Automatic extraction of relevant and reliable information from electrocardiogram (ECG) signals is essential for heart disease diag- nosis and treatment. This study proposes deep learning model based on improved one-dimensional convolutional neural network (1D-CNN) architecture for classifying heart disease using ECG data. First, we col- lect ECG recordings from patients with and without heart disease. Then, the relevant features are extracted from the ECG data, which is a critical step as the features’ quality directly impacts the predictive models’ per- formance. Next, we apply the predictive models, encompassing 1D-CNN, Support Vector Machine (SVM), and Logistic Regression, combined with fine-tuned hyperparameters and StandardScaler, to improve heart dis- ease prediction performance. The experimental results show that the proposed deep learning model using 1D-CNN combined with fine-tuning hyperparameters and StandardScaler can achieve better classification results on ECG-based heart disease classification tasks than previous studies.