ObjectiveTo explore the mechanism of Yinchenhao Tang regulating the tumor necrosis factor receptor superfamily member 6 （Fas）/cysteine protease-8 （Caspase-8）/cysteine protease-3 （Caspase-3） signaling pathway to inhibit hepatocyte apoptosis and improve cholestatic liver injury （CLI）. MethodsAmong 48 Wistar rats，12 rats were randomly selected as the blank group，and the other rats were administered alpha-naphthalene isothiocyanate （ANIT） by gavage to induce a CLI model. The modeling rats were randomly divided into the model group， the ursodeoxycholic acid group（0.1 g·kg^-1） and the Yinchenhao Tang group（9.23 g·kg^-1），with 12 rats in each group. The rats in each group were given corresponding drugs by gavage for three consecutive days. The levels of alanine aminotransferase （ALT），aspartate aminotransferase （AST），alkaline phosphatase （ALP），gamma-glutamyl transpeptidase （γ-GT），total bilirubin （TBil） and total bile acid （TBA） in serum were detected. The levels of tumor necrosis factor-α （TNF-α） and interleukin-1β （IL-1β） in liver tissue were detected. The histopathological changes of the liver were observed by hematoxylin-eosin （HE） staining. The protein and mRNA expressions of Fas，Caspase-8，Caspase-3，B-cell lymphoma-2 （Bcl-2） associated X protein （Bax） and Bcl-2 in liver tissue were detected by Western blot and real-time fluorescence quantitative polymerase chain reaction （Real-time PCR）. ResultsCompared with those in the blank group，the levels of ALT，AST，ALP，γ-GT，TBA and TBil in serum of the model group were significantly increased （P<0.01）. The levels and mRNA expressions of TNF-α and IL-1β in liver tissue were significantly increased （P<0.01）. The arrangement of hepatocytes was disordered，and inflammatory cell infiltration and bile duct epithelial cell proliferation were observed. The protein and mRNA expressions of Fas，Caspase-8，Caspase-3 and Bax in liver tissue were significantly increased（P<0.05，P<0.01），while the protein and mRNA expressions of Bcl-2 were significantly decreased （P<0.05，P<0.01）. Compared with those in the model group，the levels of ALP，γ-GT，TBA and TBil in the serum of rats in the ursodeoxycholic acid group were significantly decreased. The levels and mRNA expressions of TNF-α and IL-1β in liver tissue were significantly decreased（P<0.05，P<0.01）. The protein and mRNA expressions of Fas，Caspase-8，Caspase-3 and Bax in liver tissue were significantly decreased （P<0.05，P<0.01），while the mRNA expression of Bcl-2 was significantly increased （P<0.05，）. The levels of ALT，AST，γ-GT，TBA and TBil in the serum of rats in the Yinchenhao Tang group were significantly decreased （P<0.01）. The levels and mRNA expressions of TNF-α and IL-1β in liver tissue were significantly decreased （P<0.05，P<0.01）. The protein expression of Fas and Bax and the mRNA expression of Fas，Caspase-8，Caspase-3 and Bax in liver tissue were significantly decreased （P<0.05，P<0.01），while the protein and mRNA expression of Bcl-2 were significantly increased （P<0.05，P<0.01）. Hepatocyte injury，inflammatory cell infiltration and proliferation of bile duct epithelial cells were reduced. ConclusionYinchenhao Tang can ameliorate CLI，and its mechanism may be related to inhibiting hepatocyte apoptosis mediated by the Fas/Caspase-8/Caspase-3 signaling pathway.

ObjectiveTo evaluate the therapeutic effects of Huanglian Jiedutang on cerebral infarction injury in a mouse model of middle cerebral artery occlusion （MCAO） and to explore its mechanism of action on oligodendrocytes， particularly its potential in myelin repair. MethodsMultiple experimental approaches were used to evaluate cerebral ischemic injury and the effects of drug intervention. Laser speckle imaging was used to detect changes in cerebral blood flow， 2，3，5-Triphenyltetrazolium chloride （TTC） staining was used to measure infarct volume， and neurological function was scored according to the Zea-Longa criteria. Brain tissues were routinely embedded in paraffin and subjected to HE and Nissl staining to observe tissue structure and neuronal damage. Animals were divided into a sham group （n=24）， model group （n=24）， Huanglian Jiedutang group （n=24）， and Ginkgo biloba extract （GBE） group （n=18）. After 1 week of acclimatization， intragastric administration was initiated. The sham and model groups received normal saline， the Huanglian Jiedutang group was administered 1.82 g·kg^-1， and the GBE group was administered 0.432 g·kg^-1 after preparation as a 2.16 g·L^-1 solution. All groups were treated for 5 consecutive days at a dose of 0.2 mL·（10 g）^-¹·d^-¹. The MCAO model was established after the final administration on day 6. Single-cell RNA sequencing was used to analyze brain tissue cellular composition and changes in oligodendrocyte subpopulations. Distinct subpopulations were identified by Uniform manifold approximation and projection （UMAP） dimensionality reduction and unsupervised clustering， and marker gene expression was analyzed. Pathway enrichment and causal inference were further performed using IPA. Finally， real-time quantitative PCR was used to verify mRNA expression changes of myelin-related genes. ResultsCompared with the sham group， the model group showed significantly increased neurological function scores （P<0.01）， significantly impaired blood flow （P<0.01）， significantly enlarged cerebral infarct area （P<0.01）， and pathological changes including disordered cortical structural arrangement， aggravated cytoplasmic vacuolization， and increased Nissl bodies. Compared with the model group， the Huanglian Jiedutang and GBE groups showed significantly decreased neurological function scores （P<0.01）， markedly restored blood flow levels （P<0.01）， significantly reduced cerebral infarct area （P<0.01）， and improvement in cortical structural disorder， alleviation of cytoplasmic vacuolization， and a reduction in Nissl bodies. Single-cell data showed that a myelin-associated oligodendrocyte （Mye-OL） subpopulation existed among oligodendrocytes， which was closely related to myelin generation. Compared with the sham group， the number of Mye-OL cells decreased in the model group. Compared with the model group， the number of Mye-OL cells increased in the Huanglian Jiedutang group. This subpopulation promoted the expression of myelin-related genes， including MOG， MBP， and MAG， via transcription factors such as OLIG1， OLIG2， NKX2-2， and SOX10， thereby regulating myelin generation， restoring cognition， and exerting therapeutic effects on acute cerebral infarction. Compared with the sham group， the mRNA expression levels of OLIG1， OLIG2， NKX2-2， and SOX10 were significantly downregulated in the model group （P<0.01）， and the mRNA expression levels of myelin-related genes， including MOG， MBP， and MAG， were also significantly downregulated （P<0.01）. In contrast， compared with the model group， the Huanglian Jiedutang and GBE groups showed significantly upregulated mRNA expression levels of OLIG1， OLIG2， NKX2-2， and SOX10 （P<0.01）， and significantly upregulated mRNA expression levels of myelin-related genes， including MOG， MBP， and MAG （P<0.01）. ConclusionHuanglian Jiedutang exerts therapeutic effects on acute cerebral infarction by regulating the OLIG1/2-NKX2-2-SOX10 signaling pathway to promote myelin generation by Mye-OL cells.

ObjectiveTo investigate the characteristics of metabolic reprogramming during cerebral ischemia-reperfusion injury using single-cell transcriptome sequencing， analyze the heterogeneity of microglial populations， and evaluate the interventional effects of Huanglian Jiedutang on metabolic abnormalities and neuroinflammation. MethodsA transient middle cerebral artery occlusion （tMCAO） model was used to establish ischemic stroke in mice. Local cerebral blood flow changes were monitored by laser speckle imaging. Neurological impairment was evaluated using the Zea-Longa score， and histopathological damage in brain tissue was observed by HE and Nissl staining. Animals were divided into a sham group， model group， Huanglian Jiedutang group， and Ginkgo biloba extract （GBE） group. After 1 week of acclimatization， intragastric administration was initiated. The sham and model groups received normal saline， the Huanglian Jiedutang group was administered 1.82 g·kg^-1， and the GBE group was administered 0.432 g·kg^-1 after preparation as a 2.16 mg/mL solution. All groups were treated for 5 consecutive days （0.2 mL/10 g/day）， and the tMCAO model was established on day 6 after the final administration. At the molecular level， single-cell RNA sequencing was performed on ischemic hemisphere tissue. Non-negative matrix factorization （NMF） was used to cluster microglial subpopulations， combined with differential expression analysis， metabolic reprogramming assessment， and inflammatory factor correlation analysis to elucidate their functional characteristics in ischemia-reperfusion injury. Transcription factor enrichment analysis was further conducted to identify key regulatory nodes. Finally， PCR was used to detect mRNA expression changes of relevant genes to validate the single-cell sequencing results. ResultsCompared with the sham group， the model group showed increased neurological function scores （P<0.01）， decreased blood flow levels （P<0.01）， disordered cortical structure， increased cytoplasmic vacuolization， and increased Nissl bodies. Compared with the model group， the Huanglian Jiedutang and GBE groups showed decreased neurological function scores （P<0.01）， increased blood flow levels （P<0.01）， alleviated cortical structural disorder， reduced cytoplasmic vacuolization， and decreased Nissl bodies. Single-cell analysis showed that microglia could be divided into five subpopulations. Among them， clusters 3 and 5 exhibited significant pro-inflammatory phenotypes， with marked activation of hypoxia and NF-κB signaling pathways， and were identified as pro-inflammatory subpopulations. Clusters 1 and 2 were enriched in Wnt/β-catenin and transforming growth factor（TGF）-β signaling pathways and exhibited prominent anti-inflammatory and reparative characteristics. Meanwhile， glycolysis-related genes， such as HK2， PFKP， and LDHA， were significantly upregulated in the pro-inflammatory subpopulations. Correlation analysis showed that the expression levels of inflammatory molecules were positively correlated with glycolysis-related gene expression levels， whereas the expression levels of reparative and anti-inflammatory molecules were negatively correlated with glycolysis-related gene expression levels， indicating that microglia rely on the glycolytic pathway for energy acquisition under ischemic conditions. Further single-cell transcriptome analysis revealed that Huanglian Jiedutang effectively downregulated key genes driving metabolic reprogramming （such as HK2， PFKP， and LDHA）， significantly reduced the proportion of microglial subpopulations accompanied by glycolytic reprogramming， and inhibited their transformation toward a damage phenotype， thereby reducing inflammatory injury. Meanwhile， compared with the sham group， the mRNA expression levels of interleukin （IL）-1β， IL-6， tumor necrosis factor（TNF）-α， CCL2， CXCL2， and CSF3 were significantly upregulated （P<0.01） in the model group， whereas the mRNA expression levels of endothelial- and pericyte-related functional genes， including RGS5， PECAM1， VEGFB， and NOS3， were significantly downregulated （P<0.01）. In contrast， compared with the model group， the Huanglian Jiedutang and GBE groups showed significantly decreased mRNA expression levels of IL-1β， IL-6， TNF-α， CCL2， CXCL2， and CSF3 （P<0.01）， and significantly increased mRNA expression levels of endothelial- and pericyte-related functional genes， including RGS5， PECAM1， VEGFB， and NOS3 （P<0.01）. ConclusionHuanglian Jiedutang exerts neuroprotective effects by regulating the metabolic reprogramming state of microglia and modulating their inflammatory levels， thereby inhibiting neuroinflammatory injury.

ObjectiveTo evaluate the therapeutic effects of Huanglian Jiedutang on cerebral infarction injury in a mouse model of middle cerebral artery occlusion （MCAO） and to explore its mechanism of action on oligodendrocytes， particularly its potential in myelin repair. MethodsMultiple experimental approaches were used to evaluate cerebral ischemic injury and the effects of drug intervention. Laser speckle imaging was used to detect changes in cerebral blood flow， 2，3，5-Triphenyltetrazolium chloride （TTC） staining was used to measure infarct volume， and neurological function was scored according to the Zea-Longa criteria. Brain tissues were routinely embedded in paraffin and subjected to HE and Nissl staining to observe tissue structure and neuronal damage. Animals were divided into a sham group （n=24）， model group （n=24）， Huanglian Jiedutang group （n=24）， and Ginkgo biloba extract （GBE） group （n=18）. After 1 week of acclimatization， intragastric administration was initiated. The sham and model groups received normal saline， the Huanglian Jiedutang group was administered 1.82 g·kg^-1， and the GBE group was administered 0.432 g·kg^-1 after preparation as a 2.16 g·L^-1 solution. All groups were treated for 5 consecutive days at a dose of 0.2 mL·（10 g）^-¹·d^-¹. The MCAO model was established after the final administration on day 6. Single-cell RNA sequencing was used to analyze brain tissue cellular composition and changes in oligodendrocyte subpopulations. Distinct subpopulations were identified by Uniform manifold approximation and projection （UMAP） dimensionality reduction and unsupervised clustering， and marker gene expression was analyzed. Pathway enrichment and causal inference were further performed using IPA. Finally， real-time quantitative PCR was used to verify mRNA expression changes of myelin-related genes. ResultsCompared with the sham group， the model group showed significantly increased neurological function scores （P<0.01）， significantly impaired blood flow （P<0.01）， significantly enlarged cerebral infarct area （P<0.01）， and pathological changes including disordered cortical structural arrangement， aggravated cytoplasmic vacuolization， and increased Nissl bodies. Compared with the model group， the Huanglian Jiedutang and GBE groups showed significantly decreased neurological function scores （P<0.01）， markedly restored blood flow levels （P<0.01）， significantly reduced cerebral infarct area （P<0.01）， and improvement in cortical structural disorder， alleviation of cytoplasmic vacuolization， and a reduction in Nissl bodies. Single-cell data showed that a myelin-associated oligodendrocyte （Mye-OL） subpopulation existed among oligodendrocytes， which was closely related to myelin generation. Compared with the sham group， the number of Mye-OL cells decreased in the model group. Compared with the model group， the number of Mye-OL cells increased in the Huanglian Jiedutang group. This subpopulation promoted the expression of myelin-related genes， including MOG， MBP， and MAG， via transcription factors such as OLIG1， OLIG2， NKX2-2， and SOX10， thereby regulating myelin generation， restoring cognition， and exerting therapeutic effects on acute cerebral infarction. Compared with the sham group， the mRNA expression levels of OLIG1， OLIG2， NKX2-2， and SOX10 were significantly downregulated in the model group （P<0.01）， and the mRNA expression levels of myelin-related genes， including MOG， MBP， and MAG， were also significantly downregulated （P<0.01）. In contrast， compared with the model group， the Huanglian Jiedutang and GBE groups showed significantly upregulated mRNA expression levels of OLIG1， OLIG2， NKX2-2， and SOX10 （P<0.01）， and significantly upregulated mRNA expression levels of myelin-related genes， including MOG， MBP， and MAG （P<0.01）. ConclusionHuanglian Jiedutang exerts therapeutic effects on acute cerebral infarction by regulating the OLIG1/2-NKX2-2-SOX10 signaling pathway to promote myelin generation by Mye-OL cells.

ObjectiveTo investigate the characteristics of metabolic reprogramming during cerebral ischemia-reperfusion injury using single-cell transcriptome sequencing， analyze the heterogeneity of microglial populations， and evaluate the interventional effects of Huanglian Jiedutang on metabolic abnormalities and neuroinflammation. MethodsA transient middle cerebral artery occlusion （tMCAO） model was used to establish ischemic stroke in mice. Local cerebral blood flow changes were monitored by laser speckle imaging. Neurological impairment was evaluated using the Zea-Longa score， and histopathological damage in brain tissue was observed by HE and Nissl staining. Animals were divided into a sham group， model group， Huanglian Jiedutang group， and Ginkgo biloba extract （GBE） group. After 1 week of acclimatization， intragastric administration was initiated. The sham and model groups received normal saline， the Huanglian Jiedutang group was administered 1.82 g·kg^-1， and the GBE group was administered 0.432 g·kg^-1 after preparation as a 2.16 mg/mL solution. All groups were treated for 5 consecutive days （0.2 mL/10 g/day）， and the tMCAO model was established on day 6 after the final administration. At the molecular level， single-cell RNA sequencing was performed on ischemic hemisphere tissue. Non-negative matrix factorization （NMF） was used to cluster microglial subpopulations， combined with differential expression analysis， metabolic reprogramming assessment， and inflammatory factor correlation analysis to elucidate their functional characteristics in ischemia-reperfusion injury. Transcription factor enrichment analysis was further conducted to identify key regulatory nodes. Finally， PCR was used to detect mRNA expression changes of relevant genes to validate the single-cell sequencing results. ResultsCompared with the sham group， the model group showed increased neurological function scores （P<0.01）， decreased blood flow levels （P<0.01）， disordered cortical structure， increased cytoplasmic vacuolization， and increased Nissl bodies. Compared with the model group， the Huanglian Jiedutang and GBE groups showed decreased neurological function scores （P<0.01）， increased blood flow levels （P<0.01）， alleviated cortical structural disorder， reduced cytoplasmic vacuolization， and decreased Nissl bodies. Single-cell analysis showed that microglia could be divided into five subpopulations. Among them， clusters 3 and 5 exhibited significant pro-inflammatory phenotypes， with marked activation of hypoxia and NF-κB signaling pathways， and were identified as pro-inflammatory subpopulations. Clusters 1 and 2 were enriched in Wnt/β-catenin and transforming growth factor（TGF）-β signaling pathways and exhibited prominent anti-inflammatory and reparative characteristics. Meanwhile， glycolysis-related genes， such as HK2， PFKP， and LDHA， were significantly upregulated in the pro-inflammatory subpopulations. Correlation analysis showed that the expression levels of inflammatory molecules were positively correlated with glycolysis-related gene expression levels， whereas the expression levels of reparative and anti-inflammatory molecules were negatively correlated with glycolysis-related gene expression levels， indicating that microglia rely on the glycolytic pathway for energy acquisition under ischemic conditions. Further single-cell transcriptome analysis revealed that Huanglian Jiedutang effectively downregulated key genes driving metabolic reprogramming （such as HK2， PFKP， and LDHA）， significantly reduced the proportion of microglial subpopulations accompanied by glycolytic reprogramming， and inhibited their transformation toward a damage phenotype， thereby reducing inflammatory injury. Meanwhile， compared with the sham group， the mRNA expression levels of interleukin （IL）-1β， IL-6， tumor necrosis factor（TNF）-α， CCL2， CXCL2， and CSF3 were significantly upregulated （P<0.01） in the model group， whereas the mRNA expression levels of endothelial- and pericyte-related functional genes， including RGS5， PECAM1， VEGFB， and NOS3， were significantly downregulated （P<0.01）. In contrast， compared with the model group， the Huanglian Jiedutang and GBE groups showed significantly decreased mRNA expression levels of IL-1β， IL-6， TNF-α， CCL2， CXCL2， and CSF3 （P<0.01）， and significantly increased mRNA expression levels of endothelial- and pericyte-related functional genes， including RGS5， PECAM1， VEGFB， and NOS3 （P<0.01）. ConclusionHuanglian Jiedutang exerts neuroprotective effects by regulating the metabolic reprogramming state of microglia and modulating their inflammatory levels， thereby inhibiting neuroinflammatory injury.

ObjectiveTo construct a large language model （LLM）-based intelligent pre-consultation system for traditional Chinese medicine （TCM） to improve efficacy of clinical practice. MethodsA TCM large language model was fine-tuned using DeepSpeed ZeRO-3 distributed training strategy based on YAYI 2-30B. A weighted undirected graph network was designed and an agent-based syndrome differentiation model was established based on relationship data extracted from TCM literature and clinical records. An agent collaboration framework was developed to integrate the TCM LLM with the syndrome differentiation model. Model performance was comprehensively evaluated by Loss function， BLEU-4， and ROUGE-L metrics， through which training convergence， text generation quality， and language understanding capability were assessed. Professional knowledge test sets were developed to evaluate system proficiency in TCM physician licensure content， TCM pharmacist licensure content， TCM symptom terminology recognition， and meridian identification. Clinical tests were conducted to compare the system with attending physicians in terms of diagnostic accuracy， consultation rounds， and consultation duration. ResultsAfter 100 000 iterations， the training loss value was gradually stabilized at about 0.7±0.08， indicating that the TCM-LLM has been trained and has good generalization ability. The TCM-LLM scored 0.38 in BLEU-4 and 0.62 in ROUGE-L， suggesting that its natural language processing ability meets the standard. We obtained 2715 symptom terms， 505 relationships between diseases and syndromes， 1011 relationships between diseases and main symptoms， and 1 303 600 relationships among different symptoms， and constructed the Agent of syndrome differentiation model. The accuracy rates in the simulated tests for TCM practitioners， licensed pharmacists of Chinese materia medica， recognition of TCM symptom terminology， and meridian recognition were 94.09%， 78.00%， 87.50%， and 68.80%， respectively. In clinical tests， the syndrome differentiation accuracy of the system reached 88.33%， with fewer consultation rounds and shorter consultation time compared to the attending physicians （P＜0.01）， suggesting that the system has a certain pre- consultation ability. ConclusionThe LLM-based intelligent TCM pre-diagnosis system could simulate diagnostic thinking of TCM physicians to a certain extent. After understanding the patients' natural language， it collects all the patient's symptom through guided questioning， thereby enhancing the diagnostic and treatment efficiency of physicians as well as the consultation experience of the patients.

Objective To evaluate the performance of magnetic resonance imaging(MRI)multi-sequence feature imputation and fusion mutual model based on sequence deletion in differentiating high-grade glioma(HGG)from low-grade glioma(LGG).Methods We retrospectively collected multi-sequence MR images from 305 glioma patients,including 189 HGG patients and 116 LGG patients.The region of interest(ROI)of T1-weighted images(T1WI),T2-weighted images(T2WI),T2 fluid attenuated inversion recovery(T2_FLAIR)and post-contrast enhancement T1WI(CE_T1WI)were delineated to extract the radiomics features.A mutual-aid model of MRI multi-sequence feature imputation and fusion based on sequence deletion was used for imputation and fusion of the feature matrix with missing data.The discriminative ability of the model was evaluated using 5-fold cross-validation method and by assessing the accuracy,balanced accuracy,area under the ROC curve(AUC),specificity,and sensitivity.The proposed model was quantitatively compared with other non-holonomic multimodal classification models for discriminating HGG and LGG.Class separability experiments were performed on the latent features learned by the proposed feature imputation and fusion methods to observe the classification effect of the samples in two-dimensional plane.Convergence experiments were used to verify the feasibility of the model.Results For differentiation of HGG from LGG with a missing rate of 10%,the proposed model achieved accuracy,balanced accuracy,AUC,specificity,and sensitivity of 0.777,0.768,0.826,0.754 and 0.780,respectively.The fused latent features showed excellent performance in the class separability experiment,and the algorithm could be iterated to convergence with superior classification performance over other methods at the missing rates of 30%and 50%.Conclusion The proposed model has excellent performance in classification task of HGG and LGG and outperforms other non-holonomic multimodal classification models,demonstrating its potential for efficient processing of non-holonomic multimodal data.

Objective To evaluate the performance of magnetic resonance imaging(MRI)multi-sequence feature imputation and fusion mutual model based on sequence deletion in differentiating high-grade glioma(HGG)from low-grade glioma(LGG).Methods We retrospectively collected multi-sequence MR images from 305 glioma patients,including 189 HGG patients and 116 LGG patients.The region of interest(ROI)of T1-weighted images(T1WI),T2-weighted images(T2WI),T2 fluid attenuated inversion recovery(T2_FLAIR)and post-contrast enhancement T1WI(CE_T1WI)were delineated to extract the radiomics features.A mutual-aid model of MRI multi-sequence feature imputation and fusion based on sequence deletion was used for imputation and fusion of the feature matrix with missing data.The discriminative ability of the model was evaluated using 5-fold cross-validation method and by assessing the accuracy,balanced accuracy,area under the ROC curve(AUC),specificity,and sensitivity.The proposed model was quantitatively compared with other non-holonomic multimodal classification models for discriminating HGG and LGG.Class separability experiments were performed on the latent features learned by the proposed feature imputation and fusion methods to observe the classification effect of the samples in two-dimensional plane.Convergence experiments were used to verify the feasibility of the model.Results For differentiation of HGG from LGG with a missing rate of 10%,the proposed model achieved accuracy,balanced accuracy,AUC,specificity,and sensitivity of 0.777,0.768,0.826,0.754 and 0.780,respectively.The fused latent features showed excellent performance in the class separability experiment,and the algorithm could be iterated to convergence with superior classification performance over other methods at the missing rates of 30%and 50%.Conclusion The proposed model has excellent performance in classification task of HGG and LGG and outperforms other non-holonomic multimodal classification models,demonstrating its potential for efficient processing of non-holonomic multimodal data.

OBJECTIVES: To evaluate the performance of different multi-modality fusion models for predicting radiation-induced oral mucositis (RIOM) following radiotherapy in patients with nasopharyngeal carcinoma (NPC). METHODS: We retrospectively collected the data from 198 patients with locally advanced NPC who experienced RIOM following radiotherapy at the Affiliated Tumor Hospital of Guangzhou Medical University from September, 2022 to February, 2023. Based on oral radiation dose-volume parameters and clinical features of NPC, basic classification models were developed using different combinations of feature selection algorithms and classifiers and integrated using a multi-criterion decision-making (MCDM)-based classifier fusion (MCF) strategy and its variant, the H-MCF model. The basic classification models, MCF model, the H-MCF model with a single modality or multiple modalities and other ensemble classifiers were compared for performances for predicting RIOM by assessing the area under the ROC curve (AUC), accuracy, sensitivity, and specificity. RESULTS: The H-MCF model, which integrated multi-modality features, achieved the highest accuracy for predicting severe RIOM with an AUC of 0.883, accuracy of 0.850, sensitivity of 0.933, and specificity of 0.800. CONCLUSIONS Compared with each of the individual classifiers, the multimodal multi-classifier fusion algorithm combining clinical and dosimetric modalities demonstrates superior performance in predicting the incidence of severe RIOM in NPC patients following radiotherapy.
Humans ; Nasopharyngeal Carcinoma/radiotherapy* ; Nasopharyngeal Neoplasms/radiotherapy* ; Retrospective Studies ; Stomatitis/diagnosis* ; Algorithms ; Radiation Injuries/diagnosis* ; Female ; Male ; ROC Curve

Objective:To explore the risk factors and outcomes of central nervous system(CNS)complications associated with hematopoietic stem cell transplantation(HSCT).Methods:A total of 550 recipient after HSCT in the department of hematology of Tongji Hospital affiliated to Tongji Medical College of Huazhong University of Science and Technology from January 1 2019 to August 31 2021were enrolled.According to the occurrence of CNS complications, they were divided into the CNS group(24 cases)and the non CNS group(526 cases). The clinical information and prognosis were compared.We further analyzed the risk factors associated with CNS complications, and conducted multivariate logistic regression on statistically significant indicators.Cox regression analysis is conducted on prognostic factors such as age, gender and risk degree.Results:A total of 550 recipients were enrolled, of which 330 underwent allo-HSCT, and others received auto-HSCT.A total of 24 cases (4.36%)had CNS complications, of which 4 cases had 2 types of CNS complications.The type of CNS complications included intracranial infection(8 cases, 28.57%), transplantation-associated thrombotic microangiopathy(TA-TMA)(6 cases, 21.43%), central tumor invasion(4 cases, 14.29%), intracranial hemorrhage(4 cases, 14.29%), leucodystrophy(2 cases, 7.14%)and unexplained encephalopathy(4 cases, 14.29%). Logistic regression analysis of risk factors related to CNS complications showed that, Platelet implantation time( β=0.084, OR=1.088, P=0.048), CMV infection( β=1.295, OR=3.65, P=0.008)is positively correlated with the occurrence of CNS complications in HSCT recipients but age( β=-0.052, OR=0.949, P=0.004)is negatively correlated with it.Nine of the 24 cases(37.50%)who experienced CNS complications died, including 3 cases of intracranial infection, 3 cases of cerebral hemorrhage, 2 cases of TMA, and 1 case of unexplained encephalopathy.Platelet implantation time is an independent risk factor for poor prognosis of CNS complications in HSCT recipients. Conclusions:Our results indicated that, age, CMV infection and platelet implantation time were associated with the occurrence of CNS complications after HSCT.Platelet implantation time is an independent risk factor for poor prognosis of CNS complications in HSCT recipients.