The World Health Organization estimates that over 7 million people worldwide are infected with Chagas disease, with up to one-third of chronic infections developing cardiac arrhythmias (2025). Efficient diagnosis can help physicians provide an adequate treatment plan, but serological testing is limited. We propose a deep learning model that would test for cardiac alterations associated with Chagas and could be put on a Raspberry Pi, making it more accessible for hospitals. The use of deep learning-based models is a rapidly growing research topic in the medical field for classification and diagnostics. A great application of such software is the detection of Chagas disease in electrocardiograms (ECG), as this is one of the most widely used diagnostic instruments in medicine. Further developments suggest incorporating such models into devices for monitoring and detecting abnormalities in ECG signals. In addition, equipment is not centered around designing for low-cost materials, open-source code, and datasets to ensure better accessibility. In this work, we create a long short-term memory network to tackle this problem using the TensorFlow library on the George B. Moody PhysioNet Challenge's open-source dataset. We present a deep learning-based method for live classification on a Raspberry Pi 4 and an evaluation of our TensorFlow Model and TensorFlowLite FlatBuffers to demonstrate their minimal run-time requirements while maintaining an acceptable accuracy of 98%. By focusing on open-source and portable solutions, our approach offers a scalable method to prioritize testing and treatment for Chagas disease, particularly in resource-limited settings.