ObjectiveTo investigate the effects and underlying mechanisms of Shenqi Wenfei prescription （SQWF） on chronic obstructive pulmonary disease （COPD）. MethodsA rat model of COPD with lung Qi deficiency was established using lipopolysaccharide （LPS） combined with cigarette smoke. Forty-eight SD rats were randomly divided into a blank group， a model group， low-， medium-， and high-dose SQWF groups （2.835， 5.67， 11.34 g·kg^-1）， and a Yupingfeng group （1.35 g·kg^-1）. Drug administration began on day 29 after modeling and continued for 2 weeks. The general condition of the rats was observed， and the lung function in each group was assessed. Hematoxylin-eosin （HE） staining was used to observe pathological changes in lung tissue. The proportion of inflammatory cells in bronchoalveolar lavage fluid （BALF） was measured. Apoptosis in lung tissue was examined by terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling （TUNEL） staining. The release level of lactate dehydrogenase （LDH） in BALF was detected by a microplate assay. Reactive oxygen species （ROS） levels in lung tissue were detected using fluorescent probes. The levels of malondialdehyde （MDA）， total superoxide dismutase （SOD）， and reduced glutathione （GSH） in BALF were measured by biochemical methods. Ultrastructural changes in lung cells were observed via transmission electron microscopy. Double immunofluorescence staining was performed to detect the expression of thioredoxin-interacting protein （TXNIP） and nucleotide-binding oligomerization domain-like receptor protein 3 （NLRP3） in lung tissue. Western blot analysis was used to detect the protein expression of TXNIP， NLRP3， apoptosis-associated speck-like protein containing a CARD （ASC）， cysteinyl aspartate-specific protease-1 （Caspase-1）， Caspase-1 p20， gasdermin D （GSDMD）， GSDMD N-terminal active fragment （GSDMD-N）， interleukin-1β （IL-1β）， and IL-18 in lung tissue. Serum IL-1β and IL-18 levels were measured by ELISA. ResultsCompared with the blank group， the model group showed lassitude， fatigue， tachypnea， and audible phlegm sounds， and lung function significantly declined （P0.01）. Pulmonary emphysema and inflammatory cell infiltration were obvious. The level of inflammatory cells in BALF increased significantly （P0.05）. The number of TUNEL-positive cells increased （P0.01）. Levels of LDH， ROS， and MDA in BALF increased significantly （P0.01）， while GSH and SOD activities decreased significantly （P0.01）. Lung tissue cells showed irregular morphology， swollen mitochondria， disrupted cell membranes， and abundant vesicles， i.e.， pyroptotic bodies. Protein levels of TXNIP， NLRP3， ASC， Caspase-1， Caspase-1 p20， GSDMD， GSDMD-N， IL-1β， and IL-18 in lung tissue were significantly elevated （P0.01）， and serum IL-1β and IL-18 levels also increased significantly （P0.01）. Compared with the model group， each medication group showed alleviation of qi deficiency symptoms and improved lung function （P0.01）. Pulmonary emphysema and inflammatory cell infiltration were reduced. Inflammatory cell levels decreased （P0.05）. The number of TUNEL-positive cells decreased significantly （P0.01）. Levels of LDH， ROS， and MDA decreased significantly （P0.05）， while GSH and SOD activities significantly increased （P0.01）. Morphological and structural damage in lung tissue was improved to varying degrees. Protein levels of TXNIP， NLRP3， ASC， Caspase-1， Caspase-1 p20， GSDMD， GSDMD-N， IL-1β， and IL-18 in lung tissue significantly decreased （P0.01）， and serum IL-1β and IL-18 levels also decreased significantly （P0.05）. ConclusionSQWF can improve lung function and alleviate inflammatory responses in COPD rats. Its mechanism may be related to regulating the ROS/TXNIP/NLRP3 pathway and inhibiting pyroptosis.

ObjectiveTo investigate the effects and underlying mechanisms of Shenqi Wenfei prescription （SQWF） on chronic obstructive pulmonary disease （COPD）. MethodsA rat model of COPD with lung Qi deficiency was established using lipopolysaccharide （LPS） combined with cigarette smoke. Forty-eight SD rats were randomly divided into a blank group， a model group， low-， medium-， and high-dose SQWF groups （2.835， 5.67， 11.34 g·kg^-1）， and a Yupingfeng group （1.35 g·kg^-1）. Drug administration began on day 29 after modeling and continued for 2 weeks. The general condition of the rats was observed， and the lung function in each group was assessed. Hematoxylin-eosin （HE） staining was used to observe pathological changes in lung tissue. The proportion of inflammatory cells in bronchoalveolar lavage fluid （BALF） was measured. Apoptosis in lung tissue was examined by terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling （TUNEL） staining. The release level of lactate dehydrogenase （LDH） in BALF was detected by a microplate assay. Reactive oxygen species （ROS） levels in lung tissue were detected using fluorescent probes. The levels of malondialdehyde （MDA）， total superoxide dismutase （SOD）， and reduced glutathione （GSH） in BALF were measured by biochemical methods. Ultrastructural changes in lung cells were observed via transmission electron microscopy. Double immunofluorescence staining was performed to detect the expression of thioredoxin-interacting protein （TXNIP） and nucleotide-binding oligomerization domain-like receptor protein 3 （NLRP3） in lung tissue. Western blot analysis was used to detect the protein expression of TXNIP， NLRP3， apoptosis-associated speck-like protein containing a CARD （ASC）， cysteinyl aspartate-specific protease-1 （Caspase-1）， Caspase-1 p20， gasdermin D （GSDMD）， GSDMD N-terminal active fragment （GSDMD-N）， interleukin-1β （IL-1β）， and IL-18 in lung tissue. Serum IL-1β and IL-18 levels were measured by ELISA. ResultsCompared with the blank group， the model group showed lassitude， fatigue， tachypnea， and audible phlegm sounds， and lung function significantly declined （P0.01）. Pulmonary emphysema and inflammatory cell infiltration were obvious. The level of inflammatory cells in BALF increased significantly （P0.05）. The number of TUNEL-positive cells increased （P0.01）. Levels of LDH， ROS， and MDA in BALF increased significantly （P0.01）， while GSH and SOD activities decreased significantly （P0.01）. Lung tissue cells showed irregular morphology， swollen mitochondria， disrupted cell membranes， and abundant vesicles， i.e.， pyroptotic bodies. Protein levels of TXNIP， NLRP3， ASC， Caspase-1， Caspase-1 p20， GSDMD， GSDMD-N， IL-1β， and IL-18 in lung tissue were significantly elevated （P0.01）， and serum IL-1β and IL-18 levels also increased significantly （P0.01）. Compared with the model group， each medication group showed alleviation of qi deficiency symptoms and improved lung function （P0.01）. Pulmonary emphysema and inflammatory cell infiltration were reduced. Inflammatory cell levels decreased （P0.05）. The number of TUNEL-positive cells decreased significantly （P0.01）. Levels of LDH， ROS， and MDA decreased significantly （P0.05）， while GSH and SOD activities significantly increased （P0.01）. Morphological and structural damage in lung tissue was improved to varying degrees. Protein levels of TXNIP， NLRP3， ASC， Caspase-1， Caspase-1 p20， GSDMD， GSDMD-N， IL-1β， and IL-18 in lung tissue significantly decreased （P0.01）， and serum IL-1β and IL-18 levels also decreased significantly （P0.05）. ConclusionSQWF can improve lung function and alleviate inflammatory responses in COPD rats. Its mechanism may be related to regulating the ROS/TXNIP/NLRP3 pathway and inhibiting pyroptosis.

ObjectiveThis paper aims to clarify the material basis of Gualou Niubangtang and establish a quantitative analysis method for its main constituents， providing a reference for the overall quality control of this preparation. MethodsThe constituents in the formula were systematically characterized based on ultra-performance liquid chromatography-quadrupole time-of-flight tandem mass spectrometry （UPLC-Q-TOF-MS/MS）. Identification was performed by matching with the UNIFI 9.6 software and utilizing database platforms such as PubChem， ChemicalBook， and ChemSpider， combined with relevant literature reports. A quantitative analysis method for the seven main constituents in Gualou Niubangtang was established by using high performance liquid chromatography （HPLC）. ResultsUPLC-Q-TOF-MS/MS analysis identified 155 constituents， including 69 flavonoids， 36 terpenoids， 23 phenylpropanoids， 8 phenylethanoid glycosides， and 19 other types of constituents. In the established quantitative analysis method， the seven main constituents showed good linearity within their respective linear ranges. The precision， repeatability， stability， and spike recovery all met the required standards. The results showed that the content ranges of geniposide， liquiritin， hesperidin， arctiin， baicalin， oroxylin A-7-O-β-D-glucuronide， and wogonoside in 15 batches of Gualou Niubangtang were 13.67-21.25， 1.20-7.64， 5.45-7.45， 22.97-33.51， 29.95-39.07， 2.58-4.80， and 6.56-9.31 mg·g^-1， respectively. ConclusionThis study successfully characterizes and attributes multi-category constituents in Gualou Niubangtang， clarifying that its material basis is primarily composed of flavonoids， terpenoids， phenylethanoid glycosides， and phenylpropanoids. Furthermore， it enables the quantification of seven constituents within the formula. This work lays a foundation for research on the quality control， action mechanism， and clinical application of this formula.

ObjectiveTo integrate network pharmacology prediction with multi-level experimental verification methods， and to explore in depth the therapeutic efficacy and potential mechanism of luteolin in treating gout. MethodsDatabases were used to obtain potential pharmacodynamic targets of luteolin. Protein-protein interaction （PPI） network construction and network pharmacology analysis techniques were used to screen key core targets of luteolin in gout treatment. Further biological function enrichment analysis and signaling pathway analysis were performed on these targets. Molecular docking simulation was used to calculate the binding energy between luteolin and potential core targets， clarifying the strength of their interactions. In the in vivo experiment for hyperuricemia， 48 mice were randomly divided into a blank group， a model group， an allopurinol group （5 mg·kg^-1）， and low-dose （10 mg·kg^-1）， medium-dose （30 mg·kg^-1）， and high-dose （90 mg·kg^-1） luteolin groups. For the first three days， the blank and model groups were gavaged with an equal volume of normal saline， while the allopurinol group and luteolin groups were gavaged with corresponding drugs. From day 4 onwards， modeling was performed by intraperitoneal injection at 12：00 daily （normal saline for the blank group， and oxonic acid potassium-hypoxanthine mixture for other groups， with 300 mg·kg^-1 for each group）. Gavage intervention was administered at 18：00 daily （normal saline for the blank/model groups， and corresponding drugs for the treatment groups） until day 7. After sampling， levels of serum uric acid （UA）， alanine aminotransferase （ALT）， and aspartate aminotransferase （AST） were measured. Levels of xanthine oxidase （XO） in the liver and kidney， ATP-binding cassette transporter G2 （ABCG2） and malondialdehyde （MDA） in the kidney， and superoxide dismutase （SOD） in the liver were determined. Renal HE staining was also performed. In the pharmacodynamic study of gouty arthritis， 36 rats were randomly divided into a blank group， a model group， a colchicine group （0.315 mg·kg^-1）， and low-dose （7 mg·kg^-1）， medium-dose （21 mg·kg^-1）， and high-dose （63 mg·kg^-1） luteolin groups. The model was established by vertically injecting 100 µL of 25 g·L^-1 monosodium urate suspension into the posterior lateral aspect of the right ankle joint （the blank group was injected with an equal volume of normal saline）， with repeated injections every two days for reinforcement. From day 2 after modeling， daily gavage administration was performed （normal saline for the blank/model groups， and corresponding drugs for the treatment groups） for a total of 16 days. During the experiment， ankle swelling and pain threshold were measured regularly. After sampling， levels of serum tumor necrosis factor-α （TNF-α）， interleukin-6 （IL-6）， and interleukin-1β （IL-1β） were determined. Ankle joints were subjected to HE， Masson， and safranin O-fast green staining， and HE staining was also performed on ankle synovial tissue and various organs. Western blot was used to determine the expression levels of key proteins in gout-related signaling pathways. ResultsNetwork pharmacology analysis predicted that luteolin may regulate over 20 core targets， such as XO， ABCG2， nuclear factor erythroid 2-related factor 2 （Nrf2）， and SOD， through acting on signaling pathways including NF-κB， phosphoinositide 3-kinase/protein kinase B （PI3K/Akt）， and ABC transporters， thereby affecting uric acid metabolism and inflammatory responses. In the hyperuricemia model， compared with the blank group， the model group showed significantly increased serum UA level， liver and kidney XO activity， renal ABCG2 expression， and liver SOD activity （P<0.01）. Compared with the model group， the high-dose luteolin group significantly reduced serum UA level （P<0.01）， inhibited liver and kidney XO activity （P<0.01）， and significantly increased renal ABCG2 expression and liver SOD activity （P<0.01）， effectively alleviating renal oxidative stress damage and improving renal histopathological status. In the gouty arthritis model， compared with the blank group， the model group showed significant ankle swelling， decreased pain threshold， and significantly increased levels of IL-6， IL-1β， and TNF-α in serum and synovial tissue （P<0.01）. The high-dose luteolin group significantly reduced ankle swelling， prolonged hot plate pain threshold， effectively decreased the levels of the above inflammatory factors in serum and synovial tissue （P<0.01）， and significantly improved ankle pathological damage， showing good analgesic and anti-inflammatory effects. Western blot results further confirmed that luteolin significantly upregulated Nrf2 protein expression and downregulated XO and nucleotide-binding oligomerization domain （NOD）-like receptor protein 3 （NLRP3） expression in animals. ConclusionLuteolin can improve symptoms of hyperuricemia and gouty arthritis， and its potential mechanism may be related to inhibiting XO activity， increasing ABCG2 and SOD levels， and regulating Nrf2-mediated oxidative stress-related pathways.

Stroke is the leading cause of death and disability among adults in China， and its common complications include digestive system abnormalities， cognitive impairment， depression， stroke-associated pneumonia， and hemiplegia. The combination of traditional Chinese and Western medicine has great potential in treating post-stroke complications. Xinglou Chengqitang （XLCQT） is a representative prescription of alleviating the disease in the upper part by treating the lower part. It has definite therapeutic effect and high safety. Clinically， XLCQT is often used to treat stroke and its complications. However， the quantity and quality of clinical trials of XLCQT in treating post-stroke complications need to be improved. Additionally， since the basic research is weak， the material basis and multi-target mechanism for the efficacy of this prescription are unknown. This article reviews XLCQT in terms of the pharmacodynamic basis， medicinal properties， safety evaluation， and progress in clinical research and mechanisms in treating post-stroke complications. This article summarizes 22 key active ingredients of XLCQT in treating acute stroke complicated with syndrome of phlegm heat and fu-organ excess. Among these key active ingredients， resveratrol， kaempferol， luteolin， chrysoeriol， apigenin， （+）-catechin， and adenosine have good pharmacokinetic properties and high bioavailability. The mechanisms of XLCQT in treating post-stroke complications are complex， including inflammatory response， brain-gut axis， hypothalamic-pituitary-adrenal （HPA） axis， intestinal flora， neurotrophic factors， autophagy， oxidative stress， and free radical damage. This review helps to deeply understand the pharmacodynamic basis and mechanisms of XLCQT in treating post-stroke complications and provides a theoretical basis for the clinical application of XLCQT against post-stroke complications and the development of drugs.

OBJECTIVE To introduce the development of an intelligent prescription review decision-support system for anti-tumor drugs and assess its clinical application outcomes. METHODS Relevant data sources， including national and local pharmaceutical administration policies， clinical practice guidelines/consensus， hospital information systems data， and genetic testing results， were integrated. Adhering to the principles of structure， standardization and dynamic updating， a knowledge base covering chemotherapeutic， targeted and immunotherapeutic agents was constructed using a dual-dimensional modeling approach that combined “drug attributes” and “clinical contexts”. This knowledge base was then embedded into the hospital’s electronic medical order system to establish the prescription review decision-support system. The application and performance of the system were evaluated at Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine. RESULTS A knowledge base containing 18 318 prescription review rules for anti-tumor drugs was constructed， and a closed-loop prescription review system was successfully established， encompassing pre-prescription real-time intervention， in-process interactive review， and post-prescription evaluation and analysis. From 2021 to 2024， the system generated a total of 57 879 alerts for prescriptions of five typical categories of anti-tumor drugs. For platinum-containing prescriptions， 22 577 alerts were generated， with Cisplatin for injection （lyophilized） being the most frequently alerted drug （13 445 alerts）， and “ototoxicity risk due to combined use” alerts remained high （7 682 alerts）. For methotrexate-containing prescriptions， 3 721 alerts were recorded， primarily related to “precaution-related issues” （76.4%， 2 843/3 721）. For doxorubicin-containing prescriptions， 17 301 alerts were triggered， primarily related to “dosage and administration” （14 315 alerts）. For human epidermal growth factor receptor 2-targeted agents-containing prescriptions， 1 007 alerts were issued， mostly related to “reimbursement restrictions” （956 alerts）. For programmed death-1/programmed death-ligand 1 inhibitors-containing prescriptions， the alerts increased year by year， totaling 13 273 alerts， primarily related to “inappropriate indication” （9 118 alerts）. Over the 4 years， the physician response rates to system alerts were 21.4%， 27.1%， 33.5% and 51.6%， respectively. CONCLUSIONS An intelligent decision-support system for anti-tumor drug prescription review， encompassing a closed-loop process of “real-time pre-event intervention， interactive in-event prescription review， post-event evaluation and analysis”， has been successfully constructed and implemented throughout the entire workflow. There is a discernible trend in this hospital， where the focus on monitoring anti-tumor drugs is shifting towards immunotherapy drugs. Additionally， the acceptance rate of physicians regarding prescription review opinions has been steadily increasing year by year.

ObjectiveTo investigate the protective effect of α-ketoglutarate （α-KG） on hepatic lipid deposition in offspring caused by arsenic exposure during pregnancy. Methods8-week-old institute of cancer research （ICR） mice were mated in a ratio of 2∶1 between females and males， and the detection of vaginal plugs confirmed pregnant. A total of 32 pregnant mice were randomly divided into four groups： control group， arsenic group， α-KG group， arsenic+α-KG group. On gestational day 0-16 （GD0-GD16）， the arsenic and arsenic+α-KG groups were exposed to sodium arsenite （NaAsO₂ ，15 mg/L） in drinking water everyday， and the α-KG and arsenic+α-KG groups were gavaged with α-KG （2 g/kg） everyday. On GD16， pregnant mice were euthanized to collect fetal liver， and fetal body weight and crown-rump length were measured. Gene expression differences between the control group and the arsenic group were analyzed by transcriptome. The total triglycerides （TGs） and subtypes in fetal liver were detected by liquid chromatography tandem mass spectrometry （LC-MS/MS）. Oil red O staining was used to observe the histopathological changes in the liver. Quantitative polymerase chain reaction （qPCR） was used to detect the expression level of genes related to lipid synthesis， transport， and degradation， and phosphatidylinositol 3' -kinase/ protein kinase B （PI3K/AKT） in the liver of fetus. ResultsTranscriptomics analysis showed that 2 144 genes were downregulated and 1 675 genes were upregulated in the arsenic exposed fetal liver； body weight and crown-rump length were reduced （P_TuKey<0.05）； the level of hepatic TGs was elevated in arsenic group （P_TuKey<0.05）； oil-red O staining showed a significant increase in lipid droplets in arsenic group （P_TuKey<0.01）； the expression of lipid synthesis-related genes were significantly upregulated （P_TuKey<0.05）； the expression of β-oxidation-related genes and lipid degradation-related genes were downregulated （P_TuKey<0.05）； the expression of PI3K， AKT decreased（P_TuKey<0.05）. Compared with the arsenic group， the body weight and crown-rump length of fetus increased in the arsenic+α-KG group （P_TuKey<0.05）； the level of hepatic TGs decreased in the arsenic+α-KG group （P_TuKey<0.05）； oil red O staining showed lipid droplets significantly decreased （P_TuKey<0.01）； the expression of lipid synthesis-related genes were downregulated （P_TuKey<0.05）， the expression of β-oxidation-related genes and lipid degradation-related genes were upregulated （P_TuKey<0.05）； the expression levels of PI3K and AKT increased （P_TuKey<0.05）. Conclusionα-KG alleviated hepatic lipid deposition in offspring exposed to arsenic during pregnancy through activating PI3K/AKT signaling pathway.

ObjectiveTo delineate imaging findings using an imaging platform and investigate the correlation between MRI characteristics of spinal cord atrophy and clinical diagnosis in children with spinal cord injury (SCI). MethodsImaging data of 150 children with SCI admitted to Beijing Bo'ai Hospital, China Rehabilitation Research Center, from January, 2002 to March, 2024 were collected and imported into the imaging platform. The anteroposterior and transverse diameters of the middle part of the spinal cord at the cross-section with the most severe atrophy were measured, and the relevant indicators of the previous normal spinal cord segment were measured as controls; the radiomic features were extracted. Clinical data of the children including gender, age, cause of injury, sensory level, motor level, spinal cord injury level, injury severity and disease course were collected. ResultsSpinal cord atrophy was identified in 81 cases (54%), among which 78 cases (96%) were American Spinal Injury Association Impairment Scale (AIS) grade A and 3 cases (4%) were AIS grade C. The upper boundary of the spinal cord atrophy site strongly correlated with the injury level, motor level and sensory level (r > 0.8, P < 0.001). ConclusionMore than half of children with SCI may develop secondary spinal cord atrophy, the vast majority of whom suffer from complete spinal cord injury; the upper boundary of spinal cord atrophy is correlated with the injury level.

ObjectiveCleft palate (CP) is a common congenital deformity often associated with velopharyngeal insufficiency (VPI), which disrupts the physiological coupling between respiration and speech. Conventional clinical assessments, such as nasometry and spirometry, provide limited static data and fail to visualize the dynamic spatiotemporal distribution of lung ventilation during phonation. This study introduces spatiotemporal electrical impedance tomography (ST-EIT) to evaluate speech-respiratory functional features in CP patients compared to normal controls (NC). The aim is to characterize multi-domain respiratory patterns and to validate an interpretable machine learning framework for providing objective, quantitative evidence for clinical assessment. MethodsSeventy-five participants were enrolled in this study, comprising 37 patients with surgically repaired CP and 38 healthy volunteers matched for age, gender, and body mass index (BMI). All subjects performed standardized sustained phonation tasks while undergoing synchronous monitoring with a 16-electrode EIT system and a pneumotachograph. A comprehensive feature engineering pipeline was developed to extract physiological parameters across 3 complementary domains. (1) Temporal domain: including inspiratory/expiratory phase duration (tPhase), time constants (Tau), and inspiratory-to-expiratory time ratios (TI/TE); (2) airflow domain: comprising mean flow, peak flow, and instantaneous flow at 25%, 50%, and 75% of tidal volume; and (3) spatial domain: quantifying global and regional tidal impedance variation (TIV), global inhomogeneity (GI), and center of ventilation (CoV). Extreme Gradient Boosting (XGBoost) classifiers were trained using 5 distinct data sources (Spirometry, Nasometry, Inspiratory-EIT, Expiratory-EIT, and fused ST-EIT). Model performance was rigorously evaluated via stratified 5-fold cross-validation, and Shapley additive explanations (SHAP) were employed to quantify global and local feature contributions. ResultsThe CP group exhibited a distinct respiratory phenotype compared to controls. In the temporal domain, CP patients showed significantly shorter inspiratory (1.60 s vs.1.85 s, P<0.001) and expiratory phase durations (2.45 s vs. 3.95 s, P<0.001), indicating a rapid, shallow breathing rhythm. In the airflow domain, while inspiratory flows were comparable, the CP group demonstrated significantly elevated mean and peak flows during the expiratory phase (P<0.001), reflecting compensatory respiratory effort. Spatially, CP patients presented significant ventilation redistribution, characterized by higher regional TIV in the right-anterior (ROI1) and left-posterior (ROI4) quadrants, but lower TIV in the left-anterior (ROI2) quadrant. In terms of diagnostic accuracy, the multi-modal ST-EIT model achieved the highest performance (AUC: 0.915±0.012, Accuracy: 0.843±0.019, F1-score: 0.872±0.017), substantially outperforming models based on spirometry (AUC: 0.721) or nasometry (AUC: 0.625) alone. Interpretability analysis revealed that spatial domain features were the most critical, contributing 53.4% to the model’s decision-making, followed by temporal (25.0%) and airflow (21.6%) features. ConclusionST-EIT successfully captures the temporal, airflow, and spatial deviations in CP speech respiration that are undetectable by conventional methods—specifically, rapid phase transitions, hyperdynamic expiratory airflow, and regional ventilation heterogeneity. This study validates ST-EIT as a robust, non-invasive, and radiation-free tool for characterizing speech-respiratory dysfunction, offering high clinical value for bedside screening, rehabilitation planning, and longitudinal monitoring of patients with cleft palate.

ObjectiveCleft palate (CP) is a common congenital deformity often associated with velopharyngeal insufficiency (VPI), which disrupts the physiological coupling between respiration and speech. Conventional clinical assessments, such as nasometry and spirometry, provide limited static data and fail to visualize the dynamic spatiotemporal distribution of lung ventilation during phonation. This study introduces spatiotemporal electrical impedance tomography (ST-EIT) to evaluate speech-respiratory functional features in CP patients compared to normal controls (NC). The aim is to characterize multi-domain respiratory patterns and to validate an interpretable machine learning framework for providing objective, quantitative evidence for clinical assessment. MethodsSeventy-five participants were enrolled in this study, comprising 37 patients with surgically repaired CP and 38 healthy volunteers matched for age, gender, and body mass index (BMI). All subjects performed standardized sustained phonation tasks while undergoing synchronous monitoring with a 16-electrode EIT system and a pneumotachograph. A comprehensive feature engineering pipeline was developed to extract physiological parameters across 3 complementary domains. (1) Temporal domain: including inspiratory/expiratory phase duration (tPhase), time constants (Tau), and inspiratory-to-expiratory time ratios (TI/TE); (2) airflow domain: comprising mean flow, peak flow, and instantaneous flow at 25%, 50%, and 75% of tidal volume; and (3) spatial domain: quantifying global and regional tidal impedance variation (TIV), global inhomogeneity (GI), and center of ventilation (CoV). Extreme Gradient Boosting (XGBoost) classifiers were trained using 5 distinct data sources (Spirometry, Nasometry, Inspiratory-EIT, Expiratory-EIT, and fused ST-EIT). Model performance was rigorously evaluated via stratified 5-fold cross-validation, and Shapley additive explanations (SHAP) were employed to quantify global and local feature contributions. ResultsThe CP group exhibited a distinct respiratory phenotype compared to controls. In the temporal domain, CP patients showed significantly shorter inspiratory (1.60 s vs.1.85 s, P<0.001) and expiratory phase durations (2.45 s vs. 3.95 s, P<0.001), indicating a rapid, shallow breathing rhythm. In the airflow domain, while inspiratory flows were comparable, the CP group demonstrated significantly elevated mean and peak flows during the expiratory phase (P<0.001), reflecting compensatory respiratory effort. Spatially, CP patients presented significant ventilation redistribution, characterized by higher regional TIV in the right-anterior (ROI1) and left-posterior (ROI4) quadrants, but lower TIV in the left-anterior (ROI2) quadrant. In terms of diagnostic accuracy, the multi-modal ST-EIT model achieved the highest performance (AUC: 0.915±0.012, Accuracy: 0.843±0.019, F1-score: 0.872±0.017), substantially outperforming models based on spirometry (AUC: 0.721) or nasometry (AUC: 0.625) alone. Interpretability analysis revealed that spatial domain features were the most critical, contributing 53.4% to the model’s decision-making, followed by temporal (25.0%) and airflow (21.6%) features. ConclusionST-EIT successfully captures the temporal, airflow, and spatial deviations in CP speech respiration that are undetectable by conventional methods—specifically, rapid phase transitions, hyperdynamic expiratory airflow, and regional ventilation heterogeneity. This study validates ST-EIT as a robust, non-invasive, and radiation-free tool for characterizing speech-respiratory dysfunction, offering high clinical value for bedside screening, rehabilitation planning, and longitudinal monitoring of patients with cleft palate.