1.Explainable Machine Learning Model for Predicting Prognosis in Patients with Malignant Tumors Complicated by Acute Respiratory Failure: Based on the eICU Collaborative Research Database in the United States
Zihan NAN ; Linan HAN ; Suwei LI ; Ziyi ZHU ; Qinqin ZHU ; Yan DUAN ; Xiaoting WANG ; Lixia LIU
Medical Journal of Peking Union Medical College Hospital 2026;17(1):98-108
To develop and validate a model for predicting intensive care unit (ICU) mortality risk in patients with malignant tumors complicated by acute respiratory failure (ARF) based on an explainable machine learning framework. Clinical data of patients with malignant tumors and ARF were extracted from the eICU Collaborative Research Database in the United States, including demographic characteristics, comorbidities, vital signs, laboratory test indicators, and major interventions within the first 24 hours after ICU admission.The study outcome was ICU death.Enrolled patients were randomly divided into a training set and a validation set at a ratio of 7:3.Predictor variables were selected using least absolute shrinkage and selection operator (LASSO) regression.Five machine learning algorithms-extreme gradient boosting (XGBoost), support vector machine (SVM), Logistic regression, multilayer perceptron (MLP), and C5.0 Decision Tree-were employed to construct predictive models.Model performance was evaluated based on the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and other metrics.The optimal model was further interpreted using the Shapley additive explanations (SHAP) algorithm. A total of 3196 patients with malignant tumors complicated by ARF were included.The training set comprised 2, 261 patients and the validation set 935 patients; 683 patients died during ICU stay, while 2513 survived.LASSO regression ultimately selected 12 variables closely associated with patient ICU outcomes, including sepsis comorbidity, use of vasoactive drugs, and within the first 24 hours after ICU admission: minimum mean arterial pressure, maximum heart rate, maximum respiratory rate, minimum oxygen saturation, minimum serum bicarbonate, minimum blood urea nitrogen, maximum white blood cell count, maximum mean corpuscular volume, maximum serum potassium, and maximum blood glucose.After model evaluation, the XGBoost model demonstrated the best performance.The AUCs for predicting ICU mortality risk in the training and validation sets were 0.940 and 0.763, respectively; accuracy was 88.3% and 81.2%;sensitivity was 98.5% and 95.9%.Its predictive performance also remained optimal in sensitivity analyses.SHAP analysis indicated that the top five variables contributing to the model's predictions were minimum oxygen saturation, minimum serum bicarbonate, minimum mean arterial pressure, use of vasoactive drugs, and maximum white blood cell count. This study successfully developed a mortality risk prediction model for ICU patients with malignant tumors complicated by ARF based on a large-scale dataset and performed explainability analysis.The model aids clinicians in early identification of high-risk patients and implementing individualized interventions.
2.LINC00657 Promotes Malignant Progression of Cervical Cancer by Sponging miR-30a-5p to Regulate Skp2 Expression
Changhui ZHOU ; Jingqin REN ; Zhen CHEN ; Qi YAN ; Nan YANG ; Jiaqi ZHAO ; Rong LI
Cancer Research on Prevention and Treatment 2026;53(2):103-111
Objective To investigate the role and regulatory mechanism of LINC00657 in the progression of cervical cancer. Methods Bioinformatics analysis predicted potential binding sites between LINC00657 and miR-30a-5p and between miR-30a-5p and Skp2. These sites were verified by using RNA immunoprecipitation and dual-luciferase reporter experiments. LINC00657, miR-30a-5p, and Skp2 mRNA expression levels in cervical cancer tissues and cell lines were assessed by utilizing RT-qPCR. Western blot analysis was employed to examine the protein levels of Skp2 in cells and subcutaneous xenograft tumor models in nude mice. Immunohistochemistry was applied to analyze Skp2 expression in animal tissues. The cellular processes of cervical cancer cell lines were evaluated through CCK-8, scratch, and Transwell assays. Results LINC00657 and Skp2 presented binding sites for miR-30a-5p. In cervical cancer, LINC00657 and Skp2 showed high expression levels (P<0.05), whereas miR-30a-5p displayed low expression (P<0.05). Functional experiments demonstrated that linc00657 upregulates Skp2 expression, a process that is dependent on its sequestration of miR-30a-5p. Conclusion LINC00657 promoted the malignant progression of cervical cancer by upregulating Skp2 expression through specifically sequestering miR-30a-5p, thereby relieving its inhibitory effect on the target gene Skp2.
3.A Case Report of Pachydermoperiostosis by Multidisciplinary Diagnosis and Treatment
Jie ZHANG ; Yan ZHANG ; Li HUO ; Ke LYU ; Tao WANG ; Ze'nan XIA ; Xiao LONG ; Kexin XU ; Nan WU ; Bo YANG ; Weibo XIA ; Rongrong HU ; Limeng CHEN ; Ji LI ; Xia HONG ; Yan ZHANG ; Yagang ZUO
JOURNAL OF RARE DISEASES 2025;4(1):75-82
A 20-year-old male patient presented to the Department of Dermatology of Peking Union Medical College Hospital with complaints of an 8-year history of facial scarring, swelling of the lower limbs, and a 4-year history of scalp thickening. Physical examination showed thickening furrowing wrinkling of the skin on the face and behind the ears, ciliary body hirsutism, blepharoptosis, and cutis verticis gyrate. Both lower limbs were swollen, especially the knees and ankles. The skin of the palms and soles of the feet was keratinized and thickened. Laboratory examination using bone and joint X-ray showed periostosis of the proximal middle phalanges and metacarpals of both hands, distal ulna and radius, tibia and fibula, distal femurs, and metatarsals.Genetic testing revealed two variants in
4.Clinical and Ultrasound Manifestations of Immune Checkpoint Inhibitor-Associated Enterocolitis:Report of One Case.
Xiao-Yan ZHANG ; Jing QIN ; Xiao-Qing LI ; Guan-Nan ZHANG ; Wen-Bo LI
Acta Academiae Medicinae Sinicae 2025;47(5):771-775
Immune checkpoint inhibitor-associated enterocolitis is an immune-related adverse reaction during tumor treatment with immune checkpoint inhibitors.In this article,we present the clinical data and ultrasound manifestations of a patient with immune checkpoint inhibitor-associated enterocolitis,aiming to share diagnostic and therapeutic insights.
Humans
;
Immune Checkpoint Inhibitors/adverse effects*
;
Enterocolitis/chemically induced*
;
Ultrasonography
;
Male
5.Construction of A Chiral Separation Method Using Polystyrene-cyclodextrin Metal-organic Framework Coating for Open-tubular Capillary Electrochromatography
Yan ZHANG ; Hao-Yu LI ; Cai LIU ; Rong-Yue ZHANG ; Xiao-Nan HE ; Juan QIAO
Chinese Journal of Analytical Chemistry 2025;53(10):1751-1760
By using a strategy of leveraging the ability of metal-organic frameworks(MOFs)materials to precisely regulate the spatial orientation of cyclodextrins(CD),a polystyrene-modified γ-cyclodextrin metal-organic frameworks(PS-CD-MOFs)capillary coating was established and applied to the chiral separation of amino acids in open-tubular capillary electrochromatography(OT-CEC)based on the excellent film-forming property of polystyrene(PS).Characterization results by Fourier transform infrared spectroscopy(FT-IR),X-ray diffraction(XRD)indicated that PS was successfully grafted onto the surface of CD-MOFs.The modification significantly improved the structural stability and thermal stability of CD-MOFs while maintaining the integrity of the MOFs.The PS-CD-MOFs coated capillary electrophoresis system exhibited excellent performance in separating dansylated amino acid enantiomers(Dns-D,L-AAs).Specifically,under the optimal separation conditions(Sodium dodecyl sulfate/boric acid buffer system,20 cm capillary,and 10 kV effective voltage),good separation was achieved for D,L-methionine(D,L-Met)and D,L-serine(D,L-Ser).Further quantitative analysis showed that Dns-D,L-Met presented a good linear relationship in the concentration range of 10.0-1500.0 μmol/L,with a correlation coefficient close to 0.998,demonstrating high sensitivity and repeatability.Not only did it overcome the problem that traditional CD(used as additives in capillary electrophoresis)could not precisely control the spatial orientation of chiral resolvents,but also it solved the issues of insufficient stability and bonding amount of CD-MOFs coatings by utilizing the excellent film-forming property of polymers on the inner wall of capillaries.This study provided an efficient and controllable new strategy for construction of chiral separation stationary phases.
6.Modeling methods and evaluation criteria in animal models of steroid-induced osteonecrosis of the femoral head
Xiuli WU ; Xiaoxia YAN ; Zhiqiang REN ; Nan SUN ; Jinju LI
Chinese Journal of Tissue Engineering Research 2025;29(21):4560-4567
BACKGROUND:Steroid-induced osteonecrosis of the femoral head is a key risk factor for non-traumatic femoral head necrosis,and the incidence of it has gradually increased in recent years,but its specific pathogenesis is still unclear,and finding a reasonable animal model is essential for disease research and treatment.OBJECTIVE:To review the commonly used animal models of steroid-induced osteonecrosis of the femoral head in recent years,and to analyze the advantages and disadvantages of different modeling methods and evaluation criteria,so as to provide a reference for follow-up research. METHODS:The articles published from 2013 to 2023 were searched in CNKI,WanFang Data,and PubMed with the keywords "femoral head necrosis,osteonecrosis of the femoral head,steroid-induced osteonecrosis of the femoral head,animal model." Finally,61 articles were involved for comprehensive analysis according to the inclusion criteria,including 38 English articles and 23 Chinese articles.RESULTS AND CONCLUSION:(1) Rabbits,rats,and chickens are the animals that are widely used in the study of steroid-induced osteonecrosis of the femoral head models. (2) Hormone combined with lipopolysaccharide or horse serum modeling has low mortality,high success rate,and strong stability. (3) Histopathology is the gold standard for the evaluation model,but the experiment needs to be terminated,which is not conducive to subsequent experimental research,so finding a non-invasive alternative method is still the direction of future efforts. (4) An ideal model of steroid-induced osteonecrosis of the femoral head has not yet been explored,and future researchers need to continue to strive to achieve breakthroughs in this field as soon as possible.
7.Wu Jutong's Diagnosis and Treatment of Evil Invading Shaoyang
Manrou YAN ; Nan LI ; Jingsi FU ; Linlin QIAN ; Xiaobei MA
Journal of Nanjing University of Traditional Chinese Medicine 2025;41(4):431-435
Wu Jutong's academic thoughts are inherited from Ye Tianshi and traced back to Zhongjing,and have a unique under-standing and play of the evil invading Shaoyang.Wu Jutong believes that Shaoyang has two major functional characteristics:ascending and pivoting.The principle of treating evil invading Shaoyang is to follow the ascending nature of Shaoyang and restore the pivoting function of Shaoyang.For Shaoyang fire-heat syndrome,follow Shaoyang to ascend and disperse to penetrate heat,and often use light,scavenging and dispersing products;for Shaoyang phlegm-fluid syndrome,creat Xiangfu Xuanfuhua Decoction to benefit the liver and gallbladder and resolve phlegm-fluid;for malaria invading Shaoyang,turn the pivot to stop malaria,and use Xiaochaihu Decoction and Qinghao Biejia Decoction for the treatment;for yin evil invading Shaoyang and Jueyin,Xiao Chaihu Decoction and Bupleurum chinense DC.are often used to turn the pivot of Shaoyang and warm and unchoke Jueyin.The contraindications of Xiao Chaihu Decoction are al-so summarized.
8.Clinical Features, Prognostic Analysis and Predictive Model Construction of Central Nervous System Invasion in Peripheral T-Cell Lymphoma.
Ya-Ting MA ; Yan-Fang CHEN ; Zhi-Yuan ZHOU ; Lei ZHANG ; Xin LI ; Xin-Hua WANG ; Xiao-Rui FU ; Zhen-Chang SUN ; Yu CHANG ; Fei-Fei NAN ; Ling LI ; Ming-Zhi ZHANG
Journal of Experimental Hematology 2025;33(3):760-768
OBJECTIVE:
To investigate the clinical features and prognosis of central nervous system (CNS) invasion in peripheral T-cell lymphoma (PTCL) and construct a risk prediction model for CNS invasion.
METHODS:
Clinical data of 395 patients with PTCL diagnosed and treated in the First Affiliated Hospital of Zhengzhou University from 1st January 2013 to 31st December 2022 were analyzed retrospectively.
RESULTS:
The median follow-up time of 395 PTCL patients was 24(1-143) months. There were 13 patients diagnosed CNS invasion, and the incidence was 3.3%. The risk of CNS invasion varied according to pathological subtype. The incidence of CNS invasion in patients with anaplastic large cell lymphoma (ALCL) was significantly higher than in patients with angioimmunoblastic T-cell lymphoma (AITL) (P <0.05). The median overall survival was significantly shorter in patients with CNS invasion than in those without CNS involvement, with a median survival time of 2.4(0.6-127) months after diagnosis of CNS invasion. The results of univariate and multivariate analysis showed that more than 1 extranodal involvement (HR=4.486, 95%CI : 1.166-17.264, P =0.029), ALCL subtype (HR=9.022, 95%CI : 2.289-35.557, P =0.002) and ECOG PS >1 (HR=15.890, 95%CI : 4.409-57.262, P <0.001) were independent risk factors for CNS invasion in PTCL patients. Each of these risk factors was assigned a value of 1 point and a new prediction model was constructed. It could stratify the patients into three distinct groups: low-risk group (0-1 point), intermediate-risk group (2 points) and high-risk group (3 points). The 1-year cumulative incidence of CNS invasion in the high-risk group was as high as 50.0%. Further evaluation of the model showed good discrimination and accuracy, and the consistency index was 0.913 (95%CI : 0.843-0.984).
CONCLUSION
The new model shows a precise risk assessment for CNS invasion prediction, while its specificity and sensitivity need further data validation.
Humans
;
Lymphoma, T-Cell, Peripheral/pathology*
;
Prognosis
;
Retrospective Studies
;
Central Nervous System Neoplasms/pathology*
;
Neoplasm Invasiveness
;
Male
;
Female
;
Central Nervous System/pathology*
;
Middle Aged
;
Adult
9.Additional role of low-density lipoprotein cholesterol on the risk of osteoporosis in men with or without coronary heart disease: a real-world longitudinal study.
Jing ZENG ; Zi-Mo PAN ; Ting LI ; Ze-Yu CHEN ; Xiao-Yan CAI ; Mei-Liang GONG ; Xin-Li DENG ; Sheng-Shu WANG ; Nan LI ; Miao LIU ; Chun-Lin LI
Journal of Geriatric Cardiology 2025;22(2):219-228
BACKGROUND:
Early control of low-density lipoprotein cholesterol (LDL-C) is crucial for reducing the progress of cardiovascular disease. However, its additional role to the risk of primary osteoporosis in men with coronary heart disease was inconclusive. Our study aims to determine the association of LDL-C and its trajectories for osteoporosis risk in the middle-aged and aged men of China.
METHODS:
The retrospective cohort study of 1546 men aged 69.74 ± 11.30 years conducted in Beijing, China from 2015 to 2022. And the incidence of primary osteoporosis was annually recorded. LDL-C trajectories were further identified by latent class growth model using repeated measurements of LDL-C. The association of baseline LDL-C for osteoporosis was estimated using hazard ratio (HR) with 95% CI in Cox proportional hazard model, while mean level and trajectories of LDL-C for osteoporosis were evaluated using odds ratio (OR) with 95% CI in logistic regression model.
RESULTS:
During the median 6.2-year follow-up period, 70 men developed primary osteoporosis. The higher level of baseline LDL-C (HR = 1.539, 95% CI: 1.012-2.342) and mean LDL-C (OR = 2.190, 95% CI: 1.443-3.324) were associated with higher risk of osteoporosis in men with coronary heart disease after adjusted for covariates. Compared with those in the LDL-C trajectory of low-stable decrease, participants with medium-fluctuant trajectory, whose longitudinal LDL-C started with a medium LDL-C level and appeared an increase and then decrease, were negatively associated with osteoporosis risk (OR = 2.451, 95% CI: 1.152-5.216). And participants with initially high LDL-C level and then a rapid decrease demonstrated a tendency towards reduced risk (OR = 0.718, 95% CI: 0.212-2.437).
CONCLUSIONS
Elevated LDL-C level and its long-term fluctuation may increase the risk of primary osteoporosis in men. Early controlling a stable level of LDL-C is also essential for bone health.
10.Shenmai Injection Reduces Cardiomyocyte Apoptosis Induced by Doxorubicin through miR-30a/Bcl-2.
Xiao-Nan ZHANG ; Yan-Yang LI ; Shi-Chao LYU ; Qiu-Jin JIA ; Jun-Ping ZHANG ; Long-Tao LIU
Chinese journal of integrative medicine 2025;31(3):240-250
OBJECTIVE:
To explore the molecular mechanism of Shenmai Injection (SMI) against doxorubicin (DOX) induced cardiomyocyte apoptosis.
METHODS:
A total of 40 specific pathogen-free (SPF) male Sprague Dawley (SD) male rats were divided into 5 groups based on the random number table, including the control group, the model group, miR-30a agomir group, SMI low-dose (SMI-L) group, and SMI high-dose (SMI-H) group, with 8 rats in each group. Except for the control group, the rats were injected weekly with DOX (2 mg/kg) in the tail vein for 4 weeks to induce myocardial injury, and were given different regimens of continuous intervention for 2 weeks. Cardiac function was detected by echocardiography and myocardial pathological changes were observed by Van Gieson (VG) staining. Myocardial injury serum markers, including creatine kinase (CK), lactate dehydrogenase (LDH), troponin T (cTnT), N-terminal pro-brain natriuretic peptide (NT-proBNP), soluble ST2 (sST2), and growth differentiation factor-15 (GDF-15) were detected by enzyme linked immunosorbent assay (ELISA). Cardiomyocyte apoptosis was observed by terminal deoxynucleotidyl transferase-mediated biotinylated dUTP triphosphate nick end labeling (TUNEL) and transmission electron microscopy, and the expressions of target proteins and mRNA were detected by Western blot and quantitative real time polymerase chain reaction (qRT-RCR), respectively.
RESULTS:
The treatment with different doses of SMI reduced rat heart mass index and left ventricular mass index (P<0.05), significantly improved the left ventricular ejection fraction (P<0.05), decreased the levels of serum CK, LDH, cTnT, and NT-proBNP (P<0.05 or P<0.01), reduced the levels of serum sST2 and GDF-15 (P<0.05 or P<0.01), decreased the collagen volume fraction, reduced the expressions of rat myocardial type I and type III collagen (P<0.05 or P<0.01), and effectively alleviated myocardial fibrosis. And the study found that SMI promoted the expression levels of miR-30a and Bcl-2 in myocardium, and down-regulated the expression of Bax, which inhibited the activation of Caspase-3 and Caspase-9 (P<0.05 or P<0.01), and improved myocardial cell apoptosis.
CONCLUSIONS
SMI can alleviate myocardial injury and apoptosis caused by DOX, and its mechanism possibly by promoting the targeted expression of myocardial Bcl-2 protein through miR-30a.
Animals
;
Myocytes, Cardiac/metabolism*
;
Apoptosis/drug effects*
;
MicroRNAs/genetics*
;
Rats, Sprague-Dawley
;
Male
;
Drugs, Chinese Herbal/administration & dosage*
;
Doxorubicin/pharmacology*
;
Proto-Oncogene Proteins c-bcl-2/genetics*
;
Drug Combinations
;
Injections
;
Rats

Result Analysis
Print
Save
E-mail