This study explores the emergence of deep learning approach in predicting ionospheric total electron content (TEC); which is pivotal for space weather forecasting. Total electron content variability has profound implications for satellite communications and navigation systems, especially in regions with unique ionospheric characteristics such as equatorial Kenya. Traditional TEC prediction models which are rooted in empirical or physics-based methods, often encounter challenges in capturing the complex, non-linear behaviors inherent in equatorial ionospheric dynamics. A systematic literature review (SLR) was performed to extract and synthesize the algorithms and features that have been used in long short-term memory networks (LSTMs) techniques to model and predict the ionospheric TEC, with a focus on the unique characteristics of the geomagnetic equator at low latitudes. Several articles on ionospheric TEC prediction using deep learning were obtained from research databases where few were selected based on the inclusion/exclusion criteria used in the study. The outcome of the review from several studies embarked on deep learning using LSTMs architectures, they highlighted the importance of feature selection and feature engineering in enhancing prediction accuracy. The study also explored hybrid machine learning techniques models which showed improved forecast performance. Additionally, the suggestion on addressing data gaps and considering additional parameters which could further enhance the accuracy and reliability of TEC predictions.