Implementation of a Deep Learning Model Using Teachable Machine for Early Pneumonia Detection from X-Ray Images

Abstract

Pneumonia is one of the leading infectious lung diseases that continues to cause high morbidity and mortality rates, particularly among children and the elderly. Early detection is crucial to prevent severe complications; however, the limited availability of radiologists in Indonesia poses a significant challenge. This study implements a deep learning model using Teachable Machine to detect pneumonia from chest X-ray images. The dataset was obtained from an open-source repository on Kaggle, consisting of 1,341 normal lung images and 3,875 pneumonia lung images. The training process was carried out with 50 epochs, a learning rate of 0.001, and a batch size of 16. The experimental results demonstrated that the model achieved 100% accuracy, precision, recall, and F1-score in detecting pneumonia. These findings indicate that Teachable Machine can serve as an effective solution for developing early pneumonia detection models without requiring advanced programming skills. Nevertheless, the relatively small dataset size may lead to potential overfitting, highlighting the need for further research with larger and more diverse datasets. The contribution of this study lies in providing an alternative approach to artificial intelligence implementation that is simple, fast, and cost-effective, particularly for healthcare facilities in resource-limited regions.