Neglected diseases remain a significant threat to global health, particularly in low income countries, where limited resources and funding hinder drug discovery and treatment efforts. These diseases, which disproportionately affect impoverished populations, often receive little attention from the pharmaceutical industry due to their lack of profitability. Consequently, there is an urgent need to develop innovative and cost-effective approaches to accelerate the discovery of therapeutic solutions. Recent advances in computational science, particularly deep learning, have demonstrated remarkable potential in expediting drug discovery by predicting chemical activity and identifying new drug candidates efficiently. This study aims to leverage cutting-edge deep learning techniques to develop models capable of accurately predicting chemical activity and uncovering novel therapeutic options for neglected diseases such as Chagas disease. The research focuses on exploring various deep learning architectures, including convolutional neural networks (CNNs) and recurrent neural networks (RNNs), as well as incorporating advanced data pre-processing techniques to improve model accuracy and performance. Large and complex datasets containing drug activity information for neglected diseases will be utilized to train and evaluate the proposed models. The study will assess the impact of model choice, computational resources, and data quality on the speed and precision of drug candidate identification. By optimizing these factors, the research aims to create a framework for faster and more cost-effective drug discovery processes. The outcomes of this study have the potential to significantly enhance the pipeline for identifying treatments for neglected diseases, reducing the time and cost associated with drug development. Ultimately, this research could contribute to alleviating the suffering of millions of people affected by these overlooked diseases, offering hope for better health outcomes in underserved populations.