Value Added: the Case for Point-of-View Camera use in Orthopedic Surgical Education

Abstract

Background

Orthopedic surgical education is evolving as educators search for new ways to enhance surgical skills training. Orthopedic educators should seek new methods and technologies to augment and add value to real-time orthopedic surgical experience. This paper describes a protocol whereby we have started to capture and evaluate specific orthopedic milestone procedures with a GoPro® point-of-view video camera and a dedicated video reviewing website as a way of supplementing the current paradigm in surgical skills training. We report our experience regarding the details and feasibility of this protocol.

Methods

Upon identification of a patient undergoing surgical fixation of a hip or ankle fracture, an orthopedic resident places a GoPro® point-of-view camera on his or her forehead. All fluoroscopic images acquired during the case are saved and later incorporated into a video on the reviewing website. Surgical videos are uploaded to a secure server and are accessible for later review and assessment via a custom-built website. An electronic survey of resident participants was performed utilizing Qualtrics software. Results are reported using descriptive statistics.

Results

A total of 51 surgical videos involving 23 different residents have been captured to date. This includes 20 intertrochanteric hip fracture cases and 31 ankle fracture cases. The average duration of each surgical video was 1 hour and 16 minutes (range 40 minutes to 2 hours and 19 minutes). Of 24 orthopedic resident surgeons surveyed, 88% thought capturing a video portfolio of orthopedic milestones would benefit their education

Conclusions

There is a growing demand in orthopedic surgical education to extract more value from each surgical experience. While further work in development and refinement of such assessments is necessary, we feel that intraoperative video, particularly when captured and presented in a non-threatening, user friendly manner, can add significant value to the present and future paradigm of orthopedic surgical skill training.

Introduction

Factors both in and out of educators' control are changing orthopedic surgical education. Imposed restrictions on resident work hours have led to greater emphasis on surgical simulation and have increased scrutiny on how we provide surgical skills training. At the same time, an emphasis on surgical efficiency, as well as pressure to increase case volume and clinical throughput, has restricted opportunities for graduated responsibility. The American Board of Orthopedic Surgery (ABOS) mandate for a post-graduate year one (PGY-1) surgical skills training program has shifted baseline orthopedic skill acquisition towards simulated practice environments. Even as the quality of training simulation improves, orthopedic educators must also seek methods to enhance and add value to real-time trainee orthopedic surgical experiences in the operating room.

Orthopedic educators must foster and evaluate resident surgical skills. The Accreditation Council for Graduate Medical Education (ACGME) mandates review of resident performance based upon achieving designated milestones. There are 16 patient care milestones1 that are directly or indirectly related to a trainee's surgical skill (Figure 1). Currently, individual faculty members assess resident performance by subjectively evaluating milestone ratings based upon limited and non-uniform observation of procedural experiences. The ACGME requires residents to maintain a log of cases they have participated in, but these logs do not include any objective assessment of the individual's involvement or performance

Figure 1.

ACGME Orthopedic patient care milestones for a) hip and b) ankle fractures.¹

In many skill trades and professions, teachable moments are preserved with the use of video recording technology.^2-4 For example, it is not uncommon for athletic teams to video record games or practice scenarios for extensive review later.^5,6 Likewise, it is common practice for aviation experts and law enforcement personnel to video record scenarios, both actual and simulated experiences, for later review. The benefits of retrospective video review in training are numerous. An individual or team can review processes of care, shared or unique practices, and preferences. Individuals can identify technical shortcomings or achievements with the benefit and knowledge of hindsight regarding the outcome of such technical performances. An ability to view one's own performance lends insight into particular strengths and weaknesses, perceived or real. Another key benefit of video review is that it can facilitate peer review. The peer review process allows trainees to understand strengths or weaknesses in the context of their training level. While specific evidence remains unclear, many experts believe that such a review process increases knowledge of and indirectly benefits procedure-specific skill. It also affords the opportunity for a trainee to review surgical performance with an attending or faculty instructor in an environment that is controlled and free from the stress often experienced in surgical environments.

This paper describes the protocol whereby we capture specific orthopedic milestone procedures utilizing a GoPro® point-of-view video camera with subsequent review of video on a website. Two milestone procedures commonly encountered on our orthopedic trauma team, hip fracture and ankle fractures, served as the initial procedures of interest for the GoPro® program. We asked orthopedic residents to obtain a video recording of their participation in these procedures as primary surgeon during a two-month trauma rotation. These videos were then catalogued and remain part of the trainee's milestone portfolio. We report our experience regarding the details and feasibility of this protocol, including the acceptance of the residents and the operating room staff.

Methods

Our institutional review board (IRB) approved the protocol for resident participation in this ongoing educational project. In addition, all participating patients provided verbal and signed written informed consent before taking part in the study. Upon identification of a subject with a hip or ankle fracture requiring operative management, a resident or staff surgeon discussed the study with the patient and enrolled participants in accordance with the IRB-approved protocol. Operating room staff, including anesthesia, nursing, and surgical technologists, were notified that a GoPro® point-of-view camera use was part of an ongoing educational initiative and the staff provided verbal consent before the video recording procedures.

Prior to surgical preparation, the resident surgeon placed the GoPro® point-of-view camera on his or her head (Figure 2). The Wi-Fi feature was enabled while the adjustments were being made so that the viewing angle and area of focus could be confirmed with use of a mobile application on a portable device (e.g., an iPhone®). After surgical scrub, the camera was turned on by an operating room scrub technician. The camera recorded the point-of-view of the primary surgeon, in this case, the resident learner. During the recording, the assistance of the attending surgeon was frequently visible within the field of view. This protocol allowed the learner to record the operation, including the lessons provided by attending surgeons. Upon completion of the surgical procedure or at any point throughout the case, the camera could be turned off or removed. All of the fluoroscopic images that were acquired were time stamped and saved for later incorporation in the video.

Figure 2.

Intraoperative photograph of a resident surgeon wearing a GoPro® camera.

The camera (GoPro® Hero 3+, San Mateo, CA, USA) has a resolution of 1920 by 1080 pixels and a field of view of 69.5^˚ by 118.2^˚. The camera view was aligned with the surgeon's line of site onto the surgical field, generally pointed slightly downward at an angle of approximately 50-60^˚ from vertical. The videos were collected in the 1080p30 format setting (resolution of 1920 by 1080 pixels at 30 frames per second (fps)), which yielded a file size of approximately 140 megabytes per minute. For later review on a website, the video files were converted from 1080p30 format to a wide 480p24 format (resolution of 852 by 480 pixels at 24 fps). With a total bitrate under 1 megabyte per second (1Mbps), this format enabled streaming of the video content while retaining adequate viewing resolution.

After capturing the intraoperative video, the data were transferred from the micro SD card in the camera to a secure, dedicated website. Initially, we manually edited the fluoroscopic images (transferred separately as DICOM files from the fluoroscopy machine) into each surgical video at the appropriate time and place. This manual process took approximately two hours and 15 minutes per video.⁷ We subsequently developed a software algorithm to automate video format conversion and the superimposing of fluoroscopic images onto the streaming surgical video at the appropriate time (Figure 3). This largely avoided the need for human intervention with the exception of the synchronization of the video time with the timestamp on the images.

Figure 3.

Screenshot of a recorded procedure with embedded C-arm image accessible via a secured website.

A purpose-built, interactive website accepts and catalogs the raw GoPro® video file and completes the necessary video conversion (Figure 4). The website was built using a combination of PHP, MySQL, JavaScript, HTML and CSS programming languages. This website enhanced the utility of video review in several ways. First, it provided context for the operation, by including the surgical procedure notes as well as the before and after radiographic images. Second, the website presented the opportunity to give and receive coaching on surgical video. Anyone logged onto the website could ask a question of another member of the website. Reviewers of each video were also given the ability to leave comments at any time on the video. All questions and comments were time stamped and appear at the appropriate time during video viewing. The individual posting the question or comment can be anonymous or identified. Finally, the website allows resident surgical performance to be assessed with a global rating scale / OSATS score. This assessment metric was presented below each surgical video (Figure 5).

Figure 4.

View of the OrthoVid website log, with videos cataloged including title, date, and operating resident (names shown are stage names assigned for confidentiality).

Figure 5.

Interactive grading platform with modified OSATS. The global rating scale was adapted from the scale used in articular fracture reduction simulation.²

While yet to be standardized, videos were reviewed by individual residents and faculty members involved in the care of the patient. During these reviews, attention was directed toward various aspects of the surgical performance thought to be important to the specific procedure (e.g., preparation for the procedure, surgical dissection, reduction, fixation technique). In addition, while not formalized, select videos were later used for instruction of PGY1 level residents or others as they rotated onto the trauma team. All orthopedic residents at our institution were invited to participate in a survey regarding their opinions of utilizing the GoPro® pointof- view camera during surgical cases.

Results

The 51 surgical videos that capture the performance of 23 different residents include 20 intertrochanteric hip fracture cases and 31 ankle fracture cases. The average duration of each surgical video was 1 hour and 16 minutes, with a range of 40 minutes to 2 hours and 19 minutes. Most surgeries were performed during a patient's index admission to the hospital for their specific injury.

In total, 21/24 (87.5%) surveyed residents felt that the ability to capture intraoperative, point-of-view video could add value to their orthopedic educational experience. Of 24 residents surveyed, 19 (79%) were interested in receiving a GoPro® video camera to document their surgical procedures over the course of a residency experience and 21 (88%) percent felt that capturing a video portfolio of select orthopedic milestone procedures would aid their surgical education. Regarding resident's rank order of how they would prefer to review learning materials to aid their orthopedic education, self-review was ranked most highly, followed by review by orthopedic staff, review with orthopedic staff, and lastly peer review. Nine residents (38%) reported headache as a downside to wearing the GoPro® camera. In one case, a member of the peri-operative team indicated that they were uncomfortable with the use of intraoperative video, and for this reason, the recording was abandoned during that particular case. Subsequent discussion with operating room staff helped convey and clarify the value and purpose of the video recording protocol. To our knowledge there have been no adverse patient consequences related to the use of the GoPro® camera to capture intraoperative video.

Discussion

Changes to the residency training environment have led to restricted access to and a lack of uniformity in the surgical experience. It is clear that duty-hour restrictions, enhanced focus on patient safety, increased emphasis on high clinical throughput and operative efficiency all present challenges to the traditional paradigm. With decreasing reimbursements for individual procedures, the burden to operate more efficiently stands in direct conflict to the process of teaching and graduated responsibility. In addition, indirect metrics of surgical performance such as surgical site infection and hospital readmission rates are increasingly tied to provider and hospital reimbursement, further decreasing the tolerance for accommodating a resident learning curve. The present study describes our initial experience with a method for capturing and evaluating surgical video in an orthopedic training scenario.

Intraoperative education must remain a fundamental training component in orthopedic surgery. There is little debate that robust, objective assessments of orthopedic surgical skill simply do not exist. Yet with the advent of patient care milestones, educators feel enhanced pressure to identify such metrics. Equally discouraging is that even experts, on occasion, fail to recognize or agree upon objective assessment techniques or on what represents a dangerous, proficient, or exceptional surgical performance. In a prior study that we conducted, residents who reviewed their recorded performance with a traumatologist showed significant improvement in OSATS scores and decreases in the number of fluoroscopic images utilized.7 While yet to demonstrate robust validation, the OSATS score represents a commonly utilized scoring metric for orthopedic surgical performance. Previous attempts to utilize intraoperative video in surgical training have been met with both success and failure.^8-11 Prior camera modalities suffered from poor resolution, static positioning and no contextual focus. For many operating rooms, a permanent audio-visual system is not possible due to cost and inflexibility with multiple operating rooms at a given medical center.12 Wearable devices eliminate the need for a separate videographer and allow for the surgeon's vantage point to be recorded. For these reasons, our group trialed several point-of-view cameras including both Google Glass® and GoPro® (Table I).^13-16 Notable differences include the higher price, limited battery life, and software limitations associated with Google Glass®, as well as the higher resolution provided by GoPro® Hero 3+. In January 2015, Google indefinitely suspended its Explorer program, limiting access to Google Glass®.¹⁶ For these reasons, we identified the GoPro® as the camera line most suitable to our needs.

With our protocol, we hope to add value to the operative educational experience by capturing, cataloguing and reviewing select surgical procedures on video. Video review reinforces teaching principles in a less stressful setting than typically encountered in an operating room. Birkmeyer et al. suggested that technical surgical skill can be assessed via video peer review and correlated to postoperative complication rates.¹⁷ In that study, 20 practicing bariatric surgeons submitted surgical videos for peer review. It was shown that the technical skill of a surgeon could be stratified based upon the OSATS score derived from video review. More importantly, it was noted that skillful surgeons had fewer postoperative complications and lower rates of reoperation, readmission, and visits to the emergency department.

An additional advantage of the video recording and review protocol is resident acceptance. We found a high acceptance rate of our technology platform by residents in our program. The merits of the line of work are largely self-explanatory. It is designed to enhance surgical skill training and education and to inform educators on best practice. The benefits seen will directly impact individuals responsible for much of the work.

Several iterations of this technology have led us to conclude that an effective video presentation of surgical performance requires a coherent narrative flow, including more than just the raw point-of-view video. Consequently, the surgical notes, pre- and post-operative radiographs, and the intraoperative fluoroscopic imaging must be presented passively and in context. When the narrative flow is disrupted—due to missing radiographic information, for example—viewers tend to become distracted and have trouble re-engaging with the surgery. Also, it must be easy for the viewer to scan back and forth through the video stream to quickly find moments of interest during the surgery.

Our long-term vision for this point-of-view educational surgical experience continues to evolve. We can envision our residency training program providing all entering orthopedic residents within our program with a personal GoPro® camera. Instructions could be provided on how to upload procedural video, as well as on how to obtain and provide peer or faculty assessments. A working repository for this data holds promise for both formative and summative surgical skill assessment. It might, for instance, be required that as part of a senior resident's graduating portfolio they provide the program director with a 16-item menu of patient care milestone videos demonstrating a minimal level of proficiency

There are limitations to the present line of work. Notably, the informed consent process and potential for abuse should not go unnoticed. Using intraoperative video is relatively easy to do with this protocol; however, potential inclusion of identifiable subjects or personnel who have not been informed or consented could threaten the viability of the protocol. Additional limitations relate to the expense of the protocol. GoPro® cameras cost roughly $300.00. Their successful operation depends upon a charged battery and removable memory in the form of a micro SD card. All of these items need to be available for use at a moment's notice. The tasks of uploading large files and editing video require time and effort. In our initial experience, videos had to be manually edited to add the fluoroscopic and radiographic images using specialty video-editing software. Our newer methods avoid this painstaking step by using timestamps in the fluoroscopic image DICOM metadata to automatically overlay the images on the video, which reduces the editing time to several minutes once the data have been acquired. We are hopeful that time and improvements in technology will make this process become even more efficient.

Changes in orthopedic surgical education are inevitable. For reasons previously highlighted, there is an increased emphasis being placed today on procedural skill development, particularly for surgical simulation. With increased pressure to provide objective assessments of surgical skill there is a growing demand to extract the maximum value out of each surgical experience. While further work in development and refinement of such assessments is sure to occur, we feel that intraoperative video, particularly when captured, processed, and presented in a non-threatening, user-friendly manner, can add significant value to the present and future paradigm of orthopedic surgical skill training.