking speed(4 m MWS)and Barthel index were compared between the two groups.The correlation between TUG and each indicator was examined.Multivariate Logistic regression analysis was used to analyze the correlation between the TUG and falls in elderly patients.Results:A total of 96 eligible patients were enrolled, including 35 in the fall group and 61 in the non-fall group.The average TUG time was longer in the fall group than in the non-fall group(16.45±6.44 s vs.10.17±2.91 s, t=-6.556, P<0.001). The correlation analysis results showed that the TUG was correlated with falls and 4 m MWS( r=0.582 and 0.875, both P<0.001). Multivariate Logistic regression analysis showed that the TUG( OR=1.201, 95% CI: 1.111-1.470, P=0.004)and 4 m MWS( OR=1.146, 95% CI: 1.063-1.244, P=0.015)were risk factors for falls. Conclusions:The TUG is correlated with fall risks in elderly frail patients and should be recommended as a routine test in clinical practice.