YOLO for Medical Object Detection (2018–2024)
Akanksha Soni, Avinash Rai
Abstract
The You Only Look Once (YOLO) algorithm is a widely-utilized object detection approach, extensively employed in diverse medical object detection tasks. Specifically, peer-reviewed articles published between 2018 and 2024 were chosen through a thorough and methodical search of the PubMed database. 80 studies were found from this search, all of which used the YOLO algorithm for various medical object detection tasks including the localization and identification of lung nodules, breast cancer, tuberculosis, pneumonia, Covid-19, and brain tumor. The results indicated YOLO's efficacy in these tasks, showcasing its superior performance compared to other existing methods. Nevertheless, the review also pinpointed certain limitations of YOLO, such as the requirement for extensive and well-balanced datasets, coupled with its high computational demands. In conclusion, the review not only highlights these aspects but also outlines research gaps and suggests future directions for the application of YOLO in medical object detection.