Millions of people worldwide suffer with asthma, a chronic respiratory disease marked by inflammation and blockage of the airways. The goal of this research is to use patient health data to create a deep learning model for diagnosing asthma. Exploratory analysis was carried out to investigate associations between characteristics such as age and gender and the diagnosis of asthma after cleaning and converting categorical variables such as gender, diagnosis, and smoking status. In order to rectify the imbalance between classes, the data was randomly oversampled and divided into 70% training and 30% testing. Early stopping was used to avoid overfitting when training a deep neural network with three fully connected layers using the RMSprop optimizer and binary cross-entropy loss. The model achieved 98% accuracy, with strong precision and recall, indicating its suitability for clinical use. The trained model was saved in .h5 format for future applications.