The objective of this study is to identify directions for future ways of combining Machine Learning methods and technical indicators to accurately predict stock actions and prices. Combining Long Short-Term Memory (LSTM) with technical indicators such as Moving Average Convergence/Divergence (MACD), Relative Strength Index (RSI) and Bollinger Bands enhances accuracy, providing valuable insights for investment and risk management. A systematic literature review has been implemented to place relevant articles from the past three years and categorize the studies having similar context. The reviews indicate that deep learning models perform better than traditional ML methods in terms of prediction, and the inclusion of additional input features improves their performance further. In the study, univariate and multivariate LSTM models were implemented on NIFTY 50 stocks where the multivariate LSTM model gave an overall lower value of Mean Absolute Percentage Error (MAPE) which was 0.61. The multivariate LSTM along with RSI as one of its features provided better results than existing models. The final segment offers comprehensive conclusions and outlines avenues for future research. Our findings suggest LSTM has outstanding potential in improving stock price prediction.