Bibliographic record
Abstract
Video recording surgical procedures has gained public and industry attention recently, for good reason: filming surgical activity has far-reaching potential, but comes with significant questions about its role in the operating room (OR). Incorporating OR cameras, microphones, and other sensors is a natural next step in pursuing surgical data, so creating buy-in among the medical and patient communities is crucial. Furthermore, even with stakeholder acceptance, surgi-videography pioneers bear the significant responsibility of architecting a system that complements the existing OR environment, while respecting requirements for sterility, privacy, and data protection. WHY IT IS A GOOD IDEA FOR SURGERY Capturing OR activity via video is vastly beneficial. Recordings enhance training, provide useful data for patients’ medical records, and form the foundation to analyze OR performance and safety. Technology and photography systems allow multiview, high-resolution images of patient anatomy, the surgical team, and all activities, creating an omniscient record of a patient's treatment. These raw, unbiased accounts—both of room activity and the procedure itself—will ostensibly be closer to “truth” than the post-hoc recollections contained in surgical dictations, which frequently lack important details missed or undocumented by the surgeon.1 Recordings of the entire OR allow an objective evaluation of unrecognized risks and hazards and provide more effective root cause analysis and peer review after adverse events. Detailed, procedural videos not only benefit peer review; they more thoroughly document the specific alterations in anatomy that occurred as a result of surgery and are thus highly useful for a patient's future care. Dictations still help interpret and rapidly process surgical proceedings, but, similar to radiology records, raw images would permit secondary review should deeper understanding be needed. Unfiltered surgical data would also be a boon to those of us studying the OR. Precisely understanding OR dynamics—how the team interacts and responds to unexpected events, and how trainees acquire skills—offers a vast new field of data against which we can test hypotheses and defend suggested solutions. We will be able to systematically assess new techniques and technologies, identify best practices among numerous individual preferences, and provide evidence-based recommendations for process improvements and surgical training. PATIENTS DESERVE AND WANT VIDEO RECORDING, TOO Beyond surgeons and administrators using video for academic research and hospital operations, the patient public is, understandably, also interested in such recordings. Enabling full access to and understanding of what was done to their bodies respects patients’ autonomy and empowers them to make the best health decisions. National regulations, including HIPAA in the United States and PHIPA in Canada, expanded patients’ rights to access and use their medical records and participate in their own care. Given these national priorities to empower patients, we are obligated, to the best of our current abilities, to ensure medical records most accurately reflect the details of patients’ surgical procedures. Surgical videos would also satisfy a legitimate public desire to demystify medicine. Being anesthetized is an experience often intimidating and enigmatic to patients; it is possible that this can be ameliorated by enabling patients to watch what happened to them while they were unconscious. However, creating a means of viewing one's own procedure should not be taken lightly, as the public often has a deeply misguided understanding of surgery itself. Simultaneous with the development of better OR video technology for our internal use, we facilitate a means mass public consumption of videos that can be quickly viewed, “liked,” “shared,” and “posted,” making us responsible for preventing misinformation. Surgical videos are already easily accessible online, and Snapchat accounts displaying live surgery are widely followed.2 Rather than offering transparency, however, these videos are often highly skewed, “ideal” cases edited to eliminate the blood, struggles, and minor errors that can happen routinely in day-to-day surgery. This may set up the struggles seen in “real” surgical videos to be perceived negatively by patients and juries, and gives urgency to efforts to define the boundaries of acceptable surgical performance. Public appetite for surgical transparency should also be recognized as symptomatic of a breakdown of trust in surgeons, medicine, and hospital communities overall.3,4 Patients have inadvertently5 and intentionally6 audio-recorded their procedures, only to discover evidence implying unprofessional behavior by physicians when they thought no one was listening. Furthermore, the Boston Globe's exposé on concurrent surgery7 and first-hand accounts of questionable surgical practices published in medical journals8,9 have led to questions about activity in the privacy of the OR. We have an opportunity to use video in the OR for bridge building and public education, but if we do not lead this charge, it may instead primarily become fodder for fault-finding. PUBLIC POLICY NEEDS TO PROTECT ALL STAKEHOLDERS Massachusetts10 and Wisconsin11 have introduced bills to permit patients to request a recording of their surgical procedures. While Wisconsin's bill partially acknowledges the need for surgical team privacy (ie, patients cannot share the video without permission except for medical purposes), both bills favor creating litigation evidence and enabling surgical team monitoring rather than protecting all OR stakeholders. Without protecting the surgical team, there is a risk that the sudden availability of surgical videos could lead to misinterpretation of surgical events. We already know that surgeons vary in technical proficiency,12,13 and even surgeons in the same institution may differ in their steps and techniques for the “same” operation.14 Thus, it is unclear what constitutes “normal” surgical practices and how much variance would justify scrutiny, not to mention malpractice. With widespread OR video recording, we could measure variability and develop normative standards. However, in a “Catch 22”: without normative data, the widespread adoption of video threatens surgeons’ liability and will likely be resisted. Questions such as, “what constitutes the reasonable range of surgical abilities?,” and “how often do, and should, minor and inconsequential errors occur?” must be posed procedure-by-procedure before one-off videos could be subjected to determinations of standard of care. Furthermore, an omniscient system, advantageous for its vigilance and ability to capture moments missed by the surgical team, will uncover events that the surgeon never realized. It would be impractical for the surgeon to review the entire procedure and dictate based on the video. What if the surgeon's dictation and the video record do not coincide? Will surgeons be held to the same standard of omniscience that video can achieve? Lastly, we do not know what effect the presence of video recording and continuous “monitoring” in the OR will have on surgeon and surgical team behavior. Surgeons may feel pressure to practice “defensive” surgery, limit teaching,15 and “perform” for the camera—all of which may have unintended consequences for technical performance, response to adverse events, and resident training. Although there is a clear value to including surgical videos in the medical record, the surgical profession needs a grace period to generate normative data and study the potential negative effects of widespread video recording before surgical videos become freely available. When cockpit and flight data recording was proposed in the 1960s, pilots shared similar concerns. Pilots have strong personal motivation to enhance airline safety, yet this threat to their privacy and the risk of retribution for mistakes led to legitimate demands for carefully delineated uses of, and protections for, cockpit voice recorder data before cockpit voice recorders were introduced into airplanes. We as a profession need to similarly take charge of this issue to ensure that the coming tide of OR recording is instituted in a manner that protects surgical teams, hospitals, and most importantly, patients. Generalized OR recording, like airplane “black boxes,” will result in a better understanding of intraoperative risks, lead to improvements in surgical training and safety,16 and enhance patients’ future care. To enable sustainable, widespread use of video in the OR, we need to clearly delineate how data will be used, how the privacy of surgical teams and patients will be protected, and how hospitals and practitioners can be spared unnecessary liability (Table 1).17 As we turn the camera lens on ourselves, we may be surprised by what we find. Surgeons have long been privileged to apply cutting-edge technologies to our patients’ benefit; we can’t be afraid to embrace the power of video and uncover and learn from the OR's “truth.”TABLE 1: OR Data Types, Uses, and Proposed Protections
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".