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.A systematic review on the integrated application of evidence-based narrative education and undergraduate nursing teaching
Nannan BAI ; Meng LI ; Qian LIANG ; Chou YAO ; Yan WANG ; Ju HAN ; Chenyang HOU ; Nana XING
Chinese Medical Ethics 2026;39(2):229-237
ObjectiveTo systematically evaluate the application of narrative education in undergraduate nursing teaching, to understand the current application status of narrative education, and to provide a theoretical basis for the subsequent establishment of a sound narrative education system. MethodsA systematic search was conducted for studies published in Chinese and English databases on applying narrative education to undergraduate nursing teaching, with the search period ranging from database inception to February 23, 2025. Literature was screened, and relevant information was extracted. A rigorous quality evaluation was conducted on the included studies, and a descriptive analysis was performed on their content. ResultsA total of 20 papers were included, involving 3,180 research subjects, all of whom were undergraduate nursing students. The results of descriptive analysis showed that the teaching model of narrative education primarily encompassed reading narrative works, watching films and videos, performing narrative scenarios, and writing reflective journals. The course setting and content covered pre-teaching preparation and in-teaching implementation. The evaluation of teaching effectiveness included the evaluation of teachers’ teaching methods (student evaluation/self-evaluation) and the evaluation of students’ learning effectiveness (course grade evaluation/humanistic care scale/empathy scale assessment, and others). ConclusionNarrative education combines abstract concepts with concrete clinical situations, which not only enriches students’ learning experiences but also enhances their humanistic literacy. Meanwhile, it provides teachers with opportunities to develop their narrative teaching skills, which requires them to possess profound professional knowledge and employ narrative techniques to guide students in reflection and critical thinking, thereby improving teaching quality and learning outcomes. Future efforts should consistently deepen the connotation research of narrative education and build a systematic nursing education system.
3.Establishment of a Gastrointestinal-Brain Inter-Organ Multimodal Characterization System Based on Traditional Chinese Medicine Theory and Its Application in Refractory Diseases
Guanghui HAN ; Yan GUO ; Peijing RONG ; Bin CONG ; Shuangjiang LIU ; Shaoyuan LI ; Wei WEI
Journal of Traditional Chinese Medicine 2025;66(6):561-568
The concept of holism is the core idea of traditional Chinese medicine (TCM). Various organs and tissues coordinate with each other to maintain the body's life activities, with a close and mutual influence between the spleen, stomach, and the central nervous system (brain). The gut-brain axis plays an important bridging role between the digestive system and the central nervous system, achieving bidirectional information exchange between the brain and the gastrointestinal tract through complex neuroendocrine and immune mechanisms. The theory of cross-organ interaction involves the mutual influence, coordination, and integration between different organs and systems; multimodality, on the other hand, utilizes multiple sensory modalities, such as vision, hearing, and touch, to convey information. By combining TCM theory with the gut-brain axis theory, a cross-organ multimodal characterization system is established to explore its mechanism and application value in refractory diseases such as functional gastrointestinal disorders, precancerous gastrointestinal diseases, Alzheimer's disease, Parkinson's syndrome, type 2 diabetes, and depression.
4.Discriminating Tumor Deposits From Metastatic Lymph Nodes in Rectal Cancer: A Pilot Study Utilizing Dynamic Contrast-Enhanced MRI
Xue-han WU ; Yu-tao QUE ; Xin-yue YANG ; Zi-qiang WEN ; Yu-ru MA ; Zhi-wen ZHANG ; Quan-meng LIU ; Wen-jie FAN ; Li DING ; Yue-jiao LANG ; Yun-zhu WU ; Jian-peng YUAN ; Shen-ping YU ; Yi-yan LIU ; Yan CHEN
Korean Journal of Radiology 2025;26(5):400-410
Objective:
To evaluate the feasibility of dynamic contrast-enhanced MRI (DCE-MRI) in differentiating tumor deposits (TDs) from metastatic lymph nodes (MLNs) in rectal cancer.
Materials and Methods:
A retrospective analysis was conducted on 70 patients with rectal cancer, including 168 lesions (70 TDs and 98 MLNs confirmed by histopathology), who underwent pretreatment MRI and subsequent surgery between March 2019 and December 2022. The morphological characteristics of TDs and MLNs, along with quantitative parameters derived from DCE-MRI (K trans , kep, and v e) and DWI (ADCmin, ADCmax, and ADCmean), were analyzed and compared between the two groups.Multivariable binary logistic regression and receiver operating characteristic (ROC) curve analyses were performed to assess the diagnostic performance of significant individual quantitative parameters and combined parameters in distinguishing TDs from MLNs.
Results:
All morphological features, including size, shape, border, and signal intensity, as well as all DCE-MRI parameters showed significant differences between TDs and MLNs (all P < 0.05). However, ADC values did not demonstrate significant differences (all P > 0.05). Among the single quantitative parameters, v e had the highest diagnostic accuracy, with an area under the ROC curve (AUC) of 0.772 for distinguishing TDs from MLNs. A multivariable logistic regression model incorporating short axis, border, v e, and ADC mean improved diagnostic performance, achieving an AUC of 0.833 (P = 0.027).
Conclusion
The combination of morphological features, DCE-MRI parameters, and ADC values can effectively aid in the preoperative differentiation of TDs from MLNs in rectal cancer.
6.Correlation Between Human Body Fat Distribution and Carotid Atherosclerosis
Yanmeng QI ; Dandan LI ; Anqi CHEN ; Han YANG ; Yan LI
Medical Journal of Peking Union Medical College Hospital 2025;16(3):703-709
To explore the correlation between human body fat distribution and carotid athero-sclerosis. Adult healthy individuals who underwent physical examinations at the Health Management Center of the International Medical Department(Xidan Campus) of Peking Union Medical College Hospital between January 2020 and January 2023 were selected as study participants. Baseline data, laboratory indicators, Inbody body composition analysis, and carotid ultrasound examination results were collected. Participants were divided into carotid atherosclerosis group and control group based on carotid ultrasound examination results. Baseline data, laboratory indicators, and body composition differences between the two groups were compared, and multivariate Logistic regression analysis was conducted to investigate the correlation between carotid atherosclerosis and body composition. To further explore the correlation between fat distribution and carotid atherosclerosis, the ratio of trunk fat to limb fat(trunk fat/limb fat), visceral fat area, and waist-to-hip ratio were stratified into Q1, Q2, Q3, and Q4 groups(with Q1 group as the reference) based on quartiles(Q1 to Q4), and a forest plot was constructed to analyze the risk of carotid atherosclerosis in each subgroup. A total of 2221 participants were included in this study, including 1194 males(53.76%) and 1027 females(46.24%), with a mean age of(50.28±10.93) years. There were 1049 cases in the carotid atherosclerosis group and 1172 cases in the control group. Baseline data analysis showed that the carotid atherosclerosis group had a higher age and male proportion compared to the control group(both There is a significant correlation between trunk fat/limb fat and the occurrence of carotid atherosclerosis. In the future, the sample size should be expanded, and more precise methods for measuring fat distribution should be adopted to validate the conclusions of this study.
7.Study on toxicity-reducing and efficacy-enhancing effects of Polygala tenuifolia compatibility on sand-ironing Strychnos nux-vomica
Yi SUI ; Guo FENG ; Gang LIU ; Keyan LIU ; Xuehao WEI ; Minggang TENG ; Wei LI ; Caiyao HAN ; Yan LEI
China Pharmacy 2025;36(10):1197-1201
OBJECTIVE To explore the effects of Polygala tenuifolia compatibility on toxicity, anti-inflammatory and analgesic efficacy of sand-ironing Strychnos nux-vomica (SS). METHODS The preparation of SS single decoction, SS-P. tenuifolia core-removed (PC) (1∶2.5) or (1∶5) combined decoction, and SS-PC (1∶5) mixture were carried out to investigate their median lethal dose (LD50). Using aspirin as positive control, the number of writhing movements, analgesic rate, pain latency, ear swelling degree and inflammation inhibition rate induced by the above-mentioned medicinal liquids in mice were compared. The contents of the active and toxic components, strychnine and brucine, in the above-mentioned medicinal liquids were also determined. RESULTS The LD50 values of SS single decoction, SS-PC (1∶2.5) combined decoction, SS-PC (1∶5) combined decoction and SS- PC (1∶5) mixture were 302.00, 614.47, 1 445.44 and 1 778.28 mg/kg, respectively. Compared with control group, the number of writhing movements and ear swelling degree in the mice of the above-mentioned medicinal liquid groups were reduced or decreased significantly (P<0.05 or P<0.01); pain latency [at 90 and 120 minutes in the SS single decoction group, at 60 and 90 minutes in the SS-PC (1∶2.5) combined decoction group, and at 60,90, 120 minutes in the SS-PC (1∶5) combined decoction group and SS-PC (1∶5) mixture group] was significantly prolonged (P<0.05 or P<0.01); analgesic rates of the respective medicinal liquids were 39.30%, 70.87%, 80.00% and 82.46%, and inflammation inhibition rates were 38.08%,TD 57.89%, 76.47% and 50.46%; analgesic and anti-inflammatory effects of combined decoction and mixture were generally better than those of the single decoction (P<0.05 or P<0.01). In the above-mentioned four medicinal liquids, the total contents of strychnine were 0.71%, 0.42%, 0.47% and 0.64%, and the total contents of brucine were 0.88%, 0.63%, 0.57% and 0.88%, respectively. CONCLUSIONS The combination of P. tenuifolia can reduce the toxicity of SS and enhance its anti-inflammatory and analgesic effects. Moreover, there is a tendency for the toxicity-reducing and efficacy-enhancing effects to increase with the increasing dosage of P. tenuifolia. Additionally, the combined decoction of SS and P. tenuifolia can reduce the contents of the active and toxic components, strychnine and brucine, in SS.
8.Discriminating Tumor Deposits From Metastatic Lymph Nodes in Rectal Cancer: A Pilot Study Utilizing Dynamic Contrast-Enhanced MRI
Xue-han WU ; Yu-tao QUE ; Xin-yue YANG ; Zi-qiang WEN ; Yu-ru MA ; Zhi-wen ZHANG ; Quan-meng LIU ; Wen-jie FAN ; Li DING ; Yue-jiao LANG ; Yun-zhu WU ; Jian-peng YUAN ; Shen-ping YU ; Yi-yan LIU ; Yan CHEN
Korean Journal of Radiology 2025;26(5):400-410
Objective:
To evaluate the feasibility of dynamic contrast-enhanced MRI (DCE-MRI) in differentiating tumor deposits (TDs) from metastatic lymph nodes (MLNs) in rectal cancer.
Materials and Methods:
A retrospective analysis was conducted on 70 patients with rectal cancer, including 168 lesions (70 TDs and 98 MLNs confirmed by histopathology), who underwent pretreatment MRI and subsequent surgery between March 2019 and December 2022. The morphological characteristics of TDs and MLNs, along with quantitative parameters derived from DCE-MRI (K trans , kep, and v e) and DWI (ADCmin, ADCmax, and ADCmean), were analyzed and compared between the two groups.Multivariable binary logistic regression and receiver operating characteristic (ROC) curve analyses were performed to assess the diagnostic performance of significant individual quantitative parameters and combined parameters in distinguishing TDs from MLNs.
Results:
All morphological features, including size, shape, border, and signal intensity, as well as all DCE-MRI parameters showed significant differences between TDs and MLNs (all P < 0.05). However, ADC values did not demonstrate significant differences (all P > 0.05). Among the single quantitative parameters, v e had the highest diagnostic accuracy, with an area under the ROC curve (AUC) of 0.772 for distinguishing TDs from MLNs. A multivariable logistic regression model incorporating short axis, border, v e, and ADC mean improved diagnostic performance, achieving an AUC of 0.833 (P = 0.027).
Conclusion
The combination of morphological features, DCE-MRI parameters, and ADC values can effectively aid in the preoperative differentiation of TDs from MLNs in rectal cancer.
9.Update on the treatment navigation for functional cure of chronic hepatitis B: Expert consensus 2.0
Di WU ; Jia-Horng KAO ; Teerha PIRATVISUTH ; Xiaojing WANG ; Patrick T.F. KENNEDY ; Motoyuki OTSUKA ; Sang Hoon AHN ; Yasuhito TANAKA ; Guiqiang WANG ; Zhenghong YUAN ; Wenhui LI ; Young-Suk LIM ; Junqi NIU ; Fengmin LU ; Wenhong ZHANG ; Zhiliang GAO ; Apichat KAEWDECH ; Meifang HAN ; Weiming YAN ; Hong REN ; Peng HU ; Sainan SHU ; Paul Yien KWO ; Fu-sheng WANG ; Man-Fung YUEN ; Qin NING
Clinical and Molecular Hepatology 2025;31(Suppl):S134-S164
As new evidence emerges, treatment strategies toward the functional cure of chronic hepatitis B are evolving. In 2019, a panel of national hepatologists published a Consensus Statement on the functional cure of chronic hepatitis B. Currently, an international group of hepatologists has been assembled to evaluate research since the publication of the original consensus, and to collaboratively develop the updated statements. The 2.0 Consensus was aimed to update the original consensus with the latest available studies, and provide a comprehensive overview of the current relevant scientific literatures regarding functional cure of hepatitis B, with a particular focus on issues that are not yet fully clarified. These cover the definition of functional cure of hepatitis B, its mechanisms and barriers, the effective strategies and treatment roadmap to achieve this endpoint, in particular new surrogate biomarkers used to measure efficacy or to predict response, and the appropriate approach to pursuing a functional cure in special populations, the development of emerging antivirals and immunomodulators with potential for curing hepatitis B. The statements are primarily intended to offer international guidance for clinicians in their practice to enhance the functional cure rate of chronic hepatitis B.

Result Analysis
Print
Save
E-mail