Background Acute non-traumatic chest pain is one of the common complaints in the emergency department and is closely associated with fatal disease. Triage assessment urgently requires the use of simple, rapid tools to screen patients with chest pain for high-risk condition to improve patient outcomes. Methods After data preprocessing and feature selection, univariate and multiple logistic regression analyses were performed to identify potential predictors associated with acute non-traumatic chest pain. A nomogram was built based on the predictors, and an internal evaluation was performed using bootstrap resampling methods. The model was also externally validated in this center. Furthermore, the model results were risk-stratified using the decision tree analysis to explore the corresponding triage level. Subsequently, we developed an online visualization tool based on the model to assess the risk of high risk in patients with chest pain. Results Multiple logistic regression analysis showed that age, smoking, coronary heart disease, hypertension, diabetes, hyperlipidemia, pain site, concomitant symptoms, and electrocardiograph, all of which are independent predictors of high-risk chest pain patients. The AUC of our model in the development and validation groups was 0.919 (95%CI: 0.891 similar to 0.974) and 0.904 (95%CI: 0.855 similar to 0.952). Moreover, our model demonstrated better outcomes in terms of accuracy/sensitivity in both cohorts (81.9%/85.2% and 94.8%/78.5%). The calibration curve shows a high degree of agreement between the predicted and actual probabilities. Decision curve analysis clarified that our model had higher net gains across the entire range of clinical thresholds. Afterward, we developed an online tool, which is used in the triage link to facilitate nurses to screen people with high-risk chest pain. Conclusion We proposed an accurate model to predict the high-risk populations with chest pain, based on which a simple and rapid online tool was developed and provided substantial support for its application as a decision-making tool for the emergency department.