As the world advances towards decarbonization, hydrogen has emerged as a promising alternative to fossil fuels in various applications. However, the high cost of materials used in hydrogen production, storage, and transportation remains a significant barrier to widespread adoption. Such expensive materials are primarily employed to counteract a phenomenon known as hydrogen embrittlement (HE), which is often responsible for degrading the strength properties of materials storing it, leading to limited storage capacity and safety issues. Meanwhile, low-alloy steels are considered good candidates as the base material for their storing capacity at a moderate cost. Unfortunately, low-alloy steels are also subjected to hydrogen embrittlement. The material’s fatigue crack growth (FCG) rate is an important parameter considered during design. When comparing the FCG rate in hydrogen and air environments, low-alloy steels of tensile strength higher than 900 MPa display a significant acceleration of FCG rate, rapidly increasing along with the tensile strength. For application, fatigue testing in a hydrogen environment on materials considered is required, which is costly and time-consuming. However, predicting the FCG rate prior to testing would allow for more efficient and targeted testing, saving significant time and budget for laboratories. In this study, we aim to predict materials’ FCG rate in air and hydrogen environments from their heat treatment conditions and mechanical properties to identify optimal low-alloy steels with their associated process protocol in a minimum number of experiments. In addition to data analysis, image analysis using scanning electron microscopy (SEM) and electron backscatter diffraction (EBSD) is integrated to enhance the prediction accuracy. Convolutional Neural Networks (CNNs) are employed for image analysis, while Random Forest is used for data interpretation. Further image analysis is also conducted independently to identify microstructural patterns and generate heatmaps, providing deeper insights into the FCG resistance of low-alloy steels in hydrogen environments. This research aims to improve the comprehension of hydrogen embrittlement on low-alloy steels by bridging the gap between material science and machine learning. By integrating these two fields, we aim to create new high-strength materials that are compatible with high-pressure hydrogen environments, opening up new possibilities and potential in the field of materials engineering.