The study developed an artificial intelligence model based on a convolutional neural network with a channel attention mechanism for patients with acute chest pain in the emergency department. It included 1188 patients from Xiangya Central South University Hospital in March 2024, with clinical diagnoses from 12-lead ECG. The model achieved excellent ability to discriminate between STEMI (AUC 0.986) and NSTEMI (AUC 0.916), with high accuracy, precision, facility, F1-score and inference time of 0.24 ± 0.08 s (p < 0.001), shorter than manual interpretation by cardiologists. For unstable angina and aortic dissection, performance was suboptimal with high sensitivity but low accuracy. The model enables rapid and reliable detection of STEMI and NSTEMI to support clinical decision-making in the emergency room. Recognition of unstable angina and aortic dissection is limited by non-specific electrophysiological signs.