Background and Aim Severe acute pancreatitis (SAP) in patients progresses rapidly and can cause multiple organ failures associated with high mortality. We aimed to train a machine learning (ML) model and establish a nomogram that could identify SAP, early in the course of acute pancreatitis (AP). Methods In this retrospective study, 631 patients with AP were enrolled in the training cohort. For predicting SAP early, five supervised ML models were employed, such as random forest (RF), K‐nearest neighbors (KNN), and naive Bayes (NB), which were evaluated by accuracy (ACC) and the areas under the receiver operating characteristic curve (AUC). The nomogram was established, and the predictive ability was assessed by the calibration curve and AUC. They were externally validated by an independent cohort of 109 patients with AP. Results In the training cohort, the AUC of RF, KNN, and NB models were 0.969, 0.954, and 0.951, respectively, while the AUC of the Bedside Index for Severity in Acute Pancreatitis (BISAP), Ranson and Glasgow scores were only 0.796, 0.847, and 0.837, respectively. In the validation cohort, the RF model also showed the highest AUC, which was 0.961. The AUC for the nomogram was 0.888 and 0.955 in the training and validation cohort, respectively. Conclusions Our findings suggested that the RF model exhibited the best predictive performance, and the nomogram provided a visual scoring model for clinical practice. Our models may serve as practical tools for facilitating personalized treatment options and improving clinical outcomes through pre‐treatment stratification of patients with AP.

Background The prognostic value of cytokeratin 19 fragment (CYFRA 21 − 1) and Ki67 in advanced non-small cell lung cancer (NSCLC) patients with wild-type epidermal growth factor receptor (EGFR) remains to be explored. Methods In this study, 983 primary NSCLC patients from January 2016 to December 2019 were retrospectively reviewed. Finally, 117 advanced NSCLC patients with wild-type EGFR and 37 patients with EGFR mutation were included and prognostic value of CYFRA 21 − 1 and Ki67 were also identified. Results The patients age, smoking history and the Eastern Corporative Oncology Group (ECOG) performance scores were significantly different between CYFRA21-1 positive and negative groups (p < 0.05), while no significant differences were found in Ki67 high and low groups. The results of over survival (OS) demonstrated that patients with CYFRA21-1 positive had markedly shorter survival time than CYFRA21-1 negative (p < 0.001, For whole cohorts; p = 0.002, For wild-type EGFR). Besides, patients with wild-type EGFR also had shorter survival times than Ki67 high group. Moreover, In CYFRA 21 − 1 positive group, patients with Ki67 high had obviously shorter survival time compared to patients with Ki67 low (median: 24vs23.5 months; p = 0.048). However, Ki67 could not be used as an adverse risk factor for patients with EGFR mutation. Multivariate cox analysis showed that age (HR, 1.031; 95%CI, 1.003 ~ 1.006; p = 0.028), Histopathology (HR, 1.760; 95%CI,1.152 ~ 2.690; p = 0.009), CYFRA 21 − 1 (HR, 2.304; 95%CI,1.224 ~ 4.335; p = 0.01) and Ki67 (HR, 2.130; 95%CI,1.242 ~ 3.652; p = 0.006) served as independent prognostic risk factor for advanced NSCLC patients. Conclusions Our finding indicated that CYFRA 21 − 1 was an independent prognostic factor for advanced NSCLC patients and Ki67 status could be a risk stratification marker for CYFRA 21 − 1 positive NSCLC patients with wild-type EGFR.

Background: Immune thrombocytopenia (ITP) is characterized by non-chronic (transient, &lt;12 months) and chronic (≥12 months) decline in the number of platelets. Herpes virus infections have been shown, in many studies, to be associated with the development of ITP. However, it remains unclear whether the herpes virus infection status is associated with the chronic ITP.Methods: We reviewed 480 primary pediatric patients with ITP in the period from January 2017 to December 2019. The prevalence of herpes virus antibodies including the Cytomegalovirus (CMV), Herpes simplex virus 1 (HSV-1), Herpes simplex virus 2 (HSV-2), and Epstein Barr virus were recorded. The levels of serum complement C3 and C4, T (CD3+, CD4+, CD8+), B (CD19+) lymphocytes, and natural killer (CD16+ 56+) cells were also analyzed. Multivariate analysis was used to evaluate the associations between chronic ITP and herpes virus infection status.Results: Compared with non-chronic, patients with chronic ITP had older age (≥3 years), lower levels of hemoglobin and complement C3, and lower probability of CMV and HSV-2 infections (IgM positive; p &lt; 0.05). Patients with herpes virus infection had lower serum platelet counts (p &lt; 0.001), lower complement C3 levels and lower CD4+/CD8+ cells ratio (p &lt; 0.05). Furthermore, platelet counts were positively correlated with CD4+/CD8+ cells ratios (r = 0.519; p = 0.0078), and negatively correlated with T cells (CD3+: r = −0.458, p = 0.0213; CD8+: r = −0.489, p = 0.0131). Multivariate analysis showed that age (OR, 1.644; 95%CI, 1.007–2.684; p = 0.047) was an adverse risk factor for chronic ITP and CMV IgM positive (OR, 0.241; 95%CI, 0.072–0.814; p = 0.022) had lower risk of chronic ITP development, while other herpes virus infection statuses and clinical features were not.Conclusion: Although herpes virus infections were associated with the onset of ITP, our findings indicated that herpes virus infection status might not be a risk factor for chronic ITP.