Objective To explore the diagnostic efficacy of machine learning in predicting perineural invasion(PNI)of invasive breast cancer based on MRI imaging features of breast cancer.Methods The data of 294 patients with invasive breast cancer confirmed by surgical pathology were retrospectively analyzed,and the patients were randomly divided into training set(205 cases,PNI 77 cases)and validation set(89 cases,PNI 33 cases)at a ratio of 7∶3.10 machine learning models were constructed by selecting training set clinical and radiographic features using single factor logistic regression.The area under the curve(AUC),accuracy(ACC),sensitivity(SE),specificity(SP),positive predictive value(PPV),and negative predictive value(NPV)were used to evaluate the predictive effi-cacy of different models for PNI,and the best model was determined.SHapley Additive exPlanation(SHAP)was used to visuaize the diagnosis process of the model.Results In the validation set,the multi-layer perceptron(MLP)model performed best,with AUC,ACC,SE,SP,PPV,and NPV of 0.91,0.89,0.79,0.95,0.90,and 0.88,respectively.Conclusion The model of MRI imaging fea-tures of breast cancer constructed by MLP machine learning model can effectively predict the preoperative PNI of invasive breast cancer.

Objective:To verify the efficacy of VELscope autofluorescence examination technology in the early warning of potential oral malignant diseases.Methods:80 patients with suspected dysplasia were included and underwent oral examination,visually en-hanced lesion scope(VELscope)autofluorescence examination and histopathology examination respectively,the fluorescence imaging results were compared with the histopathological results.Results:Pathological reports showed 64 cases with mild or no dysplasia,9 cases with moderate or severe dysplasia,1 case with moderate dysplasia without malignant transformation and 6 cases with cancer.The results of VELscope showed that there were 73 positive cases and negative 7 cases of fluorescence deletion.The sensitivity and specificity of VELscope were 93.75％and 9.37％respectively,and the sensitivity could reach 100％in the cancerous tissue.Con-clusion:As a non-invasive examination method,VELscope is highly sensitive but lowly specific.

Objective To explore the diagnostic efficacy of machine learning in predicting perineural invasion(PNI)of invasive breast cancer based on MRI imaging features of breast cancer.Methods The data of 294 patients with invasive breast cancer confirmed by surgical pathology were retrospectively analyzed,and the patients were randomly divided into training set(205 cases,PNI 77 cases)and validation set(89 cases,PNI 33 cases)at a ratio of 7∶3.10 machine learning models were constructed by selecting training set clinical and radiographic features using single factor logistic regression.The area under the curve(AUC),accuracy(ACC),sensitivity(SE),specificity(SP),positive predictive value(PPV),and negative predictive value(NPV)were used to evaluate the predictive effi-cacy of different models for PNI,and the best model was determined.SHapley Additive exPlanation(SHAP)was used to visuaize the diagnosis process of the model.Results In the validation set,the multi-layer perceptron(MLP)model performed best,with AUC,ACC,SE,SP,PPV,and NPV of 0.91,0.89,0.79,0.95,0.90,and 0.88,respectively.Conclusion The model of MRI imaging fea-tures of breast cancer constructed by MLP machine learning model can effectively predict the preoperative PNI of invasive breast cancer.

Objective:To verify the efficacy of VELscope autofluorescence examination technology in the early warning of potential oral malignant diseases.Methods:80 patients with suspected dysplasia were included and underwent oral examination,visually en-hanced lesion scope(VELscope)autofluorescence examination and histopathology examination respectively,the fluorescence imaging results were compared with the histopathological results.Results:Pathological reports showed 64 cases with mild or no dysplasia,9 cases with moderate or severe dysplasia,1 case with moderate dysplasia without malignant transformation and 6 cases with cancer.The results of VELscope showed that there were 73 positive cases and negative 7 cases of fluorescence deletion.The sensitivity and specificity of VELscope were 93.75％and 9.37％respectively,and the sensitivity could reach 100％in the cancerous tissue.Con-clusion:As a non-invasive examination method,VELscope is highly sensitive but lowly specific.

A 34-year-old female patient with multiple injuries in a car accident was admitted to hospital and received inosine injection.After 11 days,leukemioid reaction mainly characterized by"abnormal increase in white blood cell count"occurred,and the white blood cell count reached up to 38.26×109·L-1.Considering that inosine injection was the cause,the white blood cell count gradually decreased to 15.84×109·L-1 after the drug was stopped.Naranjo's assessment scale was used to evaluate the association of adverse reactions,with a score of 6,and the result was"very likely to be relevant".The cases reported in this paper belong to new adverse reactions caused by inosine injection,suggesting that inosine injection as an adjunct should pay attention to drug safety.If there is an abnormal increase in white blood cell count,it is necessary to be alert to the possibility of"leukemioid reaction"and stop the drug in time if necessary.

Objective To compare the induction effects of direct treatment with low-temperature plasma(LTP)and treatment with plasma-activated medium(PAM)on immunogenic cell death(ICD)of melanoma cells.Methods After direct treatment of melanoma cell line B16F10 with LTP and treatment of it with PAM for 24 hours,cell viability was detected by MTT assay.Flow cytometry was used to detect cell apoptosis and the expression of calreticulin(CRT)on the cell surface.The adenosine triphosphate(ATP)content in the culture medium was detected by an ATP detection kit.The content of high-mobility group box 1(HMGB1)in the cell culture medium was detected by ELISA.B16F10 cells treated with LTP were co-cultured with immature dendritic cells(DC)DC2.4 cell line,and flow cytometry was used to detect DC surface molecules CD80 and CD86.Results Compared with the control group,both direct treatment and indirect treatment could lead to a decrease in the viability of B16F10 cells,an increase in the apoptosis rate,an increase in intracellular ROS,an increase in CRT expression,and an increase in the secretion of ATP and HMGB1(P＜0.05).At the same treatment time,the expression of CRT and the release of ATP in B16F10 cells directly treated with LTP were higher than those indirectly treated with PAM(P＜0.05).Compared with the DC2.4 group,the expression proportion of the DC cell maturation marker molecule CD80 was significantly increased in LTP-120s group,LTP-180s group,PAM-120s group,and PAM-180s group.The expression proportion of the DC cell maturation marker molecule CD86 was significantly increased in LTP-120s group,LTP-180s group,and PAM-180s group,and the difference was statistically significant(P＜0.05).Conclusion Both direct treatment with LTP and indirect treatment with PAM can induce ICD in melanoma cells.The direct treatment with LTP has a better induction effect.

Neuroinflammation refers to a widespread immune response within central nervous system.T lymphocytes,as main effector cells in cellular immunity,regulate activity of astrocytes through cytokines secretion by various subgroups and participate in pathological process of neuroinflammation.Astrocytes are the most abundant glial cells in central nervous system.Astrocytes activated by T lymphocytes produce pro-inflammatory cytokines and other inflammatory mediators,alter structure of tight junction,contribute to breakdown of blood-brain barrier,and provide conditions for immune cells to migrate to central nervous system.This review is based on integrity of blood-brain barrier and briefly explores crosstalk mechanism between astrocytes and T lymphocytes in neuroinflammatory diseases,including Alzheimer's disease,multiple sclerosis,amyotrophic lateral sclerosis,viral encephalitis and migraine,in order to provide new potential biomarkers and directions for treatment of neuroinflammatory diseases.

Objective:To explore the regulatory mechanism of voltage-dependent anion channel 1(VDAC1) on the proliferation, migration and invasion of lung adenocarcinoma(LUAD) cells.Methods:This study employed a combination of bioinformatics and experimental validation methods, conducting bioinformatics analysis and cytological experimental validation in the central laboratory of the School of Medicine, Anhui University of Science and Technology from February 2023 to August 2024.Clinical histological specimen validation was performed using immunohistochemistry, and a retrospective analysis was conducted on 5 cases of lung adenocarcinoma and adjacent samples from Huai′an First People′s Hospital affiliated with Nanjing Medical University. The TCGA network database was analyzed for the expression pattern, prognostic value, and functional enrichment of VDAC1 in LUAD. A549 cells with VDAC1 knockdown and H1650 cells with VDAC1 overexpression were established through lentiviral transfection. The expression difference of VDAC1 protein in LUAD and adjacent tissue specimens was detected by immunohistochemistry.The effects of VDAC1 on the proliferation, migration, and invasion capabilities were explored through CCK8 assay, scratch healing assay, and Transwell assay.The activation levels of epithelial-mesenchymal transition (EMT) marker proteins, cell cycle-dependent kinases, and molecules in the PI3K/AKT/mTOR signaling pathway were detected by Western blot.Results:Bioinformatics analysis revealed that VDAC1 was highly expressed in LUAD cells ( P<0.000 1) and was an independent risk factor for LUAD ( P<0.000 1). Functional enrichment analysis showed significant enrichment of the PI3K/AKT/mTOR, G2M checkpoint, and P53 signaling pathways ( P<0.001). Compared to adjacent control tissues, the expression level of VDAC1 protein is higher in lung adenocarcinoma tissues.Overexpression of VDAC1 promoted the proliferation ( P<0.000 1), migration, and invasion( P<0.01) of H1650 cells, while knockdown of VDAC1 inhibited the proliferation ( P<0.000 1), migration, and invasion ( P<0.05) of A549 cells.Western Blot experiments showed that compared to the control group, the expression levels of vimentin (1.10±0.11 vs 2.39±0.15, P<0.001), N-cadherin (0.94±0.12 vs 2.72±0.06, P<0.001), CDK1 (0.93±0.04 vs 1.53±0.03, P<0.000 1), CDK2 (1.04±0.13 vs 2.29±0.06, P<0.001), CDK4 (0.90±0.03 vs 2.00±0.11, P<0.01), p-PI3K (1.08±0.13 vs 1.85±0.12, P<0.01), and p-AKT (1.03±0.11 vs 1.69±0.06, P<0.001) were increased in H1650 cells overexpressing VDAC1, while E-cadherin expression decreased (2.18±0.14 vs 0.997±0.11, P<0.001).In contrast, in A549 cells with VDAC1 knockdown, the expression levels of vimentin (1.70±0.26 vs 0.97±0.09, P<0.05), N-cadherin (1.98±0.25 vs 1.03±0.06, P<0.05), CDK1 (1.13±0.03 vs 0.95±0.02, P<0.01), CDK2 (2.29±0.12 vs 0.92±0.10, P<0.001), CDK4 (1.71±0.096 vs 1.12±0.11, P<0.01), p-PI3K (1.67±0.09 vs 0.97±0.03, P<0.001), and p-AKT (1.53±0.04 vs 1.02±0.03, P<0.000 1) decreased, while E-cadherin expression increased (1.04±0.04 vs 1.85±0.26, P<0.05). Conclusions:VDAC1 may promote the proliferation, migration, and invasion of LUAD cells by activating EMT and cyclin-dependent kinases through the PI3K/AKT/mTOR pathway.

Diabetic retinopathy(DR)is a kind of microvascular disease caused by the long-term influence of diabetes mellitus(DM),and it is one of the main cause of global visual impairment and blindness.Its typical characteristics include microaneurysms,hard exudates,macular edema(DME)and neovascularization.Its pathogenesis is diverse,and the root cause is oxidative stress and advanced glycosylation end products.The nuclear red blood cell related factor 2(Nrf2)signa-ling pathway plays an important role in preventing various diseases.As one of the characteristics of DM,hyperglycemia will activate the generation of reactive oxygen species(ROS)in mitochondria.These oxidative stress factors activate the nucle-ar transcription factor κB pathway,becoming the main inducement of various complications of DM.This pathway will in-duce increased transcription of proinflammatory cytokines,including tumor necrosis factor-alpha,transforming growth fac-tor-beta and interleukin-1.The active ingredients of traditional Chinese medicine have significant antioxidant,anti-inflam-matory,anti-apoptotic and angiogenesis-promoting effects,and can block the progression of DR through various mecha-nisms.In this article,the research status of traditional Chinese medicine targeting the Nrf2 signaling pathway in the preven-tion and treatment of DR is reviewed to guide clinical and scientific research.

Neuroinflammation refers to a widespread immune response within central nervous system.T lymphocytes,as main effector cells in cellular immunity,regulate activity of astrocytes through cytokines secretion by various subgroups and participate in pathological process of neuroinflammation.Astrocytes are the most abundant glial cells in central nervous system.Astrocytes activated by T lymphocytes produce pro-inflammatory cytokines and other inflammatory mediators,alter structure of tight junction,contribute to breakdown of blood-brain barrier,and provide conditions for immune cells to migrate to central nervous system.This review is based on integrity of blood-brain barrier and briefly explores crosstalk mechanism between astrocytes and T lymphocytes in neuroinflammatory diseases,including Alzheimer's disease,multiple sclerosis,amyotrophic lateral sclerosis,viral encephalitis and migraine,in order to provide new potential biomarkers and directions for treatment of neuroinflammatory diseases.