en 4-Biomedical Signal & Image Processing Biomedical Signal & Image Processing Review Paper Review Paper <span style="font-size:10pt"><span new="" roman="" style="font-family:" times=""><span style="color:black">In recent years, many studies have used deep learning (DL) methods to estimate ejection fraction (EF) from two-dimensional echocardiogram data. DL methods offer several advantages. They automate EF estimation, overcome the limitations of manual or semi-automatic methods, and improve estimation accuracy and computational speed. Despite significant advances in DL-based EF estimation methods, the number of review papers examining their pipelines and performances is limited. In the current study, we categorized the DL-based EF estimation methods. We also reviewed their pipelines, architectures, performances, advantages, and limitations. Moreover, we discussed the current status, research trends, challenges, and future direction of this field. Finally, we addressed several open challenges for future studies. The results revealed that the field is mature yet evolving. DL-based methods have shown great potential for estimating EF. Additionally, many recent studies have focused on the clinical integration and evaluation of these methods. </span></span></span><br> <br> &nbsp; Artificial Intelligence, Deep Learning, Echocardiography, Ejection Fraction, Machine Learning, Review. 4394 4394 http://ijeee.iust.ac.ir/browse.php?a_code=A-10-205-5&slc_lang=en&sid=1 Roshanak Ghorbani roshanak_ghorbani@elec.iust.ac.ir No Department of Biomedical Engineering, School of Electrical Engineering, Iran University of Science and Technology, Tehran, Iran Hamid Behnam behnam@iust.ac.ir Yes Department of Biomedical Engineering, School of Electrical Engineering, Iran University of Science and Technology, Tehran, Iran