With the rapid development of deep learning approaches, tremendous progress has been made in computer- assisted analysis of minimally-invasive, videoscopic surgery. However, surgery through open incisions ("open surgery"), which constitutes a much larger portion of surgical procedures performed, is rarely investigated because of the difficulty in obtaining high-quality open surgical video footage. Automated detection of surgical instruments shows promise for evaluating surgical activities, and provides a foundation for quality/safety review, education, and identification of surgical performance. In this paper, we present results using YOLOv3 to successfully identify an electrocautery surgical instrument in a library of images derived from 22 open neck procedures (an 887-image training/validation set, and a 1149-image testing set) captured using a wearable surgical camera. We show that our method effectively detects the spatial bounds of the electrocautery pencil in still images and we further demonstrate the ability of our method to detect the location of this instrument in video footage. Our work serves as the first demonstration of open surgical instrument detection using first-person video footage from a wearable camera and sets the stage for further work in this field.Clinical Relevance- Detection of instrumentation in surgical video is the necessary first step towards automating surgical task identification and skills assessment, which will be useful for surgical quality improvement and training.