Transfer Learning (TL) is a method that uses convolutional neural networks to improve performance on a new task by leveraging previously acquired knowledge from related tasks. It has been particularly useful in medical image analysis, where data scarcity can be an issue. TL has helped save time and resources. However, most of the research conducted on this topic has arbitrarily applied transfer learning. This review paper aims to provide clear guidelines on how to select appropriate models and TL methods for medical image categorization. This comprehensive review thoroughly examines the current status of research on the use of deep-transfer learning algorithms for parasite imaging. We obtained 300 research papers from four databases, such as ACM, PubMed, Springer, and Web of Science. We followed the retrieve and filter process for paper selection, and 247 studies were found to be relevant to the scope of our review. Our review focused on papers related to selecting backbone models and transfer learning techniques such as feature extractor, feature, fine-tuning, and other attributes from scratch. The review covers a range of topics related to parasite imaging, including the parasites studied, the imaging methods used, the techniques for transfer learning, the datasets employed, the performance metrics analyzed, and the challenges faced. Additionally, we identify trends, advancements, and future possibilities in the area of TL parasite imaging. By conducting this study, we aim to provide insights into the effectiveness, limitations, and potential of TL techniques in parasite identification. Ultimately, our goal is to contribute to the enhancement of diagnostic technologies, leading to better healthcare outcomes for patients with parasitic disorders.