Objective:To investigate the prognostic factors for esophageal cancer after radical resection and the application value of machine learning prediction model.Methods:The retrospective cohort study was conducted. The clinicopatholigical data of 406 esophageal cancer patients who were admitted to Qilu Hospital of Shandong University from January 2018 to March 2022 were collected. There were 357 males and 49 females, aged (64±8)years. All patients underwent radical resection of esophageal cancer. The 406 patients were randomly divided into a training set of 285 cases and a validation set of 121 cases at a 7∶3 ratio based on a random number table. The training set was used to construct prediction model, and the validation set was used to validate prediction model. Patients were divided into high-risk group and low-risk group based on risk scores. Observation indicators: (1) follow-up of patients and analysis of influencing factors for prognosis; (2) construction and validation of machine learning prediction models. Comparison of measurement data with normal distribution between groups was conducted using the independent sample t test. Comparison of measurement data with skewed distribution between groups was conducted using the Mann-Whitney U test. Comparison of count data between groups was conducted using the chi-square test. Comparison of ordinal data between groups was conducted using the rank sum test. The Kaplan-Meier method was used to calculate survival rate and plot survival curve, and the Log-rank test was used for survival analysis. The Cox proportional hazard regression model was used for univariate and multivariate analyses. Independent influencing factors were included, and data processing, machine learning model construction, and visualization were performed using R packages including random survival forest (RSF), gradient boosting machine (GBM), least absolute shrinkage and selection operator Cox regression (LASSO-Cox), Cox proportional hazards model boosting (CoxBoost), survival support vector machine (survivalsvm), extreme gradient boosting (XGBoost), supervised principal component analysis (SuperPC), and Cox partial least squares regression (plsRcox). Receiver operating characteristic (ROC) curves were drawn, and sensitivity, specificity, and area under the curve (AUC) were calculated. The Delong test was used to assess the differences in AUC among different models in the training set, and the time-dependent ROC was used to compare the predictive performance of different models. Calibration curves were used to evaluate model accuracy, and decision curve analysis (DCA) was used to evaluate overall net benefit. Results:(1) Follow-up of patients and analysis of influencing factors for prognosis. All 406 patients were followed up postoperatively for 28(range, 6-36)months, with 1- and 3-year overall survival rate of 86.5% and 40.9%, respectively. The 285 patients in the training set were followed up postoperatively for 30(range, 6-36)months, with 1- and 3-year overall survival rate of 85.1% and 35.5%, respectively. The 121 patients in the validation set were followed up postoperatively for 25(range, 6-36)months, with 1- and 3-year overall survival rate of 87.0% and 43.2%, respectively. There was no significant difference in postoperative overall survival rate between the training set and the validation set ( χ2=3.20, P>0.05). Results of multivariate analysis showed that left thoracic surgical approach, preopera-tive neutrophil count, vascular invasion, perineural invasion, pathological T2-4 stage, pathological N2-3 stage, and postoperative pneumonia were independent risk factors affecting postoperative survival of 285 patients in the training set ( hazard ratio=1.466, 1.037, 1.482, 1.549, 5.268, 7.727, 22.202, 2.539, 2.686, 1.425, 95% confidence interval as 1.026-2.096, 1.003-1.073, 1.008-2.179, 1.105-2.170, 1.201-23.099, 1.833-32.576, 4.734-104.128, 1.577-4.087, 1.631-4.422, 1.018-1.994, P<0.05). (2) Construction and validation of machine learning prediction models. Independent risk factors affecting postoperative survival were included to construct RSF, GBM, LASSO-Cox, CoxBoost, survivalsvm, XGBoost, SuperPC, and plsRcox machine learning prediction models. Results of Delong test showed that there were significant differences in the AUC of RSF and GBM from the other six models ( P<0.05). Results of time-dependent ROC curve showed that all 8 machine learning predic-tion models had good discriminative ability in the training cohort, among which the RSF machine learning prediction model had the best predictive performance. Results of calibration curve showed that the RSF machine learning prediction model fitted well for predicting postoperative 1-, 2-, and 3-year overall survival in the training cohort, with high consistency with actual results. Results of decision curve analysis showed that within a threshold range of 0-0.80, the RSF machine learning prediction model provided a better overall net benefit. Further analysis showed that in the validation set, the AUC of RSF machine learning prediction model for postoperative 1-, 2-, and 3-year survival prediction were 0.786 (95% confidence interval as 0.609-0.962), 0.774 (95% confidence interval as 0.676-0.873), and 0.750 (95% confidence interval as 0.652-0.848), respectively. Results of calibration curve showed that the RSF machine learning prediction model fitted well for predicting postopera-tive 1-, 2-, and 3-year overall survival in the validation set, with high consistency with actual results. In the training set, the optimal cutoff value of the RSF machine learning prediction model risk score was 11.7. Patients with risk score ≥11.7 were classified as the high-risk group, and those with risk score <11.7 as the low-risk group. The median survival times of the two groups were 18.0 months and >36.0 months, respectively, showing a significant difference between them ( χ2=73.30, P<0.05). In the validation set, the optimal cutoff value of the RSF machine learning prediction model risk score was 11.7. Patients with risk score ≥11.7 were classified as the high-risk group, and those with risk score<11.7 as the low-risk group. The median survival times of the two groups were 17.0 months and>36.0 months for the high-risk and low-risk groups, respectively, showing a significant difference between them ( χ2=35.20, P<0.05). Conclusions:Left thoracic surgical approach, preoperative neutrophil count, vascular invasion, perineural invasion, pathological T2-4 stage, pathological N2-3 stage, and postoperative pneumonia are independent risk factors affecting survival of esophageal cancer patients after radical resection. The RSF machine learning prediction model constructed based on these factors can effectively distinguish the survival prognosis of high-risk and low-risk patients.

Accurate characterization of the chemical composition of complex traditional Chinese medicine(TCM)is an essential foundation for the modern scientific interpretation of TCM principles.Mass spectrometry is the most dominant technique in current research on the material basis of TCM,offering the highest sensitivity and the richest information provision.Establishing mass spectrometry databases represents the most effective approach to facilitating the structural analysis of TCM chemical components.This paper systematically searches and reviews literature published from January 2005 to January 2025 through online databases such as China National Knowledge Infrastructure,PubMed,and Web of Science,using"mass spectrometry database"and"traditional Chinese medicine"as keywords.It reviews the current status of seven TCM chemical component mass spectrometry databases and seven natural product mass spectrometry databases.The key advancements of these mass spectrometry databases for natural products are summarized,detailing their characteristics,search methodologies,included information,and data sources.Additionally,challenges related to data quality,standardization,timely updates,database interaction,retrieval functionality,and data sharing and security are discussed in depth.Furthermore,the paper explores prospective development directions for TCM mass spectrometry databases,emphasizing the importance of open data sharing,technological innovation,and data security.Through this analysis,the paper aims to offer theoretical guidance and practical recommendations for the precise identification of TCM components,as well as for the construction and application of these databases.

Objective:To explore the impact of the two-way referral model on compliance and prognosis in patients with heart failure.Methods:This bidirectional cohort study enrolled chronic heart failure (CHF) patients treated at Qilu Hospital of Shandong University or designated primary hospitals between March 2018 and March 2022. Patients were categorized into two groups based on referral status: two-way referral group (participating in the referral model with≥1 follow-up visit at primary hospitals) and the core hospital group (receiving treatment and follow-up exclusively at Qilu Hospital). Baseline clinical characteristics were collected and compared between groups. Patients underwent followed-up, with primary endpoints including follow-up rate, drug (β-blockers, angiotension converting enzyme inhibitor (ACEI)/angiotensin Ⅱ receptor blockers (ARB)/angiotensin receptor-neprilysin inhibitor (ARNI), sodium-glucose cotransporter 2 inhibitors and mineralocorticoid receptor antagonists) utilization rate and target dose achievement rate. Secondary endpoints encompassed changes from baseline in left ventricular ejection fraction (LVEF), left ventricular end-diastolic diameter (LVEDd), and N-terminal pro-brain natriuretic peptide (NT-proBNP), plus cardiovascular mortality and heart failure rehospitalization. Generalized linear mixed models analyzed longitudinal trends in LVEF, LVEDd, and NT-proBNP levels. Kaplan-Meier curves and Cox regression evaluated LVEF recovery rates, supplemented by subgroup analyses. Multivariate logistic regression was used to identify factors influencing target dose achievement rate for β-blockers and ACEI/ARB/ARNI therapies in CHF patients.Results:A total of 357 patients were enrolled, aged 53 (41, 63) years, including 256 males (71.7%). 157 patients were in the two-way referral group and 200 patients in the core hospital-treated group. Compared with the core hospital-treated group, the two-way referral group had lower baseline LVEF (28 (22, 34)% vs. 31 (23, 36)%, P=0.021) and systolic blood pressure (116 (104, 125) mmHg vs. 121 (109, 134) mmHg (1 mmHg=0.133 kPa), P=0.010). The 12-month follow-up rate of the two-way referral group was higher than the core hospital-treated group (73.8% vs. 56.0%, P=0.004). No significant between-group differences were observed in drug utilization rate of β-blockers, ACEI/ARB/ARNI, or sodium-glucose cotransporter 2 inhibitors during follow-up (all P>0.05), while mineralocorticoid receptor antagonists use showed a declining trend in both groups. Although the core hospital-treated group had higher target dose achievement rates for β-blockers (65.4% vs. 49.3%, P=0.042) and ACEI/ARB/ARNI (79.8% vs. 65.8%, P=0.046) than the two-way referral group, multivariate logistic regression indicated that the two-way referral model was not a negative predictor for these outcomes (all P>0.05). Both groups showed improved NT-proBNP, LVEDd, and LVEF from baseline (all P<0.001) with no significant difference in trends between groups (all P>0.05). There was no significant difference in the composite incidence (7.6% vs. 6.5%, P=0.674) and cumulative incidence (log-rank P=0.684) of cardiovascular death and heart failure rehospitalization at 12 months between two groups. Conclusion:The two-way referral model demonstrates advantages in improving medication adherence, drug utilization rates, and targetdoseachievement rates among CHF patients. This model not only promotes cardiac functional recovery but also reduces risks of cardiovascular mortality and heart failure rehospitalization, achieving comparable therapeutic and management outcomes to those observed in core hospital-treated patients.

Accurate characterization of the chemical composition of complex traditional Chinese medicine (TCM) is an essential foundation for the modern scientific interpretation of TCM principles. Mass spectrometry is the most dominant technique in current research on the material basis of TCM, offering the highest sensitivity and the richest information provision. Establishing mass spectrometry databases represents the most effective approach to facilitating the structural analysis of TCM chemical components. This paper systematically searches and reviews literature published from January 2005 to January 2025 through online databases such as China National Knowledge Infrastructure, PubMed, and Web of Science, using “mass spectrometry database” and “traditional Chinese medicine” as keywords. It reviews the current status of seven TCM chemical component mass spectrometry databases and seven natural product mass spectrometry databases. The key advancements of these mass spectrometry databases for natural products are summarized, detailing their characteristics, search methodologies, included information, and data sources. Additionally, challenges related to data quality, standardization, timely updates, database interaction, retrieval functionality, and data sharing and security are discussed in depth. Furthermore, the paper explores prospective development directions for TCM mass spectrometry databases, emphasizing the importance of open data sharing, technological innovation, and data security. Through this analysis, the paper aims to offer theoretical guidance and practical recommendations for the precise identification of TCM components, as well as for the construction and application of these databases.

Objective:To explore the value of deep learning models based on ultrafast dynamic contrast-enhanced MRI (UF-DCE MRI) in predicting malignant breast lesions.Methods:The study was a cross-sectional study. Clinical and imaging data of 347 patients with breast lesions who received treatment at Tianjin Medical University Cancer Institute and Hospital from March 2023 to January 2024 were analyzed retrospectively. A total of 347 lesions were observed in the 347 patients, including 75 benign and 272 malignant lesions. The random number method was used to divide into the training set with 243 cases and the validation set with 104 cases in a ratio of 7∶3. All patients underwent breast UF-DCE MRI and conventional dynamic-enhanced MRI (DCE-MRI). A 27-channel model (27-phase enhancement images of input UF-DCE MRI), a 3-channel model (3-phase enhancement images of input DCE-MRI), and a 1-channel model (1st-phase enhancement images of DCE-MRI) were built based on the pre-trained ResNet18 deep learning model on ImageNet. The efficacy of each model in predicting breast malignant lesions was analyzed using receiver operating characteristic curves and area under the curve (AUC). The differences of AUC were compared using DeLong test.Results:In the training and validation sets, the 27-channel model had the highest AUC for diagnosing malignant breast lesions, which were 0.848 (95% CI 0.818-0.877) and 0.784 (95% CI 0.752-0.817), respectively. DeLong test showed no statistically significant difference in the AUC values of the three models in the validation set for the diagnosis of malignant lesions of the breast in a two-by-two comparison ( P>0.05). UF-DCE MRI scans were 27 phases totaling 81 s with a temporal resolution of 3 s/phase; DCE-MRI scans were 3 phases totaling 270 s with a temporal resolution of 90 s/phase. Conclusions:The model combining UF-DCE MRI with deep learning demonstrates comparable efficacy to DCE-MRI deep learning model in diagnosing breast malignant lesions. However the UF-DCE MRI has the advantages of high temporal resolution and short scanning time, which makes this model valuable for precise diagnosis and treatment of breast cancer.

Objective To explore the application value of Auto-prescription combined with low-dose contrast and adaptive statisti-cal iterative reconstruction-Veo(ASIR-V)algorithm in the computed tomography angiography(CTA)of thoracodorsal artery(TDA).Methods A total of 100 patients who underwent TDA CTA examination were prospectively selected.A tube voltage of 120 kVp and contrast agent of 1.5 mL/kg were used for group A(50 cases),and images were reconstructed with 40％ post-set ASIR-V.The Auto-prescription for tube voltage and contrast agent of 1.2 mL/kg were used for group B(50 cases),while images were reconstruc-ted with 40％,60％,and 80％ post-set ASIR-V,labeled as subgroups B1 to B3.The objective and subjective evaluation results of the images were compared between and within groups.Results Group A had an effective dose(ED)of 2.98(2.65,4.03)mSv,while group B had an ED of 1.92(1.44,3.33)mSv.The iodine intake in group B was lower than that in group A,and the CT value of the axillary artery in group B was significantly higher than that in group A(P＜0.001).With the increased of ASIR-V level in group B,the signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR)of the images gradually increased(P＜0.05).In terms of subjec-tive scores on axial images,both subgroups B2 and B3 were superior to group A(P＜0.001);with the increased of ASIR-V level in group B,subjective scores of axial images increased first and then decreased,among which subjective score of subgroup B2 was the highest and the differences were statistically significant(P＜0.001).In terms of subjective scores on three-dimensional image quality,subgroups B1 to B3 were superior to group A(P＜0.001).Conclusion The use of Auto-prescription combined with low-dose con-trast and 60％ ASIR-V can significantly optimize the display of TDA,and reduce the radiation dose and contrast agent dose to a certain extent.

Objective:To study the relationship between urinary arsenic methylation metabolism patterns and skin keratinization and pigmentation abnormalities in population exposed to arsenic through drinking water.Methods:Using a cross-sectional study method, a survey on endemic arsenic poisoning was conducted among permanent residents of drinking water endemic arsenic poisoning areas in Bayannur City, Inner Mongolia Autonomous Region in 2004 (before water improvement). In 2017 (after water improvement), 71 arsenic exposed individuals were followed up as survey subjects. According to the "Diagnosis of Endemic Arsenism" (WS/T 211-2015), the clinical grading of skin injuries (skin keratinization, pigmentation abnormalities) in the survey subjects was evaluated. Urine samples were collected for detection of arsenic methylation metabolite levels by high-performance liquid chromatography inductively coupled plasma mass spectrometry and calibrated with urinary creatinine. The changes and amplitudes of urinary arsenic methylation indicators before and after water improvement were calculated and analyzed according to the outcome of skin keratinization and pigmentation abnormalities which were divided into reduced, unchanged, and added groups.Results:(1) The changes in urinary total arsenic (TAs), inorganic arsenic (iAs), monomethyl arsenic (MMA), and dimethyl arsenic (DMA) levels in different outcome groups of skin keratinization were compared, and the differences were statistically significant ( H = 9.08, 8.77, 9.28, 8.57, P < 0.05). The changes in urinary TAs, iAs, MMA, DMA levels, iAs percentage (iAs%), DMA percentage (DMA%), and primary methylation index (PMI) in different outcome groups of skin pigmentation abnormalities were compared, and the differences were statistically significant ( H = 8.04, 10.67, 8.29, 9.14, 6.30, 9.10, 7.20, P < 0.05). (2) The comparison of amplitudes in urinary TAs, iAs, MMA, and DMA levels in different outcome groups of skin keratinization showed statistically significant differences ( H = 6.92, 7.34, 6.66, 6.16, P < 0.05). The amplitudes in urinary iAs level, iAs%, DMA%, and PMI in different outcome groups of skin pigmentation abnormalities were compared, and the differences were statistically significant ( H = 7.94, 7.61, 9.95, 7.22, P < 0.05). Conclusion:The changes pattern of urinary TAs, iAs, MMA, DMA, iAs%, DMA%, and PMI in population exposed to arsenic through drinking water is related to the transformation of skin keratinization and pigmentation abnormalities.

Objective:To evaluate the efficacy of curative esophagectomy versus definitive chemoradiotherapy (dCRT) in patients with clinical T1bN0M0 thoracic esophageal cancer.Methods:The propensity score matching (PSM) and retrospective cohort study was conducted. The clinico-pathological data of 163 patients with clinical T1bN0M0 thoracic esophageal cancer who were admitted to Henan Provincial People′s Hospital from January 2014 to December 2020 were collected. There were 125 males and 38 females, aged (58.9±7.0)years. Of 163 patients, 124 cases undergoing curative transthoracic esophagectomy were allocated into the radical resection group, 39 cases undergoing dCRT were allocated into the dCRT group. Observation indicators:(1) PSM and compari-son of clinicopathological characteristics of patients between the two groups after matching; (2) complications in the radical resection group and treatment status in the dCRT group; (3) survival analysis; (4) analysis of factors influencing patients′ prognosis. Comparison of measurement data with normal distribution between groups was conducted using the Welch t test. Comparison of measurement data with skewed distribution between groups was conducted using the Mann-Whitney U test. Comparison of count data between groups was conducted using the chi-square test or Fisher exact probability. Comparison of ordinal data was conducted using the Mann-Whitney U test. The Cox proportional hazard model was used for univariate and multivariate analyses. The Kaplan-Meier method was used to calculate survival rate and plot survival curve, and Log-rank test was used for survival analysis. PSM was performed using the 2∶1 nearest neighbor matching method. The caliper value was set as 0.05. Results:(1) PSM and comparison of clinicopathological charac-teristics of patients between the two groups after matching. Of the 163 patients, 117 cases were successfully matched, with 78 cases in the radical resection group and 39 cases in the dCRT group. After PSM, the elimination of tumor differentiation degree confounding bias ensured comparability. (2) Complications in the radical resection group and treatment status in the dCRT group.Among the 78 patients in the curative esophagectomy group, 22 cases developed complications within 30 days after surgery. There was no death within 30 days after surgery. Among the 39 patients in the dCRT group, 25 cases received concurrent chemoradiotherapy alone, 8 cases received induction chemo-therapy followed by concurrent chemoradiotherapy, 3 cases received sequential chemoradiotherapy, and 3 cases received radiotherapy alone. Among the 33 patients who received concurrent chemo-radiotherapy, 29 cases were treated with the XP regimen, and 4 cases with the FP regimen. Efficacy evaluation showed that 37 patients achieved complete remission, and 2 patients had residual lesions. Twenty-two patients developed treatment-related adverse reactions. (3) Survival analysis. After PSM, the follow-up duration was 58(range, 13-125)months in the radical resection group and 56(range, 10-129)months in the dCRT group. The postoperative 5-year overall survival rates were 95.7% and 97.1% in the radical resection group and dCRT group, respectively, showing no significant difference between the two groups ( χ2=0.001, P>0.05). The postoperative 5-year disease-free progression survival rates were 88.2% and 94.2% in the radical resection group and dCRT group, respectively, showing no significant difference between the two groups ( χ2=0.652, P>0.05). (4) Analysis of factors influencing patients prognosis. Age and pathological TNM stage were indepen-dent influencing factors for overall survival time in patients with clinical T1bN0M0 thoracic esophageal cancer ( hazard ratio=1.312, 2.945, 95% confidence interval as 1.042-1.711, 2.204-5.517, P<0.05). Age and pathological TNM stage were independent influencing factors for disease-free survival time in patients with clinical T1bN0M0 thoracic esophageal cancer ( hazard ratio=1.215, 3.301, 95% confidence interval as 1.012-1.699, 2.012-6.321, P<0.05). Conclusions:There is no significant difference in overall survival and disease-free survival between patients with clinical T1bN0M0 thoracic esophageal cancer undergoing curative esophagectomy and dCRT. The treatment modality is not an independent prognostic factor.

Constipation,a common functional gastrointestinal disorder,not only severely impairs patients'quality of life but is also highly comorbid with psychiatric conditions such as anxiety and depression.Emerging evidence indicates that gut microbiota dysbiosis is a critical link connecting these two disease states.On one hand,dysbiosis exacerbates constipation by affecting host metabolism and intestinal function;on the other,it plays a central role in the pathophysiology of mood disorders.This complex interaction is primarily mediated through the"microbiota-gut-brain axis."Therefore,elucidating the intrinsic relationship among anxiety,depression,gut microbiota,and constipation has become a frontier of interdisciplinary research.

Objective To explore the value of contrast agent injection protocol based on individualized iodine delivery rate(IDR)applicated in enhanced spectral CT scanning of colorectal cancer.Methods Totally 131 colorectal cancer patients were prospectively collected and randomly divided into conventional group(group A,n=64,contrast agents were injected with routine method)and individualized group(group B,n=67,contrast agents were injected with individualized IDR method).Whole abdominal energy spectrum CT scanning was performed,and 70 keV images at arterial and venous phases(A70 keV)of group A and 40-70 keV images(interval 10 keV)at arterial and venous phases(B40-70 keV)of group B were reconstructed,and iodine-based material maps were reconstructed.Subjective and objective evaluation on quality of enhanced images were performed,lesions'normalized iodine concentration(NIC)were calculated,and the above indexes were compared between groups.Results Iodine intake in group B was lower than that in group A(P＜0.001).The subjective scores of B60 keV images at both arterial and venous phases were the highest,which were not different from those of A70 keV images(both P＞0.05).CT values,signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR)of superior mesenteric artery(SMA)and abdominal aorta on arterial B40-60 keV images were higher than those in arterial A70 keV images(all P＜0.05).CT values,SNR and CNR of superior mesenteric vein(SMV)on venous B40-50 keV,as well as CT values and CNR of SMV on venous B60 keV images were all higher than those on venous A70 keV images(all P＜0.05).No significant difference of lesions'NIC in arterial and venous phases was detected between groups(both P＞0.05).Conclusion Injection of contrast agents based on IDR combined with 60 keV single energy image for enhanced energy spectrum CT scanning of colorectal cancer could reduce the amount of contrast agents while do not affect imaging quality.