With the continuous deepening of the concept of precision diagnosis and treatment for lung cancer, how to achieve higher efficiency and accuracy in the screening, diagnosis, and treatment pathways in clinical practice has become an important issue that urgently needs to be overcome. The current clinical difficulty lies in the fact that despite continuous advancements in imaging and molecular diagnostic technologies, there are still limitations in manual efficiency and subjective experience when it comes to massive data analysis and multi-scale feature extraction. Artificial intelligence (AI), especially algorithm systems based on deep learning, is an innovative technology capable of deeply empowering medical big data. This method utilizes algorithms such as convolutional neural networks, combined with radiomics, pathomics, and multi-modal data fusion analysis, demonstrating immense potential in early precise detection and benign-malignant differentiation of pulmonary nodules, digital pathological subtype recognition and non-invasive prediction of driver genes, precise 3D surgical planning and automatic delineation of radiotherapy target volumes, as well as dynamic risk warning during follow-up. This innovative technology provides a brand-new solution for realizing intelligent and individualized lung cancer diagnosis and treatment models. This consensus, based on the latest evidence from evidence-based medicine and combined with the development trends in the AI field and real-world clinical needs, was ultimately formed by gathering the consensus opinions of multidisciplinary experts in radiology, pathology, thoracic surgery, and other fields. The main content covers the application specifications of AI in the three core scenarios of lung cancer screening, diagnosis, and treatment, the technical standards for data collection and algorithm validation, as well as the ethical and regulatory challenges faced at the current stage. It aims to clarify the applicable boundaries of AI as a clinical auxiliary decision support tool, providing scientific guidance and standardized exploration directions for peers currently engaged in or planning to carry out AI-assisted clinical diagnosis, treatment, and translation of lung cancer.

OBJECTIVE: To investigate the regulatory effect of electroacupuncture (EA) at "Neiguan" (PC6) on mitochondrial autophagy in rats with myocardial ischemia-reperfusion injury (MIRI) at different phases (ischemia and reperfusion phases), and to explore the bidirectional regulatory effects of EA at "Neiguan" (PC6) and its potential mechanism. METHODS: Forty-five male SD rats were randomly divided into 6 groups according to the random number table method, namely, sham-operation group (n=9), model-A group (n=6), model-B group (n=9), EA-A1 group (n=6), EA-B1 group (n=6), and EA-B2 group (n=9). Except the rats in the sham-operation group, the MIRI model was established in the other groups with the physical ligation and tube pushing method. In the model-A group, the samples were collected directly after ligation, and in the model-B group, the samples were collected after ligation and reperfusion. In the EA-A1 group, EA was delivered while the ligation was performed, and afterwards, the samples were collected. In the EA-B1 group, while the ligation was performed, EA was operated at the same time, and after reperfusion, the samples were collected. In the EA-B2 group, during ligation and the opening of the left anterior descending branch of the coronary artery, EA was delivered, and after reperfusion, the samples were collected. EA was performed at bilateral "Neiguan" (PC6), with a disperse-dense wave, a frequency of 2 Hz/100 Hz, a current of 1 mA, and a duration of 30 min. HE staining was employed to observe the morphology of cardiomyocytes, TUNEL was adopted to detect the apoptosis of cardiomyocytes, transcriptome sequencing was to detect the differentially expressed genes in the left ventricle, JC-1 flow cytometry was to detect the mitochondrial membrane potential (MMP) of cardiomyocytes, Western blot was to detect the protein expression of phosphatase and tensin homolog-induced kinase 1 (Pink1), Parkin and p62 in the left ventricle of rats, and ELISA was to detect the levels of serum creatine kinase isoenzyme (CK-MB) and cardiac troponin I (cTn-I) in the rats. RESULTS: Compared with the sham-operation group, the cardiomyocytes of rats in the model-B group were severely damaged, with disordered arrangement, unclear boundaries, broken muscle fibers, edema and loose distribution; and the cardiomyocytes in the EA-B2 group were slightly damaged, the cell structure was partially unclear, the cells were arranged more regularly, and the intact cardiomyocytes were visible. Compared with the sham-operation group, the apoptosis of cardiomyocytes increased in the model-B group (P<0.001); and when compared with the model-B group, the apoptosis alleviated in the EA-B2 group (P<0.001). The differentially expressed genes among the EA-B2 group, the sham-operation group and the model-B group were closely related to cell autophagy and mitochondrial autophagy. Compared with the sham-operation group, MMP of cardiomyocytes was reduced (P<0.001), the protein expression of Pink1, Parkin, and p62 of the left ventricle and the levels of serum CK-MB and cTn-I were elevated in the model B group (P<0.001). In comparison with model-A group, the MMP of cardiomyocytes and the levels of serum CK-MB and cTn-I were reduced (P<0.001, P<0.05), and the protein expression of Pink1 in the left ventricle rose in the EA-A1 group (P<0.01). Compared with the model-B group, MMP of cardiomyocytes increased (P<0.001), the protein expression of Pink1, Parkin, and p62 of the left ventricle, and the levels of serum CK-MB and cTn-I decreased (P<0.001) in the EA-B1 group and the EA-B2 group. When compared with the EA-A1 group, MMP of cardiomyocytes increased (P<0.001), and the protein expression of Pink1, Parkin, and p62 of the left ventricle, and the levels of serum CK-MB and cTn-I decreased in the EA-B1 group (P<0.01). CONCLUSION EA at "Neiguan" (PC6) can ameliorate MIRI in rats, which may be achieved through the Pink1/Parkin-mediated mitochondrial autophagy pathway. EA can alleviate myocardial injury by enhancing mitochondrial autophagy at the ischemia phase, and it can reduce reperfusion injury by weakening mitochondrial autophagy at the reperfusion phase.
Animals ; Electroacupuncture ; Male ; Myocardial Reperfusion Injury/metabolism* ; Rats, Sprague-Dawley ; Rats ; Acupuncture Points ; Autophagy ; Humans ; Mitochondria/genetics*

Nasojejunal feeding tubes are widely used in surgical,intensive care,and older patients.Manual blind insertion of nasojejunal feeding tubes is technically challenging,associated with a high failure rate,and prone to complications.The primary causes of suboptimal placement outcomes are the uncertainty and weak controllability of the interaction forces during the coordination between manual posterior advancement and the patient's physiological state.While current auxiliary techniques such as X-ray,ultrasound,and endoscopy can improve the success rate of nasojejunal tube placement and reduce complications to some extent,the accuracy and safety of placement remain constrained by challenges in controlling insertion forces and achieving precise positional localization.Robotic technology holds promise for addressing the uncertainties and controllability issues inherent in the placement process.By leveraging precise sensing,real-time navigation,and efficient control,robots can achieve intelligent positioning and precise control over the direction and location of the catheter tip during nasojejunal intubation.However,current research on robotic applications for nasojejunal feeding tube placement is still in an early stage,facing challenges such as high costs,operational complexity,and concerns over safety and reliability.Herein,we analyzed the limitations and causes of failure in existing placement methods and explored the application prospects of robotic technologies for precise control and intelligent positioning in nasojejunal feeding tube placement.The paper provides new insights for developing nursing techniques that enable safer and more effective,comfortable,and rapid intubation.Future efforts should focus on deepening the integration of artificial intelligence and robotics,optimizing drive technologies,and accelerating the translation of these technologies from the laboratory to clinical practice.This will drive the advancement of nasojejunal feeding tube placement techniques towards intelligent,precise,and accessible solutions.

OBJECTIVE: To explore the therapeutic effects and underlying mechanisms of Dendrobium officinale (Tiepi Shihu) extract (DOE) on insomnia. METHODS: Forty-two male Sprague-Dawley rats were randomly divided into 6 groups (n=7 per group): normal control, model control, melatonin (MT, 40 mg/kg), and 3-dose DOE (0.25, 0.50, and 1.00 g/kg) groups. Rats were raised in a strong-light (10,000 LUX) and -noise (>80 db) environment (12 h/d) for 16 weeks to induce insomnia, and from week 10 to week 16, MT and DOE were correspondingly administered to rats. The behavior tests including sodium pentobarbital-induced sleep experiment, sucrose preference test, and autonomous activity test were used to evaluate changes in sleep and emotions of rats. The metabolic-related indicators such as blood pressure, blood viscosity, blood glucose, and uric acid in rats were measured. The pathological changes in the cornu ammonis 1 (CA1) region of rat brain were evaluated using hematoxylin and eosin staining and Nissl staining. Additionally, the sleep-related factors gamma-aminobutyric acid (GABA), glutamate (GA), 5-hydroxytryptamine (5-HT), and interleukin-6 (IL-6) were measured using enzyme linked immunosorbent assay. Finally, we screened potential sleep-improving receptors of DOE using polymerase chain reaction (PCR) array and validated the results with quantitative PCR and immunohistochemistry. RESULTS: DOE significantly improved rats' sleep and mood, increased the sodium pentobarbital-induced sleep time and sucrose preference index, and reduced autonomic activity times (P<0.05 or P<0.01). DOE also had a good effect on metabolic abnormalities, significantly reducing triglyceride, blood glucose, blood pressure, and blood viscosity indicators (P<0.05 or P<0.01). DOE significantly increased the GABA content in hippocampus and reduced the GA/GABA ratio and IL-6 level (P<0.05 or P<0.01). In addition, DOE improved the pathological changes such as the disorder of cell arrangement in the hippocampus and the decrease of Nissel bodies. Seven differential genes were screened by PCR array, and the GABA_A receptors (Gabra5, Gabra6, Gabrq) were selected for verification. The results showed that DOE could up-regulate their expressions (P<0.05 or P<0.01). CONCLUSION DOE demonstrated remarkable potential for improving insomnia, which may be through regulating GABA_A receptors expressions and GA/GABA ratio.
Animals ; Dendrobium/chemistry* ; Rats, Sprague-Dawley ; Male ; Sleep Initiation and Maintenance Disorders/blood* ; Plant Extracts/therapeutic use* ; Receptors, GABA-A/metabolism* ; Noise/adverse effects* ; Light/adverse effects* ; gamma-Aminobutyric Acid/metabolism* ; Sleep/drug effects* ; Rats ; Receptors, GABA/metabolism*

Computational analysis can accurately detect drug-gene interactions (DGIs) cost-effectively. However, transductive learning models are the hotspot to reveal the promising performance for unknown DGIs (both drugs and genes are present in the training model), without special attention to the unseen DGIs (both drugs and genes are absent in the training model). In view of this, this study, for the first time, proposed an inductive learning-based model for the precise identification of unseen DGIs. In our study, by integrating disease nodes to avoid data sparsity, a multi-relational drug-disease-gene (DDG) graph was constructed to achieve effective fusion of data on DDG intro-relationships and inter-actions. Following the extraction of graph features by utilizing graph embedding algorithms, our next step was the retrieval of the attributes of individual gene and drug nodes. In this way, a hybrid feature characterization was represented by integrating graph features and node attributes. Machine learning (ML) models were built, enabling the fulfillment of transductive predictions of unknown DGIs. To realize inductive learning, this study generated an innovative idea of transforming known node vectors derived from the DDG graph into representations of unseen nodes using node similarities as weights, enabling inductive predictions for the unseen DGIs. Consequently, the final model was superior to existing models, with significant improvement in predicting both external unknown and unseen DGIs. The practical feasibility of our model was further confirmed through case study and molecular docking. In summary, this study establishes an efficient data-driven approach through the proposed modeling, suggesting its value as a promising tool for accelerating drug discovery and repurposing.

Objectives:To investigate the impact of bone hyperplasia in the thoracolumbar spine caused by diffuse idiopathic skeletal hyperostosis(DISH)on the changes of spinopelvic sagittal parameters between stand-ing and sitting positions.Methods:A total of 61 DISH patients[DISH group,42 males and 19 females,50-76(65.1±6.3)years]who underwent surgical treatment for lumbar spinal stenosis in our hospital between Jan-uary 2019 and December 2023 were retrospectively analyzed.100 age-and sex-matched non-DISH patients undergone the same surgical procedure during the same period were included as control[N-DISH group,63 males and 37 females,54-77(67.5±7.2)years].According to the distribution of osteophytes,the patients in the DISH group with ectopic ossification limited to the thoracic spine were categorized into the T-DISH group,while those with involvement of both thoracic and lumbar spines were divided in the L-DISH group.Preoper-atively,full-spine anteroposterior and lateral X-rays were taken in both standing and sitting positions.All patients were measured for spinopelvic sagittal parameters in standing and sitting positions,including sagittal vertical axis(SVA),pelvic tilt(PT),sacral slope(SS),pelvic incidence(PI),thoracic kyphosis(TK),lumbar lordosis(LL),and proximal femoral angle(PFA).The differences in standing and sitting positions and changes between DISH and N-DISH groups,T-DISH and L-DISH groups were compared.Results:In the standing position,the SVA(P=0.008)and TK(P=0.028)in the DISH group were significantly higher than those in the N-DISH group,while no significant differences were observed in PI,PT,SS,LL,and PFA(P＞0.05).In the sitting position,the TK(P=0.003)and LL(P=0.007)in the DISH group were significantly higher than those in the N-DISH group,whereas no significant differences were noted inSVA,PT,PI,SS,and PFA(P＞0.05).When transitioning from standing to sitting,the changes in SVA(P=0.021),PT(P=0.008),SS(P=0.001),TK(P=0.002),and LL(P＜0.001)in DISH group of patients were significantly smaller than those in N-DISH group of patients.Among DISH patients,the L-DISH group had significantly lower PI(P=0.016),SS(P=0.011),and LL(P=0.006)in the standing position compared to the T-DISH group,while no significant differences were observed in SVA,PT,TK,and PFA(P＞0.05).In the sitting position,the PI(P=0.008)and SS(P=0.007)of the L-DISH group were significantly lower than those of the T-DISH group,while no significant differences were observed in SVA,PT,TK,LL,and PFA(P＞0.05).The changes in LL when transitioning from standing to sitting were significantly lower in the L-DISH group compared to the T-DISH group(P=0.033),while the changes of other sagittal parameters showed no significant difference(P＞0.05).Conclusions:Bone hyperplasia in DISH patients significantly limits spinal mobility,and the restriction is more pronounced in patients with osteophytes extending to the lumbar spine compared to those with isolated thoracic involvement.

Computational analysis can accurately detect drug-gene interactions(DGIs)cost-effectively.However,transductive learning models are the hotspot to reveal the promising performance for unknown DGIs(both drugs and genes are present in the training model),without special attention to the unseen DGIs(both drugs and genes are absent in the training model).In view of this,this study,for the first time,proposed an inductive learning-based model for the precise identification of unseen DGIs.In our study,by integrating disease nodes to avoid data sparsity,a multi-relational drug-disease-gene(DDG)graph was constructed to achieve effective fusion of data on DDG intro-relationships and inter-actions.Following the extraction of graph features by utilizing graph embedding algorithms,our next step was the retrieval of the attributes of individual gene and drug nodes.In this way,a hybrid feature charac-terization was represented by integrating graph features and node attributes.Machine learning(ML)models were built,enabling the fulfillment of transductive predictions of unknown DGIs.To realize inductive learning,this study generated an innovative idea of transforming known node vectors derived from the DDG graph into representations of unseen nodes using node similarities as weights,enabling inductive predictions for the unseen DGIs.Consequently,the final model was superior to existing models,with significant improvement in predicting both external unknown and unseen DGIs.The practical feasibility of our model was further confirmed through case study and molecular docking.In summary,this study establishes an efficient data-driven approach through the proposed modeling,suggesting its value as a promising tool for accelerating drug discovery and repurposing.

Objective To explore the resistance mechanism of tigecycline insensitivity of multidrug-resistant Acine-tobacter baumannii,and provide reference for clinical rational antimicrobial use as well as prevention and control of healthcare-associated infection.Methods 22 strains of tigecycline insensitive multidrug-resistant Acinetobacter bau-mannii(TIS-MDR-AB)and 22 strains of tigecycline sensitive multidrug-resistant Acinetobacter baumannii(TS-MDR-AB)isolated clinically from the First Affiliated Hospital of Guangzhou Medical University from April 2022 to May 2023 were collected.Efflux pump phenotype inhibition assay was performed using efflux pump inhibitor car-bonyl cyanide m-chlorophenylhydrazone(CCCP).The main efflux pump genes(adeB,adeG,adeJ),as well as tigecycline-resistant gene tet(X),were screened by polymerase chain reaction(PCR)technique,and their mRNA expression levels were detected by real-time fluorescence quantitative PCR.Mutations in the efflux pump regulatory gene adeRS were analyzed by Sanger sequencing analysis.Results The detection rates of efflux pump genes adeB,adeG and adeJ were all above 95％in two MDR-AB groups,and tet(X)gene was not detected.Efflux pump inhibi-tor assay showed that the minimum inhibitory concentration(MIC)of TIS-MDR-AB strains decreased after adding CCCP,3 strains showed positive efflux pump phenotype.The mRNA expression level of MDR-AB adeB in the TIS-MDR-AB group was higher than that in the TS-MDR-AB group(P＜0.01),while the expression of adeG and adeJ genes was no statistically different.Multiple mutations were found in the adeR and adeS genes,the adeS of 2 strains was inserted ISAba1,and 3 strains were inserted ISAba13.Conclusion The overexpression of adeABC in the efflux pump system may play an important role in the mechanism of reduced sensitivity of MDR-AB to tigecycline,and its overexpression may be related to the insertion sequence or mutation in the efflux pump regulatory gene adeRS.