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.Expert consensus on neoadjuvant PD-1 inhibitors for locally advanced oral squamous cell carcinoma (2026)
LI Jinsong ; LIAO Guiqing ; LI Longjiang ; ZHANG Chenping ; SHANG Chenping ; ZHANG Jie ; ZHONG Laiping ; LIU Bing ; CHEN Gang ; WEI Jianhua ; JI Tong ; LI Chunjie ; LIN Lisong ; REN Guoxin ; LI Yi ; SHANG Wei ; HAN Bing ; JIANG Canhua ; ZHANG Sheng ; SONG Ming ; LIU Xuekui ; WANG Anxun ; LIU Shuguang ; CHEN Zhanhong ; WANG Youyuan ; LIN Zhaoyu ; LI Haigang ; DUAN Xiaohui ; YE Ling ; ZHENG Jun ; WANG Jun ; LV Xiaozhi ; ZHU Lijun ; CAO Haotian
Journal of Prevention and Treatment for Stomatological Diseases 2026;34(2):105-118
Oral squamous cell carcinoma (OSCC) is a common head and neck malignancy. Approximately 50% to 60% of patients with OSCC are diagnosed at a locally advanced stage (clinical staging III-IVa). Even with comprehensive and sequential treatment primarily based on surgery, the 5-year overall survival rate remains below 50%, and patients often suffer from postoperative functional impairments such as difficulties with speaking and swallowing. Programmed death receptor-1 (PD-1) inhibitors are increasingly used in the neoadjuvant treatment of locally advanced OSCC and have shown encouraging efficacy. However, clinical practice still faces key challenges, including the definition of indications, optimization of combination regimens, and standards for efficacy evaluation. Based on the latest research advances worldwide and the clinical experience of the expert group, this expert consensus systematically evaluates the application of PD-1 inhibitors in the neoadjuvant treatment of locally advanced OSCC, covering combination strategies, treatment cycles and surgical timing, efficacy assessment, use of biomarkers, management of special populations and immune related adverse events, principles for immunotherapy rechallenge, and function preservation strategies. After multiple rounds of panel discussion and through anonymous voting using the Delphi method, the following consensus statements have been formulated: 1) Neoadjuvant therapy with PD-1 inhibitors can be used preoperatively in patients with locally advanced OSCC. The preferred regimen is a PD-1 inhibitor combined with platinum based chemotherapy, administered for 2-3 cycles. 2) During the efficacy evaluation of neoadjuvant therapy, radiographic assessment should follow the dual criteria of Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 and immune RECIST (iRECIST). After surgery, systematic pathological evaluation of both the primary lesion and regional lymph nodes is required. For combination chemotherapy regimens, PD-L1 expression and combined positive score need not be used as mandatory inclusion or exclusion criteria. 3) For special populations such as the elderly (≥ 70 years), individuals with stable HIV viral load, and carriers of chronic HBV/HCV, PD-1 inhibitors may be used cautiously under the guidance of a multidisciplinary team (MDT), with close monitoring for adverse events. 4) For patients with a poor response to neoadjuvant therapy, continuation of the original treatment regimen is not recommended; the subsequent treatment plan should be adjusted promptly after MDT assessment. Organ transplant recipients and patients with active autoimmune diseases are not recommended to receive neoadjuvant PD-1 inhibitor therapy due to the high risk of immune related activation. Rechallenge is generally not advised for patients who have experienced high risk immune related adverse events such as immune mediated myocarditis, neurotoxicity, or pneumonitis. 5) For patients with a good pathological response, individualized de escalation surgery and function preservation strategies can be explored. This consensus aims to promote the standardized, safe, and precise application of neoadjuvant PD-1 inhibitor strategies in the management of locally advanced OSCC patients.
3.Genetic disease diagnosis and treatment in Shanghai: Survey and countermeasures for clinical genetics specialist training.
Xiaoju HUANG ; Lin HAN ; Li CAO ; Taosheng HUANG ; Duan MA ; Jian WANG ; Wenjuan QIU ; Fanyi ZENG ; Luming SUN ; Chenming XU ; Songchang CHEN ; Xinyu KUANG ; Hong TIAN
Chinese Journal of Medical Genetics 2026;43(4):241-247
OBJECTIVE:
To investigate the current status of clinical genetics specialization development and the diagnostic and therapeutic capabilities for hereditary diseases across medical institutions in Shanghai, and to assess the necessity and feasibility of establishing training bases for clinical genetics specialists.
METHODS:
By employing a cross-sectional survey design, the Clinical Genetics Committee of Shanghai Medical Association has conducted questionnaire surveys from March to April 2025 across 54 healthcare institutions in Shanghai (including 33 tertiary hospitals and 21 secondary hospitals). The survey involved administrative departments and medical personnel from 15 clinical specialties. The survey has covered current genetic disease diagnosis and treatment practices, relevant and specialised disease types, genetic department establishment, testing capabilities, personnel teams, and training requirements.
RESULTS:
The results revealed that 78.0% of clinical departments surveyed had treated patients with hereditary disorders. Shanghai possesses diagnostic and therapeutic expertise for over 95% of hereditary diseases listed in its rare disease catalogue, reflecting both the practical clinical demand for such conditions and the city's overall diagnostic and therapeutic strengths in this field. Nevertheless, significant disparities exist in the development of genetics departments across different tiers of healthcare institutions. Resources for genetic testing capabilities (including molecular, cellular, and biochemical testing) are also unevenly distributed across different tiers of hospitals. The survey further revealed that only 26.0% of departments believe that their current physician structure fully meets the diagnostic and treatment demands. Over 90% of departments consider standard training for clinical genetic specialists necessary, with 74.0% expressing willingness to participate in establishing training bases. Based on above findings and thorough deliberation, the Clinical Genetics Committee of the Shanghai Medical Association proposes advancing specialist training and discipline development through establishing a standard training system. The committee has drafted a three-year training protocol featuring a "joint training"-centered model, recommending a pilot-first, dynamically optimized strategy for steadily advancing training base development.
CONCLUSION
Shanghai faces substantial demand for genetic disease diagnosis and treatment, yet exhibits shortcomings in clinical genetics specialization development, resource allocation, and talent pipeline cultivation. To establish a standard training system holds significant practical importance and is underpinned by a broad demand.
Humans
;
China
;
Surveys and Questionnaires
;
Genetic Diseases, Inborn/genetics*
;
Cross-Sectional Studies
;
Genetics, Medical/education*
;
Genetic Testing
4.Application of machine learning in the diagnosis and treatment of chronic hepatitis C
Hua HAN ; Zhongping DUAN ; Yang WANG
Journal of Clinical Hepatology 2025;41(1):141-144
With the development of artificial intelligence, machine learning has shown great potential in the field of medical health. Machine learning conducts a comprehensive analysis of patient data including clinical features, blood tests, and imaging examinations and establishes corresponding mathematical models to achieve the diagnosis and treatment of diseases and the prediction of disease conditions, thereby guiding disease management. With reference to the latest research findings, this article reviews the application of machine learning in chronic hepatitis C and related research advances.
5.Salidroside promotes osteogenic differentiation of MC3T3-E1 cells:an in vitro experiment
Zhaohui LIU ; Xiaoqian HAN ; Xin DUAN ; Pengda GUO ; Yuntao ZHANG
Chinese Journal of Tissue Engineering Research 2025;29(2):231-237
BACKGROUND:Bone defects can directly affect the success rate and long-term stability of dental implants.Studies have shown that salidroside has the ability to promote the proliferation and differentiation of osteoblasts,but less is reported on its pathways related to osteogenic differentiation. OBJECTIVE:To investigate the effects of salidroside on the proliferation and differentiation of MC3T3-E1 cells and the expression of related genes and proteins through in vitro cell experiments. METHODS:Cell counting kit-8 test and alkaline phosphatase test were used to determine the optimal concentration of salidroside(0.5,1,5,10,and 50 μmol/L)in promoting the proliferation and differentiation of MC3T3-E1 cells.There were four groups in the experiment:control group,salidroside group,salidroside+LY294002 group,and LY294002 group,which were cultured with osteogenic induction solution,osteogenic induction solution containing 10 μmol/L salidroside,osteogenic induction solution containing 10 μmol/L salidroside+10 μmol/L LY294002,and osteogenic induction solution containing 10 μmol/L LY294002,respectively.The effects of salidroside and LY294002,an inhibitor of the PI3K/Akt signaling pathway,on the expressions of genes and proteins related to osteogenesis were observed. RESULTS AND CONCLUSION:Cell counting kit-8 assay and alkaline phosphatase assay showed that salidroside promoted the proliferation of MC3T3-E1 cells most significantly at 10 μmol/L.Compared with the control group,salidroside could promote mineralization,promote cell adhesion,reduce cell death,increase mRNA expression of Runx-2,osteocalcin and osteopontin(P<0.01),and increase protein expression of Runx-2 and p-Akt(P<0.01).However,the addition of LY294002 reversed the above results.These findings indicate that salidroside can promote the mineralization of MC3T3-E1 cells and the expression of osteogenesis-related genes and proteins,which may be related to the activation of PI3K/Akt signaling pathway.
6.Evolution of global disease burden,risk factors and trend projection of chronic obstructive pulmonary disease among individuals under 50 years old from 1990 to 2021
Yide WANG ; Hongxia DUAN ; Yingqi WANG ; Yidie BAO ; Linhong JIANG ; Xiaoyu HAN ; Peijun LI ; Weibing WU ; Xiaodan LIU
Journal of Army Medical University 2025;47(12):1388-1400
Objective To systematically assess the spatiotemporal distribution,risk factors,and future trends of chronic obstructive pulmonary disease(COPD)among individuals under 50 years of age globally from 1990 to 2021 based on Global Burden of Disease(GBD)data in order to provide support for the formulation of prevention and control strategies of the disease.Methods The GBD data from 1990 to 2021 were analyzed for the incidence,mortality,disability-adjusted life years(DALYs),and estimated annual percentage change(EAPC)of COPD in<50-year-old individuals across 204 countries and regions.The data were stratified by age,sex,region,country,and sociodemographic index(SDI).The COPD trends until 2035 were predicted.Results In 2021,the global incidence of early-onset COPD was estimated at 2.5 million cases(95%uncertainty interval:2.09~2.96 million),representing a 50.55%increase compared to 1990.Significant regional heterogeneity was observed,with low SDI regions experiencing a 134.08%increase,whereas high SDI regions exhibited a rise-then-fall trend.Risk factor analysis identified environmental and occupational exposures(air pollution,ambient ozone pollution,household air pollution from solid fuels,etc.)and smoking as the primary etiological factors.Notably,household solid fuel exposure accounted for 50.90%of COPD-related deaths in low SDI regions,compared to only 0.03%in high SDI regions.Projections indicated that by 2035,the global burden of early-onset COPD will increase to 2.59 million cases.Conclusion The global disease burden of COPD among people under 50 years increased significantly from 1990 to 2021,with pronounced disparities across regions and socioeconomic levels.COPD deaths in low-SDI regions are strongly associated with solid fuel exposure and particulate matter pollution,and these regions are expected to remain the main drivers of global COPD incidence growth through 2035.
7.Study on the correlation between serum hyperphosphorylated Tau protein,β-amyloid protein and mild cognitive impairment in patients with obstructive sleep apnea syndrome
Li FENG ; Yi DUAN ; Na LI ; Xiaonan HAN ; Shanshan DI
Chinese Archives of Otolaryngology-Head and Neck Surgery 2025;32(2):116-120
OBJECTIVE To investigate the correlation of serum P-Tau protein and β-amyloid protein expression levels with obstructive sleep apnea syndrome(OSAS)and mild cognitive impairment(MCI)patients,and their diagnostic value.METHODS From December 2020 to December 2023,120 patients with OSAS admitted to Third Hospital of Shijiazhuang were collected as the case group.According to the diagnostic criteria for MCI,patients were grouped into OSAS without MCI group(40 cases)and OSAS with MCI group(80 cases).ELISA method was applied to detect the levels of serum P-Tau protein and β-amyloid protein.Spearman method was applied to analyze the correlation between serum P-Tau protein,β-amyloid protein,and MCI.Multivariate logistic regression was applied to analyze the influencing factors of OSAS patients with MCI.ROC curve was applied to evaluate the diagnostic efficacy of serum P-Tau protein and β-amyloid protein in OSAS patients with MCI.RESULTS The Montreal Cognitive Assessment(MoCA)score in the OSAS with MCI group was obviously lower than that in the OSAS without MCI group(P<0.05).The expression levels of P-Tau protein and β-amyloid protein in the OSAS with MCI group were obviously higher than those in the OSAS without MCI group(P<0.05).The expression levels of serum P-Tau protein and β-amyloid protein in OSAS patients were negatively correlated with MoCA score(r=-0.346,-0.565,P<0.001).Serum P-Tau protein and β-amyloid protein were risk factors for OSAS with MCI(P<0.05).The AUC of the expression levels of serum P-Tau protein,β-amyloid protein,and their combination for OSAS with MCI was 0.751,0.848,and 0.928,respectively.The combined evaluation of the two showed better results(Zcombination-P-Tau protein=4.102,P<0.001;Zcombination-β amyloid protein=2.147,P=0.032).CONCLUSION The expression of serum P-Tau protein and β-amyloid protein is upregulated in OSAS patients with MCI,they are risk factors for the development of MCI in OSAS patients.The combined detection of the two has higher diagnostic efficacy.
8.Analysis on the compositional differences of different processing products of Atractylodes lancea Rhizoma based on HS-GC-MS and UPLC-Q-Orbitrap HRMS
Li WANG ; Rong LUO ; Xuyang HAN ; Kaijing WANG ; Wei XIAO ; Dechun JIANG ; Songleng DUAN ; Peng ZHANG ; Yanxin ZHAI ; Jiankun WU
International Journal of Traditional Chinese Medicine 2025;47(6):833-842
Objective:To compare the differences in chemical compositions before and after processing by different processing methods; To optimize the processing method of Atractylodes lancea Rhizoma.Methods:Atractylodes lancea Rhizoma was processed by stir-frying with bran and treating with rice washing water. The volatile and non-volatile components of raw Atractylodes lancea Rhizoma, bran-fried Atractylodes lancea Rhizoma and rice washing water treated Atractylodes lancea Rhizome were qualitatively analyzed by headspace gas chromatography-mass spectrometry (HS-GC-MS) and ultra-performance liquid chromatography-quadrupole-electrostatic field orbitrap high-resolution mass spectrometry (UPLC-Q-Orbitrap HRMS), and the differences in chemical composition before and after processing were compared.Results:The volatile components of the three different products were determined to have 18 common components, such as agarospirol, β-eudesol, etc. In addition, 86 non-volatile components were determined. The peak area response value of atractylodin, the index component prescribed by pharmacopoeia, decreased after processing, but there was little difference in bran stir-frying and rice-washed water frying.Conclusions:Different processing methods have certain effects on the chemical composition of Atractylodes lancea Rhizoma. Among them, the bran-frying method is superior in improving the quality of preparations, reducing production costs and improving production efficiency. The bran-fried product can be used as raw material for preparation production.
9.The systemic inflammatory response index as a risk factor for all-cause and cardiovascular mortality among individuals with coronary artery disease: evidence from the cohort study of NHANES 1999-2018.
Dao-Shen LIU ; Dan LIU ; Hai-Xu SONG ; Jing LI ; Miao-Han QIU ; Chao-Qun MA ; Xue-Fei MU ; Shang-Xun ZHOU ; Yi-Xuan DUAN ; Yu-Ying LI ; Yi LI ; Ya-Ling HAN
Journal of Geriatric Cardiology 2025;22(7):668-677
BACKGROUND:
The association of systemic inflammatory response index (SIRI) with prognosis of coronary artery disease (CAD) patients has never been investigated in a large sample with long-term follow-up. This study aimed to explore the association of SIRI with all-cause and cause-specific mortality in a nationally representative sample of CAD patients from United States.
METHODS:
A total of 3386 participants with CAD from the National Health and Nutrition Examination Survey (NHANES) 1999-2018 were included in this study. Cox proportional hazards model, restricted cubic spline (RCS), and receiver operating characteristic curve (ROC) were performed to investigate the association of SIRI with all-cause and cause-specific mortality. Piece-wise linear regression and sensitivity analyses were also performed.
RESULTS:
During a median follow-up of 7.7 years, 1454 all-cause mortality occurred. After adjusting for confounding factors, higher lnSIRI was significantly associated with higher risk of all-cause (HR = 1.16, 95% CI: 1.09-1.23) and CVD mortality (HR = 1.17, 95% CI: 1.05-1.30) but not cancer mortality (HR = 1.17, 95% CI: 0.99-1.38). The associations of SIRI with all-cause and CVD mortality were detected as J-shaped with threshold values of 1.05935 and 1.122946 for SIRI, respectively. ROC curves showed that lnSIRI had robust predictive effect both in short and long terms.
CONCLUSIONS
SIRI was independently associated with all-cause and CVD mortality, and the dose-response relationship was J-shaped. SIRI might serve as a valid predictor for all-cause and CVD mortality both in the short and long terms.
10.Cerium single-atom catalysts-armed Lactobacillus reuteri for multipronged anti-inflammatory/anti-fibrotic therapy of inflammatory bowel disease.
Yinying PU ; Shaorong HUANG ; Shuang GAO ; Yangying DUAN ; Wenhao LI ; Qiyue LI ; Han LIN ; Kun ZHANG ; Min ZHOU ; Wencheng WU
Acta Pharmaceutica Sinica B 2025;15(10):5400-5415
Simultaneous management of intestinal mucosal barrier dysfunction and gut microbiota dysregulation represents a significant challenge in the treatment of inflammatory bowel disease (IBD). Herein, we report a novel system that integrates multi-enzyme mimicking cerium single-atom nanocatalysts (CeSACs) with Lactobacillus reuteri probiotics (LR@CeSACs) for multipronged management of IBD. In this system, CeSACs demonstrate robust multi-enzyme activities across a broad pH range, effectively scavenging elevated reactive oxygen species, downregulating pro-inflammatory cytokines, and suppressing the expression of fibrosis-related genes. Moreover, probiotics promote the targeting and retention of the CeSACs for sustained catalytic antioxidant therapy. In turn, the inflammation relief enabled by CeSACs promotes bacterial viability, allowing for the rapid reshaping of intestinal barrier function and the restoration of gut microbiota. Therefore, LR@CeSACs exhibit excellent catalytic anti-inflammatory and anti-fibrotic therapeutic effects, as well as a certain prophylactic effect, as demonstrated in several murine models.


Result Analysis
Print
Save
E-mail