Letter to the Editor Surgical Innovation 2020, Vol. 0(0) 1–2 © The Author(s) 2020 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/1553350620956425 journals.sagepub.com/home/sri Analysis of Computer Vision Methods for Counting Surgical Instruments Gustavo Chavez, BA 1, , David Y. Zhao, MS 2, , Albert Haque, MS 2, , Rahim Nazerali, MD 3 , and Derek F. Amanatullah, MD, PhD 1 Dear Editor, The surgical count is the primary method to account for and manage surgical instruments, needles, and sponges during operative procedures. 1 However, a miscount, when there is a discrepancy between counted and deployed in- struments, is estimated to occur in roughly 1 in 140 cases. 1 These events often result in significant costs to hospitals and patients per year due to lost operating room (OR) time and secondary imaging procedures. 1 An estimated 1 in 70 miscounts result in a retained surgical instrument, resulting in patient harm, as well costly reoperation and litigation. 1 Given the high cost of OR time and the current burden of counting procedures, it is clear that a more accurate and less labor-intensive counting system is needed. In recent years, there is tremendous growth in machine learning technology applied to healthcare. Computer vi- sion techniques are now applied across many medical domains and are most visible in the context of minimally invasive surgery and endoscopic surgery. We conducted a literature search of computer vision studies on the de- tection or localization of surgical instruments outside of the surgical field. We are highlighting 4 studies that provide insight into both the feasibility and challenges of utilizing existing computer vision techniques to build a system that can perform the surgical count. The 4 studies, summarized in Table 1, implemented a wide range of computer vision techniques to localize different types of surgical items, with relatively high detection accuracies ranging from 89% to 95%. 2-5 Various algorithms were tested, including: instrument barcoding with template matching, 2 random forest, 5 and convolu- tional neural networks. 3 Three studies presented their object detection models in the context of a robot man- ipulator which could pick up the detected instrument. 2,4,5 If computer vision is to have widespread adoption as a modality to perform the surgical count, the underlying object detection and tracking algorithms must be robust to the large number and types of surgical objects. The cat- egories of objects present in an operative setting include soft disposables, such as laparotomy sponges, hard dis- posables, such as surgical needles, and instruments, such as hemostats. However, all 4 of the studies only con- sidered object detection on a limited set of instruments and hard disposables. In particular, none of the studies at- tempted detection of surgical sponges and needles, 2 of the most commonly miscounted items in the OR. 1 Data standardization and algorithmic benchmarking is another concern, as only 2 studies released their datasets publicly. 4,5 Both datasets are limited in size (3200 and 3009 images, respectively), number of object categories, and types of objects, as Zhou and Wachs considered 5 instrument categories (scalpel, retractor, hemostat, scis- sors and babcock forceps), 5 while Lavado considered 4 instrument categories (scalpel, straight dissection clamp, straight mayo scissor and curved mayo scissor). 3 Further- more, neither dataset contains annotations with object identifiers for tracking objects across a sequence of images. A collaborative effort between research groups should be established to create and publicly release a sufficiently large and varied dataset for detection and tracking. This would allow for a standardized process to evaluate the performance of computer vision algorithms at the surgical counting task. Author Contributions Critical revision of the manuscript: Gustavo Chavez, Albert Haque, David Y. Zhao, Rahim Nazerali, Derek F. Amanatullah Study concept and design: Rahim Nazerali and Derek F. Amanatuah Acquisition of data: Gustavo Chavez, David Y. Zhao, and Albert Haque Study supervision: Derek F. Amanatullah Analysis and interpretation: Gustavo Chavez, David Y. Zhao, and Albert Haque 1 Department of Orthopaedic Surgery, Stanford Medicine, Redwood City, CA, USA 2 Department of Computer Science, Stanford University, CA, USA 3 Division of Plastic and Reconstructive Surgery, Department of Surgery, Stanford Medicine, Palo Alto, CA, USA These authors contributed equally to this work. Corresponding Author: Derek F. Amanatullah, Department of Orthopaedic Surgery, Stanford Medicine, 450 Broadway Street, Pavillion C, Redwood City, CA, USA. Email: dfa@stanford.edu