Pancreatic cancer is a highly malignant tumor of the digestive tract， and its precision diagnosis and treatment is a major challenge in clinical practice. With the development of science and technology， artificial intelligence has been applied in the early diagnosis， treatment and prognosis prediction of some diseases， which has high clinical value. Medical imaging is a disciplie to evaluate human health， which consists of two relatively independent directions of medical imaging system and medical image processing. Artificial intelligence based on medical imaging plays an important role in the clinical diagnosis and treatment of pancreatic cancer. In this article， research progress in the application of artificial intelligence in precision diagnosis and treatment of pancreatic cancer was reviewed from the aspects of imaging diagnosis， efficacy evaluation， survival prediction and radiogenomics， etc.

The research of artificial intelligence deep learning technology has become a hot topic in the field of medical imaging in recent years. As an important branch of nuclear medicine imaging diagnosis， single-photon emission computer tomography （SPECT） has unique advantages in providing reliable image information for clinicians， especially in functional imaging. In this article， the main research directions， values and application status of deep learning in SPECT in recent years were reviewed， the main problems existing in current research were summarized， and the future research was predicted.

The incidence rate of vertigo-related diseases is high， involving multiple systems of the whole body. It is difficult to make a diagnosis due to complicated pathogenesis， which is a major challenge in daily clinical work. In recent years， rapid development of artificial intelligence technology has gradually become one of the important tools in the clinical diagnosis and treatment of vertigo-related diseases. In this article， the application of artificial intelligence in the diagnosis， evaluation， and treatment of vertigo-related diseases and recent progress and frontiers of artificial intelligence in vertigo-related diseases were reviewed， the advantages and disadvantages of artificial intelligence in the diagnosis and treatment of vertigo-related diseases were identified， and the development prospects and directions of artificial intelligence in the diagnosis and treatment of vertigo-related diseases were predicted.

Objective To evaluate the performance of prediction models for acute penetrating artery territory occlusive cerebral infarction based on machine learning algorithms and select the optimal model， aiming to provide evidence for clinical management of acute penetrating artery territory infarction. Methods A total of 441 patients diagnosed with acute perforator artery territory infarction were enrolled in this study. Patients with incomplete clinical information （n = 10） and multiple cerebral infarctions （n = 28） were excluded， resulting in a final sample size of 403 patients. The outcome variables were divided into two groups： good prognosis （mRS scores of 0-2） and poor prognosis （mRS scores>2）. Univariate and multi-variate Logistic regression （LR） using the stepwise regression method were employed to identify prediction variables. LR， random forest （RF） and support vector machine （SVM） models were utilized to develop a prognostic prediction model. The dataset was further divided randomly into a training set and a test set in a 7：3 ratio. In the test set， the predictive performance of the model for 90-day functional prognosis in patients with BAD （with poor prognosis defined as mRS scores > 2） was evaluated using metric such as the area under the receiver operating characteristic （ROC） curve （AUC）， accuracy， sensitivity and specificity， etc. Results Among 403 patients with BAD， 68.73% of them were male， with an average age of （60.4±11.4） years. Using the stepwise regression method， 7 prediction variables were selected from a pool of 44 variables： white blood cell count， platelet count， blood glucose， cholesterol， history of diabetes mellitus， history of taking hypoglycemic drugs， and history of smoking （all P < 0.05）. The AUC of LR， RF and SVN for predicting clinical prognosis was 0.610， 0.690， and 0.780， respectively. Conclusions Machine learning algorithms have demonstrated certain predictive ability for acute penetrating artery territory infarction. The performance of RF and SVM models （nonlinear models） is superior to traditional logistic regression model （linear model）.

Objective To explore the predictive value of different machine learning models based on MRI radiomics combined with clinical features for histological grade of cervical squamous cell carcinoma. Methods Clinical data of 150 patients with cervical squamous cell carcinoma confirmed by pathological biopsy were retrospectively analyzed. They were randomly divided into the training set and validation set at a ratio of 4∶1. Features were extracted from the regions of interest of T₂WI fat suppression sequence （FS-T₂WI） and enhanced T₁WI （delayed phase）. After dimensionality reduction and feature selection， logistic regression （LR）， support vector machine （SVM）， naïve Bayes （NB）， random forest （RF）， Light Gradient Boosting Machine （LightGBM）， K-nearest neighbor （KNN） were used to construct a radiomics model for predicting the histological grade of cervical squamous cell carcinoma. The area under the receiver operating characteristic （ROC） curve （AUC） was used to evaluate the predictive performance of the six models. Univariate and multivariate logistic regression analyses were performed to predict the independent risk factors， and a combined model of clinical and radiomics was established. The differences of each model were compared by AUC， and the clinical value of the model was evaluated by decision curve （DCA）. Results In the radiomics model， the LightGBM model had the largest AUC （0.910 in the training set， and 0.839 in the validation set）. The AUC of clinical features combined with LightGBM model was the largest （0.935 in the training set， and 0.888 in the validation set）， which was higher than those of clinical model （0.762 in the training set， and 0.710 in the validation set） and LightGBM radiomics model. Conclusions The LightGBM model has a high predictive value in the radiomics model. The combined model has the optimal DCA effect and the highest clinical net benefit. The combined prediction model combining radiomics and clinical features has good predictive value for cervical squamous cell carcinoma with low differentiation， providing a non-invasive and efficient method for clinical decision-making.

Isthmin-1 （ISM1） belongs to the Isthmin gene family. It is expressed in various tissues and has multiple biological functions. ISM1 is involved in growth and development， metabolism， immunity， tumorigenesis and other physiological and pathological processes， which is associated with a variety of lung diseases， including acute respiratory distress syndrome， chronic obstructive pulmonary disease and pulmonary fibrosis， etc. In this article， the structure， biological function of ISM1 and its role in pulmonary diseases were reviewed， aiming to provide a theoretical basis for the diagnosis and treatment of ISM1-related diseases.

Low triiodothyronine syndrome （LT₃S） refers to the abnormal metabolism of thyroid hormones associated with critical illness， but without obvious symptoms of hypothyroidism. LT₃S is closely associated with many critical diseases， such as severe burns， etc. During the occurrence and development of critical illness， many factors can induce LT₃S by affecting the secretion， transfer and metabolism of thyroid hormones. LT₃S is a risk factor for the development of critical illness and the occurrence of complications. It is a good indicator for evaluating the severity and prognosis of critical diseases. In this article， the pathogenesis， relationship with critical illness and treatment progress of LT₃S were reviewed.

Deep sternal wound infection （DSWI） is a rare but seriously damaging complication after median sternotomy. The quality of life and psychological health of patients may be affected due to long management period and unsatisfactory prognosis of DSWI. The improvement of this situation requires multidisciplinary cooperation. Cardiac surgeons should identify risk factors in a timely manner before surgery， take effective preventive measures， and treat incisional infections when detected after surgery to avoid the progression of infection to the sternum and mediastinum. With a definite diagnosis of DSWI， plastic surgeons are required to choose appropriate management protocols to control the infection and achieve sternal reconstruction in a timely manner. At present， prevention of DSWI has not received sufficient attention. DSWI treatment is mostly dependent on doctors’ own experience and understanding， and consensus is lacking. This article provide a review of the progress in the prevention and treatment of DSWI.

Objective To evaluate the prognostic value of early blood inflammatory indicators in patients with respiratory failure caused by severe community-acquired pneumonia. Methods A total of 186 patients diagnosed with respiratory failure due to severe pneumonia on admission were recruited in this study. According to clinical prognosis， they were divided into the death group （n = 42） and improvement group （n = 144）. Twelve cytokines， ferritin， C-reactive protein， procalcitonin， blood routine examination and blood gas analysis results of the patients within two days after admission were analyzed. Pneumonia severity index （PSI） and APACHE Ⅱ scores of the patients were collected. The correlation between the above indicators and prognosis was explored. Results Compared with the improvement group， the levels of IFN-γ， IL-10， IL-6， IL-8， C-reactive protein （CRP）， procalcitonin and ferritin were significantly higher in the death group （all P < 0.05）. Upon admission， the hemoglobin level and platelet count in the death group were significantly lower than those in the improvement group （both P < 0.05）. The PSI and APACHE Ⅱ scores on admission in the death group were significantly higher compared with those in the improvement group （both P < 0.05）. Patients in the death group had higher lactic acid level and lower oxygenation index （OI） compared with their counterparts in the improvement group （both P < 0.05）. Multivariate Logistic regression analysis revealed that IL-6， ferritin， PSI and APACHE Ⅱ scores and lactic acid level were the risk factors for poor prognosis， whereas platelet count was the protective factor for poor prognosis （all P < 0.05）. The receiver operating characteristic （ROC） curve indicated that IL-6，procalcitonin and ferritin were favorable predictors for clinical prognosis of patients. When IL-6 and ferritin levels exceed 11.41 ng/L and 659 μg/L， and platelet count was less than 148.5×10⁹/L， patients were at risk of death. Conclusions In patients with respiratory failure caused by severe community-acquired pneumonia， the increase of blood inflammatory indicators， such as IL-6， ferritin and procalcitonin， suggests high risk death in patients， especially for patients with IL-6 >11.41 ng/L， ferritin>659 μg/L， and platelet count<148.5×10⁹/L. Extensive attention should be paid to lowering the risk of death in these patients.

Objective To evaluate clinical efficacy of the combination of Vonoprazan and high-dose Amoxicillin in the initial eradication treatment of Helicobacter pylori （Hp）. Methods Select 200 patients with Hp infection who have not undergone Hp eradication treatment in the past as the research subjects. All patients were randomly divided into the observation group （n = 100） and control group （n = 100） using random number table method. In the observation group， a combination regimen of Vonoprazan （20 mg twice a day） and Amoxicillin （1 g three times a day） was given for 14 days. In the control group， a bismuth quadruple therapy was delivered for 14 days， which included Esomeprazole （20 mg twice a day）， Bismuth Potassium Citrate （220 mg twice a day）， Amoxicillin （1 g twice a day） and Clarithromycin （0.5 g twice a day）. The Hp eradication rate and incidence of adverse reactions after medication were compared between two groups. Results Of 200 patients， 11 patients were excluded from subsequent analysis， including 8 cases lost to follow-up and 3 cases with poor compliance of medication. The remaining 189 patients （96 in the observation group and 93 in the control group） completed the entire procedure. According to intention-to-treat （ITT） and per-protocol （PP） set analyses， the Hp eradication rate in the observation group was significantly higher than that in the control group （91.0% vs. 79.0%， P_ITT =0.017； 94.8% vs. 84.9%， P_PP= 0.024）. The incidence of adverse reactions did not significantly differ between two groups （4% vs. 5%，χ ² < 0.001，P > 0.999）. Conclusion The combination therapy of Vonoprazan and high-dose Amoxicillin is an efficacious and safe therapeutic option for patients with initial Hp infection， which is worthy of promotion in grassroot-level hospitals.

Objective To explore the effect of the precise diagnosis and treatment of a case of acute graft-versus-host disease （aGVHD） with rash as the main manifestation under the guidance of immune monitoring. Methods Clinical data and diagnosis and treatment of a child with high-risk acute myeloid leukemia who developed intractable rash after receiving unrelated umbilical cord blood transplantation were retrospectively analyzed. Using the keywords of “hematopoietic stem cell transplantation”“immune reconstitution” and “acute graft-versus-host disease”， relevant literatures were searched from PubMed， Web of Science， CNKI and Wanfang databases， and relevant cases were collected and analyzed. Results Hematopoietic reconstruction was successfully achieved in this 1 year 9 months old girl. On day 22 after transplantation， the skin of head and back neck of the patient showed dense red papules （approximately 19% in area）， with obvious itching. Under the treatment of oral tacrolimus and topical hormone drugs， the rash was not alleviated （area>90%）. Meanwhile， the proportion of CD3⁺ cells， CD8⁺ cells， CD3⁺CD69⁺cells and CD3⁺HLA-DR⁺cells was significantly increased， and the proportion of Treg cells was decreased. The immune indexes supported T lymphocyte activation， then diagnosis of gradeⅡaGVHD （skin grade 3） was made， and the immunosuppressive regimen was strengthened. During this period， the skin rash of the child subsided， but recurred， and accompanied by obvious skin desquamation. The proportion of CD3⁺ cells， CD8⁺ cells， and CD3⁺HLA-DR⁺cells remains relatively high， supporting the diagnosis of aGVHD. Therefore， the anti-GVHD regimen was adjusted and strengthened， and the aGVHD of the child’s skin was mitigated. Regular monitoring showed that the primary disease was completely relieved， the implantation rate was 100%， and the immune reconstitution was stable. During 32-month follow-up after transplantation， the child obtained disease-free survival. Nine relevant literatures were retrieved， all of which demonstrated that the increased proportion of early CD3⁺ cells， CD8⁺ cells， activated T lymphocytes and Treg downregulation were associated with the occurrence of aGVHD. Conclusions The dynamic monitoring of the immune status of patients after allogeneic hematopoietic stem cell transplantation is helpful to evaluate their immune reconstitution and disease status （such as， aGVHD）， thereby assisting clinicians to formulate a reasonable immunosuppressive regimen and bring favorable prognosis to patients.

Objective To investigate the role of transient receptor potential melatonin 2 （TRPM2） ion channels in microglia and astrocytes in temporal lobe epilepsy （TLE） -associated neuroinflammation. Methods Rats were randomly divided into the control group and epilepsy group. Seizure models were induced by lithium chloride-pilocarpine in the epilepsy group， and those in the control group were injected with the same dose of saline. The TLE rats were randomly divided into 7 subgroups according to the observation time after model establishment： acute phase （3 h group， 6 h group， 1 d group， 2 d group）， latent phase （14 d group） and chronic phase （30 d group， 90 d group） subgroups （n = 5）. The expression level of TRPM2 in the hippocampus of TLE rats at different stages were detected， and the cellular localization of TRPM2 in the brain of TLE rats was investigated. BV2 and C8 cell lines were induced by hydrogen peroxide （H₂O₂）. The levels of IL1-β， IL-6 and TNF-α released by BV2 and C8 cells were observed by ELISA. The levels of TRPM2， poly （ADP-ribose） polymerase 1 （PARP-1）， glycogen synthase kinase-3β （GSK-3β） and phosphorylated nuclear factor-κb p65 （p-NF-κb p65） were measured in H₂O₂-induced BV2 and C8 cells. Meantime， the activity of Calcineurin （CaN） in H₂O₂-induced BV2 was observed. Results The TRPM2 expression in 2d epileptic rat hippocampus was increased （P < 0.05）. The expression of TRPM2 and the release of inflammatory cytokines were increased in H₂O₂ induced-BV2 and C8 cells （both P < 0.05）. H₂O₂ could up-regulate the expression of PARP-1 and p-NF-κB p65 and enhance the activity of CaN in BV2 cells （all P < 0.05）. Conclusions Oxidative stress can up-regulate the expression levels of TRPM2 and pro-inflammatory cytokines in microglia and astrocytes. Oxidative stress caused by recurrent epilepsy may mediate the occurrence of TLE-associated neuroinflammation through the TRPM2/ CaN/p-NF-κB p65 pathway in microglia.

Objective To investigate the effect of high keratin 23 （KRT23） expression promoting cell proliferation on liver cancer. Methods The expression of KRT23 in liver cancer was analyzed by bioinformatics. After overexpression and inhibition of KRT23， the effect of KRT23 on the proliferation of liver cancer cells was analyzed by MTT and clone formation assay. The regulatory relationship between KRT23 and AKT signaling pathway was analyzed by software prediction， qPCRand Western blot. Results The expression level of KRT23 in liver cancer was significantly higher than that in normal control tissues. MTT and clone formation assay showed that overexpression of KRT23 promoted cell proliferation， while downregulation of KRT23 inhibited cell proliferation. GSEA analysis， qPCR and Western blot showed that KRT23 could activate the AKT signaling pathway and promote cell proliferation. Meanwhile， inhibition of AKT signaling pathway in overexpressing KRT23 cells significantly reduced cell proliferation. Conclusions KRT23 affects the proliferation of liver cancer cells by regulating the AKT signaling pathway. These results suggest that KRT23 exerts significant effects upon the malignant procession of liver cancer and may play an important role in the occurrence and development of liver cancer.

Objective To analyze the value of thromboelastogram （TEG） parameters in predicting deep venous thrombosis of the lower limbs in patients with chronic obstructive pulmonary disease（COPD）. Methods Thirty-five COPD patients complicated with deep venous thrombosis of the lower limbs were assigned into the observation group， and 35 COPD patients without deep venous thrombosis of the lower limbs of the same period were recruited in the control group. TEG parameters （R value of coagulation reaction time， K value of blood coagulation time， α angle of coagulation and MA value of maximum clot intensity）， routine blood test， blood gas analysis and baseline data were collected within 24 hours after admission. Logistic regression analysis and receiver operating characteristic （ROC） curve analysis were used to analyze the predictive value of all parameters of TEG for deep venous thrombosis of the lower limbs in patients with COPD. Results R value， K value and α angle were significantly correlated with deep venous thrombosis of the lower limbs in patients with COPD （all P < 0.05）. The area under the ROC curve （AUC） of R value was 0.787 （95%CI： 0.679-0.895）， 0.758 for K value （95%CI： 0.646-0.870）， 0.689 for α angle （95%CI： 0.565-0.812）， and 0.660 for MA value （95%CI： 0.533-0.787）， respectively. The combination of four parameters yielded higher sensitivity and specificity for predicting deep venous thrombosis of the lower limbs （AUC：0.882， 95%CI：0.796-0.969， all P < 0.001）， the cut-off value was 0.436， the sensitivity was 94.3% and the specificity was 80%， respectively. Conclusions R value， K value and α angle in TEG are the independent predictors of deep venous thrombosis of the lower limbs in patients with COPD. R value， K value and α angle can properly predict deep venous thrombosis of the lower limbs in patients with COPD， and the combination of R value， K value， α angle and MA value yields higher sensitivity and specificity.

Objective To analyze the changes in tinnitus after stapedotomy combined with artificial stapes implantation， and further explore the factors influencing the efficacy of postoperative tinnitus in patients with otosclerosis. Methods A total of 54 patients （65 ears） with otosclerosis complicated with tinnitus symptoms who underwent stapedotomy combined with artificial stapes implantation were enrolled. The hearing test results of all patients before surgery， 1 and 6 months after surgery were recorded. The scores of Tinnitus Handicap Inventory （THI） and Clinical Global Impression Scale （CGI-S） before surgery， 1 and 6 months after surgery were compared. Results Before surgery， low-frequency （<1 kHz） tinnitus was observed in 31 ears （47.7%）， medium-frequency （1-4 kHz） tinnitus in 14 ears （21.5%）， and high-frequency （≥4 kHz） tinnitus in 20 ears （30.8%）. At 1 month after surgery， tinnitus symptom was alleviated in 48 ears （73.8%） （complete healing and mitigation of tinnitus）， no changes in 16 ears （24.6%）， and tinnitus symptom was aggravated in 1 ear （1.5%）. At 6 months after surgery， tinnitus symptom was eased in 50 ears （76.9%）， no changes in 13 ears （20.0%）， and tinnitus symptom was aggravated in 1 ear （1.5%）. Patients with low- and medium-frequency tinnitus obtained more significant improvement in tinnitus compared with their counterparts with high-frequency tinnitus （P < 0.001 and P = 0.013）. At postoperative 1 month， the changes in subjective tinnitus， such as the average air conduction and bone conduction thresholds significantly differed from those before surgery （all P < 0.001）， whereas there was no statistical significance in air-bone gap （P > 0.05）. Conclusions Stapedotomy combined with artificial stapes implantation can not only restore the hearing of patients with otosclerosis， but also mitigate the tinnitus symptoms in most patients. Patients with low- and medium-frequency tinnitus before surgery can obtain more significant improvements in tinnitus after surgery than those with high-frequency tinnitus.

Subdural abscesses are located between the dura mater and the subarachnoid space， which are primarily caused by Staphylococci and Streptococci， etc. Subdural abscesses caused by Salmonella Typhi are extremely rare. In this article， a patient with acquired immunodeficiency syndrome （AIDS） presented with a subdural abscess with fever and headache. Blood culture and next-generation sequencing （NGS） of cerebrospinal fluid were indicative of Salmonella Typhi infection. He was treated with sensitive antibiotics based on drug sensitivity testing combined with surgical drainage， and Salmonella Typhi was cultured in the drainage sample. During 4-month postoperative follow-up， relevant symptoms were mitigated and imaging examination results were normal. This case prompts that intracranial infection caused by Salmonella Typhi is extremely rare， which can occur in immunocompromised population. Blood culture， pus culture and NGS are of significant value in differential diagnosis of Salmonella Typhi infection. After the diagnosis is confirmed， sensitive antibiotics combined with surgical treatment can achieve high efficacy.