Approximately 30%-40% of epilepsy patients do not respond well to adequate anti-seizure medications (ASMs), a condition known as pharmacoresistant epilepsy. The management of pharmacoresistant epilepsy remains an intractable issue in the clinic. Its early prediction is important for prevention and diagnosis. However, it still lacks effective predictors and approaches. Here, a classical model of pharmacoresistant temporal lobe epilepsy (TLE) was established to screen pharmacoresistant and pharmaco-responsive individuals by applying phenytoin to amygdaloid-kindled rats. Ictal electroencephalograms (EEGs) recorded before phenytoin treatment were analyzed. Based on ictal EEGs from pharmacoresistant and pharmaco-responsive rats, a convolutional neural network predictive model was constructed to predict pharmacoresistance, and achieved 78% prediction accuracy. We further found the ictal EEGs from pharmacoresistant rats have a lower gamma-band power, which was verified in seizure EEGs from pharmacoresistant TLE patients. Prospectively, therapies targeting the subiculum in those predicted as "pharmacoresistant" individual rats significantly reduced the subsequent occurrence of pharmacoresistance. These results demonstrate a new methodology to predict whether TLE individuals become resistant to ASMs in a classic pharmacoresistant TLE model. This may be of translational importance for the precise management of pharmacoresistant TLE.
Epilepsy, Temporal Lobe/diagnosis* ; Animals ; Drug Resistant Epilepsy/drug therapy* ; Electroencephalography/methods* ; Rats ; Anticonvulsants/pharmacology* ; Neural Networks, Computer ; Male ; Humans ; Phenytoin/pharmacology* ; Adult ; Disease Models, Animal ; Female ; Rats, Sprague-Dawley ; Young Adult ; Convolutional Neural Networks

The accurate prediction of drug absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties represents a crucial step in early drug development for reducing failure risk. Current deep learning approaches face challenges with data sparsity and information loss due to single-molecule representation limitations and isolated predictive tasks. This research proposes molecular properties prediction with parallel-view and collaborative learning (MolP-PC), a multi-view fusion and multi-task deep learning framework that integrates 1D molecular fingerprints (MFs), 2D molecular graphs, and 3D geometric representations, incorporating an attention-gated fusion mechanism and multi-task adaptive learning strategy for precise ADMET property predictions. Experimental results demonstrate that MolP-PC achieves optimal performance in 27 of 54 tasks, with its multi-task learning (MTL) mechanism significantly enhancing predictive performance on small-scale datasets and surpassing single-task models in 41 of 54 tasks. Additional ablation studies and interpretability analyses confirm the significance of multi-view fusion in capturing multi-dimensional molecular information and enhancing model generalization. A case study examining the anticancer compound Oroxylin A demonstrates MolP-PC's effective generalization in predicting key pharmacokinetic parameters such as half-life (T0.5) and clearance (CL), indicating its practical utility in drug modeling. However, the model exhibits a tendency to underestimate volume of distribution (VD), indicating potential for improvement in analyzing compounds with high tissue distribution. This study presents an efficient and interpretable approach for ADMET property prediction, establishing a novel framework for molecular optimization and risk assessment in drug development.
Deep Learning

Paraplegia caused by spinal cord injury is a serious neurological complication, for which surgery is currently the main treatment method. Due to different surgical approaches, patients are usually expected to maintain a passive prone position for a long time or switch between the supine and prone positions. Affected by multiple factors such as neurogenic sensory disorders, pathological changes in muscle tone and operative duration, the risk of intraoperative acquired pressure injury (IAPI) is significantly increased. Current clinical prevention strategies for IAPI in these patients predominantly focus on localized pressure relief during positioning, lacking systematic, standardized comprehensive prevention protocols or evidence-based guidelines. To address it, Department of Nursing, Orthopedics Branch, China International Exchange and Promotive Association for Medical and Health Care, Spinal Trauma Professional Committee, Orthopedics Branch, Chinese Medical Doctor Association, Nursing Group of Spine and Spinal Cord Professional Committee of Chinese Association of Rehabilitation Medicine organized experts in relevant fields to formulate Guideline for the prevention of intraoperative acquired pressure injury in paraplegic patients with spinal cord injury ( version 2025), based on evidence-based medical evidence and latest research results and clinical practice at home and abroad. Eleven recommendations were put forward from the aspects of preoperative risk assessment, intraoperative prevention strategies, postoperative handover and monitoring, and supportive mechanisms for IAPI prevention, aiming to standardize the prevention measures and management strategies of IAPI in paraplegic patients with spinal cord injury and accelerate the recovery of patients and improve the therapeutic effect.

Panax ginseng, a perennial herbaceous plant and a representative of the Panax genus, is renowned for its exceptional medicinal value and economic benefits, often referred to as the “King of Herbs.” With the increasing market demand and the limited availability of suitable cultivation land, the issue of continuous cropping obstacles for P. ginseng has become increasingly prominent, directly hindering the sustainable development of the ginseng industry. This article summarizes the concept and hazards of continuous cropping obstacles and, drawing on the latest research, provides an in-depth analysis of the causes and response mechanisms. This work aims to establish a solid foundation for future research into the mechanisms of continuous cropping obstacles in P. ginseng.

Panax ginseng, a perennial herbaceous plant and a representative of the Panax genus, is renowned for its exceptional medicinal value and economic benefits, often referred to as the “King of Herbs.” With the increasing market demand and the limited availability of suitable cultivation land, the issue of continuous cropping obstacles for P. ginseng has become increasingly prominent, directly hindering the sustainable development of the ginseng industry. This article summarizes the concept and hazards of continuous cropping obstacles and, drawing on the latest research, provides an in-depth analysis of the causes and response mechanisms. This work aims to establish a solid foundation for future research into the mechanisms of continuous cropping obstacles in P. ginseng.

Panax ginseng, a perennial herbaceous plant and a representative of the Panax genus, is renowned for its exceptional medicinal value and economic benefits, often referred to as the “King of Herbs.” With the increasing market demand and the limited availability of suitable cultivation land, the issue of continuous cropping obstacles for P. ginseng has become increasingly prominent, directly hindering the sustainable development of the ginseng industry. This article summarizes the concept and hazards of continuous cropping obstacles and, drawing on the latest research, provides an in-depth analysis of the causes and response mechanisms. This work aims to establish a solid foundation for future research into the mechanisms of continuous cropping obstacles in P. ginseng.

OBJECTIVE To explore and evaluate the machine learning model for prediction of bacterial bloodstream infections established based on routine test data.METHODS By means of retrospective survey,a total of 5 421 pa-tients who were hospitalized in 3 medical institutions from Jan.2015 to Dec.2022 were recruited as the research subjects,1 914 of whom were assigned as the bloodstream infection group,and 3 507 were assigned as the non-bloodstream infection group.The baseline data including gender and age and the results of routine laboratory tests were collected from the enrolled patients.The 3 types of machine learning algorithms,logistic regression,support vector machine and random forest,were respectively used for the screening of the optimal prediction model;the contribution of feature variables to the predictive capability of the model was interpreted through SHAP.The fea-ture variables of the model were optimized by using recursive feature elimination method,and the predictive effi-ciency of the model was evaluated by the area under the curve(AUC)of receiver operating characteristic(ROC)curves.RESULTS Totally 26 variables involving age,gender and blood routine test indexes were included.The random forest was chosen as the optimal machine learning algorithm for the establishment of prediction model for bloodstream infections,and the accuracy of the model was 0.709,with the AUC 0.706.The result of SHAP ex-planation indicated that the age,hematokrit and erythrocyte volume distribution width-CV had remarkable effect on the model's making right decisions.17 variables of the prediction model showed more remarkable effect than 26 variable on distinguishing from the gram-positive bacteria bloodstream infections from the gram-negative bacteria bloodstream infections,with the AUC 0.715,the sensitivity 0.701,the specificity 0.632.CONCLUSIONS The prediction model that is established based on the blood routine test indexes by machine learning algorithm can pre-dict the bacterial bloodstream infection.Meanwhile,the feature selection strategy can further improve the predic-tive efficiency of the model on basis of lowering the dimensionality.

OBJECTIVE To explore and evaluate the machine learning model for prediction of bacterial bloodstream infections established based on routine test data.METHODS By means of retrospective survey,a total of 5 421 pa-tients who were hospitalized in 3 medical institutions from Jan.2015 to Dec.2022 were recruited as the research subjects,1 914 of whom were assigned as the bloodstream infection group,and 3 507 were assigned as the non-bloodstream infection group.The baseline data including gender and age and the results of routine laboratory tests were collected from the enrolled patients.The 3 types of machine learning algorithms,logistic regression,support vector machine and random forest,were respectively used for the screening of the optimal prediction model;the contribution of feature variables to the predictive capability of the model was interpreted through SHAP.The fea-ture variables of the model were optimized by using recursive feature elimination method,and the predictive effi-ciency of the model was evaluated by the area under the curve(AUC)of receiver operating characteristic(ROC)curves.RESULTS Totally 26 variables involving age,gender and blood routine test indexes were included.The random forest was chosen as the optimal machine learning algorithm for the establishment of prediction model for bloodstream infections,and the accuracy of the model was 0.709,with the AUC 0.706.The result of SHAP ex-planation indicated that the age,hematokrit and erythrocyte volume distribution width-CV had remarkable effect on the model's making right decisions.17 variables of the prediction model showed more remarkable effect than 26 variable on distinguishing from the gram-positive bacteria bloodstream infections from the gram-negative bacteria bloodstream infections,with the AUC 0.715,the sensitivity 0.701,the specificity 0.632.CONCLUSIONS The prediction model that is established based on the blood routine test indexes by machine learning algorithm can pre-dict the bacterial bloodstream infection.Meanwhile,the feature selection strategy can further improve the predic-tive efficiency of the model on basis of lowering the dimensionality.

Paraplegia caused by spinal cord injury is a serious neurological complication, for which surgery is currently the main treatment method. Due to different surgical approaches, patients are usually expected to maintain a passive prone position for a long time or switch between the supine and prone positions. Affected by multiple factors such as neurogenic sensory disorders, pathological changes in muscle tone and operative duration, the risk of intraoperative acquired pressure injury (IAPI) is significantly increased. Current clinical prevention strategies for IAPI in these patients predominantly focus on localized pressure relief during positioning, lacking systematic, standardized comprehensive prevention protocols or evidence-based guidelines. To address it, Department of Nursing, Orthopedics Branch, China International Exchange and Promotive Association for Medical and Health Care, Spinal Trauma Professional Committee, Orthopedics Branch, Chinese Medical Doctor Association, Nursing Group of Spine and Spinal Cord Professional Committee of Chinese Association of Rehabilitation Medicine organized experts in relevant fields to formulate Guideline for the prevention of intraoperative acquired pressure injury in paraplegic patients with spinal cord injury ( version 2025), based on evidence-based medical evidence and latest research results and clinical practice at home and abroad. Eleven recommendations were put forward from the aspects of preoperative risk assessment, intraoperative prevention strategies, postoperative handover and monitoring, and supportive mechanisms for IAPI prevention, aiming to standardize the prevention measures and management strategies of IAPI in paraplegic patients with spinal cord injury and accelerate the recovery of patients and improve the therapeutic effect.

Objective:To explore the protective effect and mechanism of Tongfu Lifei decoction (TFL) on human monocytic leukemia cell THP-1 induced by lipopolysaccharide (LPS).Methods:① THP-1 cells were cultured in vitro, and incubated with 1 mg/L LPS for 18 hours to construct an in vitro THP-1 cell inflammation model. Other THP-1 cells were taken as blank control group. Enzyme-linked immunosorbent assay (ELISA) was used to detect the levels of tumor necrosis factor-α(TNF-α) and interleukin-6 (IL-6) secreted by cells. ② THP-1 cells were divided into seven groups and treated with 0, 0.005, 0.01, 0.02, 0.04, 0.08, and 0.16 mL/mL TFL for 24 hours (added different dosages of TFL solution per milliliter of culture medium, with a crude drug content of 1 kg/L). The cell survival rate was detected using methyl thiazolyl tetrazolium (MTT) colorimetric method, and the intervention dosage of TFL for its non-toxic effect on THP-1 cells was screened. ③ Another THP-1 cells were divide into inflammatory model group and 0.01, 0.02, and 0.04 mL/mL TFL groups according to the intervention dosage of TFL screened by MTT colorimetry. After 24 hours of intervention, the levels of TNF-α and IL-6 secreted by cells were measured using ELISA. Western blotting was used to detect the expressions of programmed death-1/programmed death-ligand 1 (PD-1/PD-L1) signaling pathway proteins in cells. Real time fluorescence quantitative reverse transcription-polymerase chain reaction (RT-PCR) was used to detect the expressions of microRNAs (miR-146a, miR-146b, miR-155) in cells. ④ The maximum non-toxic concentration of TFL (0.04 mL/mL) on the THP-1 cell was selected as the intervention dose. THP-1 cells were divided into inflammation model group, TFL group, TFL+miR-146a inhibitor group, TFL+miR-146b inhibitor group, and TFL+miR-155 inhibitor group. The inflammation model group was not given any drug intervention. The other inhibitor groups were added 100 nmol/L corresponding inhibitor. After 24 hours of intervention, the levels of TNF-α and IL-6 secreted by cells were measured using ELISA. Western blotting was used to detect the expressions of PD-1/PD-L1 signaling pathway proteins in cells. Results:① Compared with the blank control group, the levels of TNF-α and IL-6 secreted by cells in the inflammatory model group were significantly increased, indicating the successful construction of the THP-1 inflammatory cell model in vitro. ② 0-0.04 mL/mL TFL had no toxic effect on THP-1 cells. However, the survival rates of cells in the 0.08 mL/mL and 0.16 mL/mL TFL groups were significantly lower than those in the inflammation model group, indicating that TFL dosages exceeding 0.04 mL/mL had toxic effects on THP-1 cells. ③ Compared with the inflammation model group, 0.01 mL/mL TFL had no significant effect on the levels of TNF-α and IL-6 secreted by THP-1 cells, while intervention with 0.02 mL/mL and 0.04 mL/mL TFL significantly reduced the levels of TNF-α and IL-6 secreted by cells [TNF-α(ng/L): 95.89±8.55, 70.73±11.70 vs. 137.10±7.19, IL-6 (ng/L): 23.03±2.55, 16.58±1.72 vs. 32.60±2.55, all P < 0.01]. Compared with the inflammation model group, the expressions of PD-1/PD-L1 signaling pathway proteins in THP-1 cells in different dosages of TFL groups were significantly reduced, and showed a certain dosage dependence. The expressions of the pathway proteins in the 0.04 mL/mL TFL group were significantly lower than those in the inflammation model group [PD-1 protein (PD-1/β-actin): 0.28±0.04 vs. 1.00±0.10, PD-L1 protein (PD-L1/β-actin): 0.54±0.05 vs. 1.00±0.08, phosphoinositide 3-kinase (PI3K) protein (PI3K/β-actin): 0.28±0.03 vs. 1.00±0.08, phosphorylated protein kinase B (p-Akt) protein (p-Akt/Akt): 0.38±0.04 vs. 1.00±0.10, all P < 0.01]. Compared with the inflammation model group, the expression of miR-146a in THP-1 cells in the 0.01, 0.02, and 0.04 mL/mL TFL groups was significantly reduced (2 -ΔΔCt: 0.46±0.11, 0.31±0.13, 0.23±0.14 vs. 1.01±0.18, all P < 0.01), while there was no significant change in the expressions of miR-146b and miR-155. ④ Compared with the inflammation model group, the TFL group showed a significant decrease in the levels of TNF-α and IL-6 secreted by THP-1 cells. The miR-146a inhibitor could significantly reverse the inhibitory effect of TFL on inflammatory factors, and the difference was statistically significant as compared with the TFL group [TNF-α (ng/L): 138.55±10.30 vs. 72.33±10.59, IL-6 (ng/L): 31.35±3.98 vs. 15.75±3.76, both P < 0.01]. Compared with the inflammation model group, the expressions of PD-1/PD-L1 signaling pathway proteins in THP-1 cells in the TFL group were significantly reduced. The expressions of pathway proteins in cells in the TFL+miR-146a inhibitor group were significantly higher than those in the TFL group [PD-1 protein (PD-1/β-actin): 0.85±0.09 vs. 0.37±0.04, PD-L1 protein (PD-L1/β-actin): 0.83±0.08 vs. 0.55±0.06, PI3K protein (PI3K/β-actin): 0.85±0.09 vs. 0.63±0.06, p-Akt protein (p-Akt/Akt): 0.98±0.10 vs. 0.75±0.07, all P < 0.05]. Conclusion:TFL regulates the expression of miR-146a to inhibit the PD-1/PD-L1 signaling pathway in THP-1 cells, regulates the immune barrier of sepsis induced in cell inflammation model in vitro, and thus protects LPS induced THP-1 cells.